Search Results: "zed"

10 August 2020

Arnaud Rebillout: GoAccess 1.4, a detailed tutorial

GoAccess v1.4 was just released a few weeks ago! Let's take this chance to write a loooong tutorial. We'll go over every steps to install and operate GoAccess. This is a tutorial aimed at those who don't play sysadmin every day, and that's why it's so long, I did my best to provide thorough explanations all along, so that it's more than just a "copy-and-paste" kind of tutorial. And for those who do play sysadmin everyday: please try not to fall asleep while reading, and don't hesitate to drop me an e-mail if you spot anything inaccurate in here. Thanks! Introduction So what's GoAccess already? GoAccess is a web log analyzer, and it allows you to visualize the traffic for your website, and get to know a bit more about your visitors: how many visitors and hits, for which pages, coming from where (geolocation, operating system, web browser...), etc... It does so by parsing the access logs from your web server, be it Apache, NGINX or whatever. GoAccess gives you different options to display the statistics, and in this tutorial we'll focus on producing a HTML report. Meaning that you can see the statistics for your website straight in your web browser, under the form of a single HTML page. For an example, you can have a look at the stats of my blog here: https://goaccess.arnaudr.io. GoAccess is written in C, it has very few dependencies, it had been around for about 10 years, and it's distributed under the MIT license. Assumptions This tutorial is about installing and configuring, so I'll assume that all the commands are run as root. I won't prefix each of them with sudo. I use the Apache web server, running on a Debian system. I don't think it matters so much for this tutorial though. If you're using NGINX it's fine, you can keep reading. Also, I will just use the name SITE for the name of the website that we want to analyze with GoAccess. Just replace that with the real name of your site. I also assume the following locations for your stuff: If you have your stuff in /srv/SITE/ log,www instead, no worries, just adjust the paths accordingly, I bet you can do it. Installation The latest version of GoAccess is v1.4, and it's not yet available in the Debian repositories. So for this part, you can follow the instructions from the official GoAccess download page. Install steps are explained in details, so there's nothing left for me to say :) When this is done, let's get started with the basics. We're talking about the latest version v1.4 here, let's make sure: 
$ goaccess --version
GoAccess - 1.4.
...
Now let's try to create a HTML report. I assume that you already have a website up and running. GoAccess needs to parse the access logs. These logs are optional, they might or might not be created by your web server, depending on how it's configured. Usually, these log files are named access.log, unsurprisingly. You can check if those logs exist on your system by running this command: 
find /var/log -name access.log
Another important thing to know is that these logs can be in different formats. In this tutorial we'll assume that we work with the combined log format, because it seems to be the most common default. To check what kind of access logs your web server produces, you must look at the configuration for your site. For an Apache web server, you should have such a line in the file /etc/apache2/sites-enabled/SITE.conf: 
CustomLog $ APACHE_LOG_DIR /SITE/access.log combined
For NGINX, it's quite similar. The configuration file would be something like /etc/nginx/sites-enabled/SITE, and the line to enable access logs would be something like: 
access_log /var/log/nginx/SITE/access.log
Note that NGINX writes the access logs in the combined format by default, that's why you don't see the word combined anywhere in the line above: it's implicit. Alright, so from now on we assume that yes, you have access log files available, and yes, they are in the combined log format. If that's the case, then you can already run GoAccess and generate a report, for example for the log file /var/log/apache2/access.log 
goaccess \
    --log-format COMBINED \
    --output /tmp/report.html \
    /var/log/apache2/access.log
It's possible to give GoAccess more than one log files to process, so if you have for example the file access.log.1 around, you can use it as well: 
goaccess \
    --log-format COMBINED \
    --output /tmp/report.html \
    /var/log/apache2/access.log \
    /var/log/apache2/access.log.1
If GoAccess succeeds (and it should), you're on the right track! All is left to do to complete this test is to have a look at the HTML report created. It's a single HTML page, so you can easily scp it to your machine, or just move it to the document root of your site, and then open it in your web browser. Looks good? So let's move on to more interesting things. Web server configuration This part is very short, because in terms of configuration of the web server, there's very little to do. As I said above, the only thing you want from the web server is to create access log files. Then you want to be sure that GoAccess and your web server agree on the format for these files. In the part above we used the combined log format, but GoAccess supports many other common log formats out of the box, and even allows you to parse custom log formats. For more details, refer to the option --log-format in the GoAccess manual page. Another common log format is named, well, common. It even has its own Wikipedia page. But compared to combined, the common log format contains less information, it doesn't include the referrer and user-agent values, meaning that you won't have it in the GoAccess report. So at this point you should understand that, unsurprisingly, GoAccess can only tell you about what's in the access logs, no more no less. And that's all in term of web server configuration. Configuration to run GoAccess unprivileged Now we're going to create a user and group for GoAccess, so that we don't have to run it as root. The reason is that, well, for everything running unattended on your server, the less code runs as root, the better. It's good practice and common sense. In this case, GoAccess is simply a log analyzer. So it just needs to read the logs files from your web server, and there is no need to be root for that, an unprivileged user can do the job just as well, assuming it has read permissions on /var/log/apache2 or /var/log/nginx. The log files of the web server are usually part of the adm group (though it might depend on your distro, I'm not sure). This is something you can check easily with the following command: 
ls -l /var/log   grep -e apache2 -e nginx
As a result you should get something like that: 
drwxr-x--- 2 root adm 20480 Jul 22 00:00 /var/log/apache2/
And as you can see, the directory apache2 belongs to the group adm. It means that you don't need to be root to read the logs, instead any unprivileged user that belongs to the group adm can do it. So, let's create the goaccess user, and add it to the adm group: 
adduser --system --group --no-create-home goaccess
addgroup goaccess adm
And now, let's run GoAccess unprivileged, and verify that it can still read the log files: 
setpriv \
    --reuid=goaccess --regid=goaccess \
    --init-groups --inh-caps=-all \
    -- \
    goaccess \
    --log-format COMBINED \
    --output /tmp/report2.html \
    /var/log/apache2/access.log
setpriv is the command used to drop privileges. The syntax is quite verbose, it's not super friendly for tutorials, but don't be scared and read the manual page to learn what it does. In any case, this command should work, and at this point, it means that you have a goaccess user ready, and we'll use it to run GoAccess unprivileged. Integration, option A - Run GoAccess once a day, from a logrotate hook In this part we wire things together, so that GoAccess processes the log files once a day, adds the new logs to its internal database, and generates a report from all that aggregated data. The result will be a single HTML page. Introducing logrotate In order to do that, we'll use a logrotate hook. logrotate is a little tool that should already be installed on your server, and that runs once a day, and that is in charge of rotating the log files. "Rotating the logs" means moving access.log to access.log.1 and so on. With logrotate, a new log file is created every day, and log files that are too old are deleted. That's what prevents your logs from filling up your disk basically :) You can check that logrotate is indeed installed and enabled with this command (assuming that your init system is systemd): 
systemctl status logrotate.timer
What's interesting for us is that logrotate allows you to run scripts before and after the rotation is performed, so it's an ideal place from where to run GoAccess. In short, we want to run GoAccess just before the logs are rotated away, in the prerotate hook. But let's do things in order. At first, we need to write a little wrapper script that will be in charge of running GoAccess with the right arguments, and that will process all of your sites. The wrapper script This wrapper is made to process more than one site, but if you have only one site it works just as well, of course. So let me just drop it on you like that, and I'll explain afterward. Here's my wrapper script: 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
#!/bin/bash
# Process log files /var/www/apache2/SITE/access.log,
# only if /var/lib/goaccess-db/SITE exists.
# Create HTML reports in $1, a directory that must exist.
set -eu
OUTDIR=
LOGDIR=/var/log/apache2
DBDIR=/var/lib/goaccess-db
fail()   echo >&2 "$@"; exit 1;  
[ $# -eq 1 ]   fail "Usage: $(basename $0) OUTPUT_DIRECTORY"
OUTDIR=$1
[ -d "$OUTDIR" ]   fail "'$OUTDIR' is not a directory"
[ -d "$LOGDIR" ]   fail "'$LOGDIR' is not a directory"
[ -d "$DBDIR"  ]   fail "'$DBDIR' is not a directory"
for d in $(find "$LOGDIR" -mindepth 1 -maxdepth 1 -type d); do
    site=$(basename "$sitedir")
    dbdir=$DBDIR/$site
    logfile=$d/access.log
    outfile=$OUTDIR/$site.html
    if [ ! -d "$dbdir" ]   [ ! -e "$logfile" ]; then
        echo "  Skipping site '$site'"
        continue
    else
        echo "  Processing site '$site'"
    fi
    setpriv \
        --reuid=goaccess --regid=goaccess \
        --init-groups --inh-caps=-all \
        -- \
    goaccess \
        --agent-list \
        --anonymize-ip \
        --persist \
        --restore \
        --config-file /etc/goaccess/goaccess.conf \
        --db-path "$dbdir" \
        --log-format "COMBINED" \
        --output "$outfile" \
        "$logfile"
done
So you'd install this script at /usr/local/bin/goaccess-wrapper for example, and make it executable: 
chmod +x /usr/local/bin/goaccess-wrapper
A few things to note: As is, the script makes the assumption that the logs for your site are logged in a sub-directory /var/log/apache2/SITE/. If it's not the case, adjust that in the wrapper accordingly. The name of this sub-directory is then used to find the GoAccess database directory /var/lib/goaccess-db/SITE/. This directory is expected to exist, meaning that if you don't create it yourself, the wrapper won't process this particular site. It's a simple way to control which sites are processed by this GoAccess wrapper, and which sites are not. So if you want goaccess-wrapper to process the site SITE, just create a directory with the name of this site under /var/lib/goaccess-db: 
mkdir -p /var/lib/goaccess-db/SITE
chown goaccess:goaccess /var/lib/goaccess-db/SITE
Now let's create an output directory: 
mkdir /tmp/goaccess-reports
chown goaccess:goaccess /tmp/goaccess-reports
And let's give a try to the wrapper script: 
goaccess-wrapper /tmp/goaccess-reports
ls /tmp/goaccess-reports
Which should give you: 
SITE.html
At the same time, you can check that GoAccess populated the database with a bunch of files: 
ls /var/lib/goaccess-db/SITE
Setting up the logrotate prerotate hook At this point, we have the wrapper in place. Let's now add a pre-rotate hook so that goaccess-wrapper runs once a day, just before the logs are rotated away. The logrotate config file for Apache2 is located at /etc/logrotate.d/apache2, and for NGINX it's at /etc/logrotate.d/nginx. Among the many things you'll see in this file, here's what is of interest for us: In the config file, there is also this snippet: 
prerotate
    if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
        run-parts /etc/logrotate.d/httpd-prerotate; \
    fi; \
endscript
It indicates that scripts in the directory /etc/logrotate.d/httpd-prerotate/ will be executed before the rotation takes place. Refer to the man page run-parts(8) for more details... Putting all of that together, it means that logs from the web server are rotated once a day, and if we want to run scripts just before the rotation, we can just drop them in the httpd-prerotate directory. Simple, right? Let's first create this directory if it doesn't exist: 
mkdir -p /etc/logrotate.d/httpd-prerotate/
And let's create a tiny script at /etc/logrotate.d/httpd-prerotate/goaccess: 
1
2
#!/bin/sh
exec goaccess-wrapper /tmp/goaccess-reports
Don't forget to make it executable: 
chmod +x /etc/logrotate.d/httpd-prerotate/goaccess
As you can see, the only thing that this script does is to invoke the wrapper with the right argument, ie. the output directory for the HTML reports that are generated. And that's all. Now you can just come back tomorrow, check the logs, and make sure that the hook was executed and succeeded. For example, this kind of command will tell you quickly if it worked: 
journalctl   grep logrotate
Integration, option B - Run GoAccess once a day, from a systemd service OK so we've just seen how to use a logrotate hook. One downside with that is that we have to drop privileges in the wrapper script, because logrotate runs as root, and we don't want to run GoAccess as root. Hence the rather convoluted syntax with setpriv. Rather than embedding this kind of thing in a wrapper script, we can instead run the wrapper script from a [systemd][] service, and define which user runs the wrapper straight in the systemd service file. Introducing systemd niceties So we can create a systemd service, along with a systemd timer that fires daily. We can then set the user and group that execute the script straight in the systemd service, and there's no need for setpriv anymore. It's a bit more streamlined. We can even go a bit further, and use systemd parameterized units (also called templates), so that we have one service per site (instead of one service that process all of our sites). That will simplify the wrapper script a lot, and it also looks nicer in the logs. With this approach however, it seems that we can't really run exactly before the logs are rotated away, like we did in the section above. But that's OK. What we'll do is that we'll run once a day, no matter the time, and we'll just make sure to process both log files access.log and access.log.1 (ie. the current logs and the logs from yesterday). This way, we're sure not to miss any line from the logs. Note that GoAccess is smart enough to only consider newer entries from the log files, and discard entries that are already in the database. In other words, it's safe to parse the same log file more than once, GoAccess will do the right thing. For more details see "INCREMENTAL LOG PROCESSING" from man goaccess. systemd]: https://freedesktop.org/wiki/Software/systemd/ Implementation And here's how it all looks like. First, a little wrapper script for GoAccess: 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
#!/bin/bash
# Usage: $0 SITE DBDIR LOGDIR OUTDIR
set -eu
SITE=$1
DBDIR=$2
LOGDIR=$3
OUTDIR=$4
LOGFILES=()
for ext in log log.1; do
    logfile="$LOGDIR/access.$ext"
    [ -e "$logfile" ] && LOGFILES+=("$logfile")
done
if [ $ #LOGFILES[@]  -eq 0 ]; then
    echo "No log files in '$LOGDIR'"
    exit 0
fi
goaccess \
    --agent-list \
    --anonymize-ip \
    --persist \
    --restore \
    --config-file /etc/goaccess/goaccess.conf \
    --db-path "$DBDIR" \
    --log-format "COMBINED" \
    --output "$OUTDIR/$SITE.html" \
    "$ LOGFILES[@] "
This wrapper does very little. Actually, the only thing it does is to check for the existence of the two log files access.log and access.log.1, to be sure that we don't ask GoAccess to process a file that does not exist (GoAccess would not be happy about that). Save this file under /usr/local/bin/goaccess-wrapper, don't forget to make it executable: 
chmod +x /usr/local/bin/goaccess-wrapper
Then, create a systemd parameterized unit file, so that we can run this wrapper as a systemd service. Save it under /etc/systemd/system/goaccess@.service: 
[Unit]
Description=Update GoAccess report - %i
ConditionPathIsDirectory=/var/lib/goaccess-db/%i
ConditionPathIsDirectory=/var/log/apache2/%i
ConditionPathIsDirectory=/tmp/goaccess-reports
PartOf=goaccess.service
[Service]
Type=oneshot
User=goaccess
Group=goaccess
Nice=19
ExecStart=/usr/local/bin/goaccess-wrapper \
 %i \
 /var/lib/goaccess-db/%i \
 /var/log/apache2/%i \
 /tmp/goaccess-reports
So, what is a systemd parameterized unit? It's a service to which you can pass an argument when you enable it. The %i in the unit definition will be replaced by this argument. In our case, the argument will be the name of the site that we want to process. As you can see, we use the directive ConditionPathIsDirectory= extensively, so that if ever one of the required directories does not exist, the unit will just be skipped (and marked as such in the logs). It's a graceful way to fail. We run the wrapper as the user and group goaccess, thanks to User= and Group=. We also use Nice= to give a low priority to the process. At this point, it's already possible to test. Just make sure that you created a directory for the GoAccess database: 
mkdir -p /var/lib/goaccess-db/SITE
chown goaccess:goaccess /var/lib/goaccess-db/SITE
Also make sure that the output directory exists: 
mkdir /tmp/goaccess-reports
chown goaccess:goaccess /tmp/goaccess-reports
Then reload systemd and fire the unit to see if it works: 
systemctl daemon-reload
systemctl start goaccess@SITE.service
journalctl   tail
And that should work already. As you can see, the argument, SITE, is passed in the systemctl start command. We just append it after the @, in the name of the unit. Now, let's create another GoAccess service file, which sole purpose is to group all the parameterized units together, so that we can start them all in one go. Note that we don't use a systemd target for that, because ultimately we want to run it once a day, and that would not be possible with a target. So instead we use a dummy oneshot service. So here it is, saved under /etc/systemd/system/goaccess.service: 
[Unit]
Description=Update GoAccess reports
Requires= \
 goaccess@SITE1.service \
 goaccess@SITE2.service
[Service]
Type=oneshot
ExecStart=true
As you can see, we simply list the sites that we want to process in the Requires= directive. In this example we have two sites named SITE1 and SITE2. Let's ensure that everything is still good: 
systemctl daemon-reload
systemctl start goaccess.service
journalctl   tail
Check the logs, both sites SITE1 and SITE2 should have been processed. And finally, let's create a timer, so that systemd runs goaccess.service once a day. Save it under /etc/systemd/system/goaccess.timer. 
[Unit]
Description=Daily update of GoAccess reports
[Timer]
OnCalendar=daily
RandomizedDelaySec=1h
Persistent=true
[Install]
WantedBy=timers.target
Finally, enable the timer: 
systemctl daemon-reload
systemctl enable --now goaccess.timer
At this point, everything should be OK. Just come back tomorrow and check the logs with something like: 
journalctl   grep goaccess
Last word: if you have only one site to process, of course you can simplify, for example you can hardcode all the paths in the file goaccess.service instead of using a parameterized unit. Up to you. Daily operations So in this part, we assume that you have GoAccess all setup and running, once a day or so. Let's just go over a few things worth noting. Serve your report Up to now in this tutorial, we created the reports in /tmp/goaccess-reports, but that was just for the sake of the example. You will probably want to save your reports in a directory that is served by your web server, so that, well, you can actually look at it in your web browser, that was the point, right? So how to do that is a bit out of scope here, and I guess that if you want to monitor your website, you already have a website, so you will have no trouble serving the GoAccess HTML report. However there's an important detail to be aware of: GoAccess shows all the IP addresses of your visitors in the report. As long as the report is private it's OK, but if ever you make your GoAccess report public, then you should definitely invoke GoAccess with the option --anonymize-ip. Keep an eye on the logs In this tutorial, the reports we create, along with the GoAccess databases, will grow bigger every day, forever. It also means that the GoAccess processing time will grow a bit each day. So maybe the first thing to do is to keep an eye on the logs, to see how long it takes to GoAccess to do its job every day. Also, maybe you'd like to keep an eye on the size of the GoAccess database with: 
du -sh /var/lib/goaccess-db/SITE
If your site has few visitors, I suspect it won't be a problem though. You could also be a bit pro-active in preventing this problem in the future, and for example you could break the reports into, say, monthly reports. Meaning that every month, you would create a new database in a new directory, and also start a new HTML report. This way you'd have monthly reports, and you make sure to limit the GoAccess processing time, by limiting the database size to a month. This can be achieved very easily, by including something like YEAR-MONTH in the database directory, and in the HTML report. You can handle that automatically in the wrapper script, for example: 
sfx=$(date +'%Y-%m')
mkdir -p $DBDIR/$sfx
goaccess \
    --db-path $DBDIR/$sfx \
    --output "$OUTDIR/$SITE-$sfx.html" \
    ...
You get the idea. Further notes Migration from older versions With the --persist option, GoAccess keeps all the information from the logs in a database, so that it can re-use it later. In prior versions, GoAccess used the Tokyo Cabinet key-value store for that. However starting from v1.4, GoAccess dropped this dependency and now uses its own database format. As a result, the previous database can't be used anymore, you will have to remove it and restart from zero. At the moment there is no way to convert the data from the old database to the new one. If you're interested, this is discussed upstream at [#1783][bug-1783]. Another thing that changed with this new version is the name for some of the command-line options. For example, --load-from-disk was dropped in favor of --restore, and --keep-db-files became --persist. So you'll have to look at the documentation a bit, and update your script(s) accordingly. Other ways to use GoAccess It's also possible to do it completely differently. You could keep GoAccess running, pretty much like a daemon, with the --real-time-html option, and have it process the logs continuously, rather than calling it on a regular basis. It's also possible to see the GoAccess report straight in the terminal, thanks to libncurses, rather than creating a HTML report. And much more, GoAccess is packed with features. Conclusion I hope that this tutorial helped some of you folks. Feel free to drop an e-mail for comments.

8 August 2020

Holger Levsen: 20200808-debconf8

DebConf8 This tshirt is 12 years old and from DebConf8. DebConf8 was my 6th DebConf and took place in Mar de la Plata, Argentina. Also this is my 6th post in this series of posts about DebConfs and for the last two days for the first time I failed my plan to do one post per day. And while two days ago I still planned to catch up on this by doing more than one post in a day, I have now decided to give in to realities, which mostly translates to sudden fantastic weather in Hamburg and other summer related changes in life. So yeah, I still plan to do short posts about all the DebConfs I was lucky to attend, but there might be days without a blog post. Anyhow, Mar de la Plata. When we held DebConf in Argentina it was winter there, meaning locals and other folks would wear jackets, scarfs, probably gloves, while many Debian folks not so much. Andreas Tille freaked out and/or amazed local people by going swimming in the sea every morning. And when I told Stephen Gran that even I would find it a bit cold with just a tshirt he replied "na, the weather is fine, just like british summer", while it was 14 celcius and mildly raining. DebConf8 was the first time I've met Valessio Brito, who I had worked together since at least DebConf6. That meeting was really super nice, Valessio is such a lovely person. Back in 2008 however, there was just one problem: his spoken English was worse than his written one, and that was already hard to parse sometimes. Fast forward eleven years to Curitiba last year and boom, Valessio speaks really nice English now. And, you might wonder why I'm telling this, especially if you were exposed to my Spanish back then and also now. So my point in telling this story about Valessio is to illustrate two things: a.) one can contribute to Debian without speaking/writing much English, Valessio did lots of great artwork since DebConf6 and b.) one can learn English by doing Debian stuff. It worked for me too! During set up of the conference there was one very memorable moment, some time after the openssl maintainer, Kurt Roeckx arrived at the venue: Shortly before DebConf8 Luciano Bello, from Argentina no less, had found CVE-2008-0166 which basically compromised the security of sshd of all Debian and Ubuntu installations done in the last 4 years (IIRC two Debian releases were affected) and which was commented heavily and noticed everywhere. So poor Kurt arrived and wondered whether we would all hate him, how many toilets he would have to clean and what not... And then, someone rather quickly noticed this, approached some people and suddenly a bunch of people at DebConf were group-hugging Kurt and then we were all smiling and continuing doing set up of the conference. That moment is one of my most joyful memories of all DebConfs and partly explains why I remember little about the conference itself, everything else pales in comparison and most things pale over the years anyway. As I remember it, the conference ran very smoothly in the end, despite quite some organisational problems right before the start. But as usual, once the geeks arrive and are happily geeking, things start to run smooth, also because Debian people are kind and smart and give hands and brain were needed. And like other DebConfs, Mar de la Plata also had moments which I want to share but I will only hint about, so it's up to you to imagine the special leaves which were brought to that cheese and wine party! ;-) Update: added another xkcd link, spelled out Kurt's name after talking to him and added a link to a video of the group hug.

7 August 2020

Antonio Terceiro: When community service providers fail

I'm starting a new blog, and instead of going into the technical details on how it's made, or on how I migrated my content from the previous ones, I want focus on why I did it. It's been a while since I have written a blog post. I wanted to get back into it, but also wanted to finally self-host my blog/homepage because I have been let down before. And sadly, I was not let down by a for-profit, privacy invading corporation, but by a free software organization. The sad story of wiki.softwarelivre.org My first blog that was hosted in a blog engine written by me, which was hosted in a TWiki, and later Foswiki instance previously available at wiki.softwarelivre.org, hosted by ASL.org. I was the one who introduced the tool to the organization in the first place. I had come from a previous, very fruitful experience on the use of wikis for creation of educational material while in university, which ultimately led me to become a core TWiki, and then Foswiki developer. In 2004, I had just moved to Porto Alegre, got involved in ASL.org, and there was a demand for a tool like that. 2 years later, I left Porto Alegre, and some time after that the daily operations of ASL.org when it became clear that it was not really prepared for remote participation. I was still maintaing the wiki software and the OS for quite some years after, until I wasn't anymore. In 2016, the server that hosted it went haywire, and there were no backups. A lot of people and free software groups lost their content forever. My blog was the least important content in there. To mention just a few examples, here are some groups that lost their content in there: Some of this can still be reached via the Internet Archive Wayback Machine, but that is only useful for recovering content, not for it to be used by the public. The announced tragedy of softwarelivre.org My next blog after that was hosted at softwarelivre.org, a Noosfero instance also hosted by ASL.org. When it was introduced in 2010, this Noosfero instance became responsible for the main domain softwarelivre.org name. This was a bold move by ASL.org, and was a demonstration of trust in a local free software project, led by a local free software cooperative (Colivre). I was a lead developer in the Noosfero project for a long time, and I was also involved in maintaining that server as well. However, for several years there is little to no investment in maintaining that service. I already expect that it will probably blow up at some point as the wiki did, or that it may be just shut down on purpose. On the responsibility of organizations Today, a large part of wast mot people consider "the internet" is controlled by a handful of corporations. Most popular services on the internet might look like they are gratis (free as in beer), but running those services is definitely not without costs. So when you use services provided by for-profit companies and are not paying for them with money, you are paying with your privacy and attention. Society needs independent organizations to provide alternatives. The market can solve a part of the problem by providing ethical services and charging for them. This is legitimate, and as long as there is transparency about how peoples' data and communications are handled, there is nothing wrong with it. But that only solves part of the problem, as there will always be people who can't afford to pay, and people and communities who can afford to pay, but would rather rely on a nonprofit. That's where community-based services, provided by nonprofits, are also important. We should have more of them, not less. So it makes me worry to realize ASL.org left the community in the dark. Losing the wiki wasn't even the first event of its kind, as the listas.softwarelivre.org mailing list server, with years and years of community communications archived in it, broke with no backups in 2012. I do not intend to blame the ASL.org leadership personally, they are all well meaning and good people. But as an organization, it failed to recognize the importance of this role of service provider. I can even include myself in it: I was member of the ASL.org board some 15 years ago; I was involved in the deployment of both the wiki and Noosfero, the former as a volunteer and the later professionally. Yet, I did nothing to plan the maintenance of the infrastructure going forward. When well meaning organizations fail, people who are not willing to have their data and communications be exploited for profit are left to their own devices. I can afford a virtual private server, and have the technical knowledge to import my old content into a static website generator, so I did it. But what about all the people who can't, or don't? Of course, these organizations have to solve the challenge of being sustainable, and being able to pay professionals to maintain the services that the community relies on. We should be thankful to these organizations, and their leadership needs to recognize the importance of those services, and actively plan for them to be kept alive.

Antonio Terceiro: When community service providers fail

I'm starting a new blog, and instead of going into the technical details on how it's made, or on how I migrated my content from the previous ones, I want to focus on why I did it. It's been a while since I have written a blog post. I wanted to get back into it, but also wanted to finally self-host my blog/homepage because I have been let down before. And sadly, I was not let down by a for-profit, privacy invading corporation, but by a free software organization. The sad story of wiki.softwarelivre.org My first blog that was hosted in a blog engine written by me, which was hosted in a TWiki, and later Foswiki instance previously available at wiki.softwarelivre.org, hosted by ASL.org. I was the one who introduced the tool to the organization in the first place. I had come from a previous, very fruitful experience on the use of wikis for creation of educational material while in university, which ultimately led me to become a core TWiki, and then Foswiki developer. In 2004, I had just moved to Porto Alegre, got involved in ASL.org, and there was a demand for a tool like that. 2 years later, I left Porto Alegre, and some time after that the daily operations of ASL.org when it became clear that it was not really prepared for remote participation. I was still maintaing the wiki software and the OS for quite some years after, until I wasn't anymore. In 2016, the server that hosted it went haywire, and there were no backups. A lot of people and free software groups lost their content forever. My blog was the least important content in there. To mention just a few examples, here are some groups that lost their content in there: Some of this can still be reached via the Internet Archive Wayback Machine, but that is only useful for recovering content, not for it to be used by the public. The announced tragedy of softwarelivre.org My next blog after that was hosted at softwarelivre.org, a Noosfero instance also hosted by ASL.org. When it was introduced in 2010, this Noosfero instance became responsible for the main domain softwarelivre.org name. This was a bold move by ASL.org, and was a demonstration of trust in a local free software project, led by a local free software cooperative (Colivre). I was a lead developer in the Noosfero project for a long time, and I was also involved in maintaining that server as well. However, for several years there is little to no investment in maintaining that service. I already expect that it will probably blow up at some point as the wiki did, or that it may be just shut down on purpose. On the responsibility of organizations Today, a large part of wast mot people consider "the internet" is controlled by a handful of corporations. Most popular services on the internet might look like they are gratis (free as in beer), but running those services is definitely not without costs. So when you use services provided by for-profit companies and are not paying for them with money, you are paying with your privacy and attention. Society needs independent organizations to provide alternatives. The market can solve a part of the problem by providing ethical services and charging for them. This is legitimate, and as long as there is transparency about how peoples' data and communications are handled, there is nothing wrong with it. But that only solves part of the problem, as there will always be people who can't afford to pay, and people and communities who can afford to pay, but would rather rely on a nonprofit. That's where community-based services, provided by nonprofits, are also important. We should have more of them, not less. So it makes me worry to realize ASL.org left the community in the dark. Losing the wiki wasn't even the first event of its kind, as the listas.softwarelivre.org mailing list server, with years and years of community communications archived in it, broke with no backups in 2012. I do not intend to blame the ASL.org leadership personally, they are all well meaning and good people. But as an organization, it failed to recognize the importance of this role of service provider. I can even include myself in it: I was member of the ASL.org board some 15 years ago; I was involved in the deployment of both the wiki and Noosfero, the former as a volunteer and the later professionally. Yet, I did nothing to plan the maintenance of the infrastructure going forward. When well meaning organizations fail, people who are not willing to have their data and communications be exploited for profit are left to their own devices. I can afford a virtual private server, and have the technical knowledge to import my old content into a static website generator, so I did it. But what about all the people who can't, or don't? Of course, these organizations have to solve the challenge of being sustainable, and being able to pay professionals to maintain the services that the community relies on. We should be thankful to these organizations, and their leadership needs to recognize the importance of those services, and actively plan for them to be kept alive.

3 August 2020

Arnaud Rebillout: GoAccess 1.4, a detailed tutorial

GoAccess v1.4 was just released a few weeks ago! Let's take this chance to write a loooong tutorial. We'll go over every steps to install and operate GoAccess. This is a tutorial aimed at those who don't play sysadmin every day, and that's why it's so long, I did my best to provide thorough explanations all along, so that it's more than just a "copy-and-paste" kind of tutorial. And for those who do play sysadmin everyday: please try not to fall asleep while reading, and don't hesitate to drop me an e-mail if you spot anything inaccurate in here. Thanks! Introduction So what's GoAccess already? GoAccess is a web log analyzer, and it allows you to visualize the traffic for your website, and get to know a bit more about your visitors: how many visitors and hits, for which pages, coming from where (geolocation, operating system, web browser...), etc... It does so by parsing the access logs from your web server, be it Apache, NGINX or whatever. GoAccess gives you different options to display the statistics, and in this tutorial we'll focus on producing a HTML report. Meaning that you can see the statistics for your website straight in your web browser, under the form of a single HTML page. For an example, you can have a look at the stats of my blog here: http://goaccess.arnaudr.io. GoAccess is written in C, it has very few dependencies, it had been around for about 10 years, and it's distributed under the MIT license. Assumptions This tutorial is about installing and configuring, so I'll assume that all the commands are run as root. I won't prefix each of them with sudo. I use the Apache web server, running on a Debian system. I don't think it matters so much for this tutorial though. If you're using NGINX it's fine, you can keep reading. Also, I will just use the name SITE for the name of the website that we want to analyze with GoAccess. Just replace that with the real name of your site. I also assume the following locations for your stuff: If you have your stuff in /srv/SITE/ log,www instead, no worries, just adjust the paths accordingly, I bet you can do it. Installation The latest version of GoAccess is v1.4, and it's not yet available in the Debian repositories. So for this part, you can follow the instructions from the official GoAccess download page. Install steps are explained in details, so there's nothing left for me to say :) When this is done, let's get started with the basics. We're talking about the latest version v1.4 here, let's make sure: 
$ goaccess --version
GoAccess - 1.4.
...
Now let's try to create a HTML report. I assume that you already have a website up and running. GoAccess needs to parse the access logs. These logs are optional, they might or might not be created by your web server, depending on how it's configured. Usually, these log files are named access.log, unsurprisingly. You can check if those logs exist on your system by running this command: 
find /var/log -name access.log
Another important thing to know is that these logs can be in different formats. In this tutorial we'll assume that we work with the combined log format, because it seems to be the most common default. To check what kind of access logs your web server produces, you must look at the configuration for your site. For an Apache web server, you should have such a line in the file /etc/apache2/sites-enabled/SITE.conf: 
CustomLog $ APACHE_LOG_DIR /SITE/access.log combined
For NGINX, it's quite similar. The configuration file would be something like /etc/nginx/sites-enabled/SITE, and the line to enable access logs would be something like: 
access_log /var/log/nginx/SITE/access.log
Note that NGINX writes the access logs in the combined format by default, that's why you don't see the word combined anywhere in the line above: it's implicit. Alright, so from now on we assume that yes, you have access log files available, and yes, they are in the combined log format. If that's the case, then you can already run GoAccess and generate a report, for example for the log file /var/log/apache2/access.log 
goaccess \
    --log-format COMBINED \
    --output /tmp/report.html \
    /var/log/apache2/access.log
It's possible to give GoAccess more than one log files to process, so if you have for example the file access.log.1 around, you can use it as well: 
goaccess \
    --log-format COMBINED \
    --output /tmp/report.html \
    /var/log/apache2/access.log \
    /var/log/apache2/access.log.1
If GoAccess succeeds (and it should), you're on the right track! All is left to do to complete this test is to have a look at the HTML report created. It's a single HTML page, so you can easily scp it to your machine, or just move it to the document root of your site, and then open it in your web browser. Looks good? So let's move on to more interesting things. Web server configuration This part is very short, because in terms of configuration of the web server, there's very little to do. As I said above, the only thing you want from the web server is to create access log files. Then you want to be sure that GoAccess and your web server agree on the format for these files. In the part above we used the combined log format, but GoAccess supports many other common log formats out of the box, and even allows you to parse custom log formats. For more details, refer to the option --log-format in the GoAccess manual page. Another common log format is named, well, common. It even has its own Wikipedia page. But compared to combined, the common log format contains less information, it doesn't include the referrer and user-agent values, meaning that you won't have it in the GoAccess report. So at this point you should understand that, unsurprisingly, GoAccess can only tell you about what's in the access logs, no more no less. And that's all in term of web server configuration. Configuration to run GoAccess unprivileged Now we're going to create a user and group for GoAccess, so that we don't have to run it as root. The reason is that, well, for everything running unattended on your server, the less code runs as root, the better. It's good practice and common sense. In this case, GoAccess is simply a log analyzer. So it just needs to read the logs files from your web server, and there is no need to be root for that, an unprivileged user can do the job just as well, assuming it has read permissions on /var/log/apache2 or /var/log/nginx. The log files of the web server are usually part of the adm group (though it might depend on your distro, I'm not sure). This is something you can check easily with the following command: 
ls -l /var/log   grep -e apache2 -e nginx
As a result you should get something like that: 
drwxr-x--- 2 root adm 20480 Jul 22 00:00 /var/log/apache2/
And as you can see, the directory apache2 belongs to the group adm. It means that you don't need to be root to read the logs, instead any unprivileged user that belongs to the group adm can do it. So, let's create the goaccess user, and add it to the adm group: 
adduser --system --group --no-create-home goaccess
addgroup goaccess adm
And now, let's run GoAccess unprivileged, and verify that it can still read the log files: 
setpriv \
    --reuid=goaccess --regid=goaccess \
    --init-groups --inh-caps=-all \
    -- \
    goaccess \
    --log-format COMBINED \
    --output /tmp/report2.html \
    /var/log/apache2/access.log
setpriv is the command used to drop privileges. The syntax is quite verbose, it's not super friendly for tutorials, but don't be scared and read the manual page to learn what it does. In any case, this command should work, and at this point, it means that you have a goaccess user ready, and we'll use it to run GoAccess unprivileged. Integration, option A - Run GoAccess once a day, from a logrotate hook In this part we wire things together, so that GoAccess processes the log files once a day, adds the new logs to its internal database, and generates a report from all that aggregated data. The result will be a single HTML page. Introducing logrotate In order to do that, we'll use a logrotate hook. logrotate is a little tool that should already be installed on your server, and that runs once a day, and that is in charge of rotating the log files. "Rotating the logs" means moving access.log to access.log.1 and so on. With logrotate, a new log file is created every day, and log files that are too old are deleted. That's what prevents your logs from filling up your disk basically :) You can check that logrotate is indeed installed and enabled with this command (assuming that your init system is systemd): 
systemctl status logrotate.timer
What's interesting for us is that logrotate allows you to run scripts before and after the rotation is performed, so it's an ideal place from where to run GoAccess. In short, we want to run GoAccess just before the logs are rotated away, in the prerotate hook. But let's do things in order. At first, we need to write a little wrapper script that will be in charge of running GoAccess with the right arguments, and that will process all of your sites. The wrapper script This wrapper is made to process more than one site, but if you have only one site it works just as well, of course. So let me just drop it on you like that, and I'll explain afterward. Here's my wrapper script: 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
#!/bin/bash
# Process log files /var/www/apache2/SITE/access.log,
# only if /var/lib/goaccess-db/SITE exists.
# Create HTML reports in $1, a directory that must exist.
set -eu
OUTDIR=
LOGDIR=/var/log/apache2
DBDIR=/var/lib/goaccess-db
fail()   echo >&2 "$@"; exit 1;  
[ $# -eq 1 ]   fail "Usage: $(basename $0) OUTPUT_DIRECTORY"
OUTDIR=$1
[ -d "$OUTDIR" ]   fail "'$OUTDIR' is not a directory"
[ -d "$LOGDIR" ]   fail "'$LOGDIR' is not a directory"
[ -d "$DBDIR"  ]   fail "'$DBDIR' is not a directory"
for d in $(find "$LOGDIR" -mindepth 1 -maxdepth 1 -type d); do
    site=$(basename "$sitedir")
    dbdir=$DBDIR/$site
    logfile=$d/access.log
    outfile=$OUTDIR/$site.html
    if [ ! -d "$dbdir" ]   [ ! -e "$logfile" ]; then
        echo "  Skipping site '$site'"
        continue
    else
        echo "  Processing site '$site'"
    fi
    setpriv \
        --reuid=goaccess --regid=goaccess \
        --init-groups --inh-caps=-all \
        -- \
    goaccess \
        --agent-list \
        --anonymize-ip \
        --persist \
        --restore \
        --config-file /etc/goaccess/goaccess.conf \
        --db-path "$dbdir" \
        --log-format "COMBINED" \
        --output "$outfile" \
        "$logfile"
done
So you'd install this script at /usr/local/bin/goaccess-wrapper for example, and make it executable: 
chmod +x /usr/local/bin/goaccess-wrapper
A few things to note: As is, the script makes the assumption that the logs for your site are logged in a sub-directory /var/log/apache2/SITE/. If it's not the case, adjust that in the wrapper accordingly. The name of this sub-directory is then used to find the GoAccess database directory /var/lib/goaccess-db/SITE/. This directory is expected to exist, meaning that if you don't create it yourself, the wrapper won't process this particular site. It's a simple way to control which sites are processed by this GoAccess wrapper, and which sites are not. So if you want goaccess-wrapper to process the site SITE, just create a directory with the name of this site under /var/lib/goaccess-db: 
mkdir -p /var/lib/goaccess-db/SITE
chown goaccess:goaccess /var/lib/goaccess-db/SITE
Now let's create an output directory: 
mkdir /tmp/goaccess-reports
chown goaccess:goaccess /tmp/goaccess-reports
And let's give a try to the wrapper script: 
goaccess-wrapper /tmp/goaccess-reports
ls /tmp/goaccess-reports
Which should give you: 
SITE.html
At the same time, you can check that GoAccess populated the database with a bunch of files: 
ls /var/lib/goaccess-db/SITE
Setting up the logrotate prerotate hook At this point, we have the wrapper in place. Let's now add a pre-rotate hook so that goaccess-wrapper runs once a day, just before the logs are rotated away. The logrotate config file for Apache2 is located at /etc/logrotate.d/apache2, and for NGINX it's at /etc/logrotate.d/nginx. Among the many things you'll see in this file, here's what is of interest for us: In the config file, there is also this snippet: 
prerotate
    if [ -d /etc/logrotate.d/httpd-prerotate ]; then \
        run-parts /etc/logrotate.d/httpd-prerotate; \
    fi; \
endscript
It indicates that scripts in the directory /etc/logrotate.d/httpd-prerotate/ will be executed before the rotation takes place. Refer to the man page run-parts(8) for more details... Putting all of that together, it means that logs from the web server are rotated once a day, and if we want to run scripts just before the rotation, we can just drop them in the httpd-prerotate directory. Simple, right? Let's first create this directory if it doesn't exist: 
mkdir -p /etc/logrotate.d/httpd-prerotate/
And let's create a tiny script at /etc/logrotate.d/httpd-prerotate/goaccess: 
1
2
#!/bin/sh
exec goaccess-wrapper /tmp/goaccess-reports
Don't forget to make it executable: 
chmod +x /etc/logrotate.d/httpd-prerotate/goaccess
As you can see, the only thing that this script does is to invoke the wrapper with the right argument, ie. the output directory for the HTML reports that are generated. And that's all. Now you can just come back tomorrow, check the logs, and make sure that the hook was executed and succeeded. For example, this kind of command will tell you quickly if it worked: 
journalctl   grep logrotate
Integration, option B - Run GoAccess once a day, from a systemd service OK so we've just seen how to use a logrotate hook. One downside with that is that we have to drop privileges in the wrapper script, because logrotate runs as root, and we don't want to run GoAccess as root. Hence the rather convoluted syntax with setpriv. Rather than embedding this kind of thing in a wrapper script, we can instead run the wrapper script from a [systemd][] service, and define which user runs the wrapper straight in the systemd service file. Introducing systemd niceties So we can create a systemd service, along with a systemd timer that fires daily. We can then set the user and group that execute the script straight in the systemd service, and there's no need for setpriv anymore. It's a bit more streamlined. We can even go a bit further, and use systemd parameterized units (also called templates), so that we have one service per site (instead of one service that process all of our sites). That will simplify the wrapper script a lot, and it also looks nicer in the logs. With this approach however, it seems that we can't really run exactly before the logs are rotated away, like we did in the section above. But that's OK. What we'll do is that we'll run once a day, no matter the time, and we'll just make sure to process both log files access.log and access.log.1 (ie. the current logs and the logs from yesterday). This way, we're sure not to miss any line from the logs. Note that GoAccess is smart enough to only consider newer entries from the log files, and discard entries that are already in the database. In other words, it's safe to parse the same log file more than once, GoAccess will do the right thing. For more details see "INCREMENTAL LOG PROCESSING" from man goaccess. systemd]: https://freedesktop.org/wiki/Software/systemd/ Implementation And here's how it all looks like. First, a little wrapper script for GoAccess: 
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
#!/bin/bash
# Usage: $0 SITE DBDIR LOGDIR OUTDIR
set -eu
SITE=$1
DBDIR=$2
LOGDIR=$3
OUTDIR=$4
LOGFILES=()
for ext in log log.1; do
    logfile="$LOGDIR/access.$ext"
    [ -e "$logfile" ] && LOGFILES+=("$logfile")
done
if [ $ #LOGFILES[@]  -eq 0 ]; then
    echo "No log files in '$LOGDIR'"
    exit 0
fi
goaccess \
    --agent-list \
    --anonymize-ip \
    --persist \
    --restore \
    --config-file /etc/goaccess/goaccess.conf \
    --db-path "$DBDIR" \
    --log-format "COMBINED" \
    --output "$OUTDIR/$SITE.html" \
    "$ LOGFILES[@] "
This wrapper does very little. Actually, the only thing it does is to check for the existence of the two log files access.log and access.log.1, to be sure that we don't ask GoAccess to process a file that does not exist (GoAccess would not be happy about that). Save this file under /usr/local/bin/goaccess-wrapper, don't forget to make it executable: 
chmod +x /usr/local/bin/goaccess-wrapper
Then, create a systemd parameterized unit file, so that we can run this wrapper as a systemd service. Save it under /etc/systemd/system/goaccess@.service: 
[Unit]
Description=Update GoAccess report - %i
ConditionPathIsDirectory=/var/lib/goaccess-db/%i
ConditionPathIsDirectory=/var/log/apache2/%i
ConditionPathIsDirectory=/tmp/goaccess-reports
PartOf=goaccess.service
[Service]
Type=oneshot
User=goaccess
Group=goaccess
Nice=19
ExecStart=/usr/local/bin/goaccess-wrapper \
 %i \
 /var/lib/goaccess-db/%i \
 /var/log/apache2/%i \
 /tmp/goaccess-reports
So, what is a systemd parameterized unit? It's a service to which you can pass an argument when you enable it. The %i in the unit definition will be replaced by this argument. In our case, the argument will be the name of the site that we want to process. As you can see, we use the directive ConditionPathIsDirectory= extensively, so that if ever one of the required directories does not exist, the unit will just be skipped (and marked as such in the logs). It's a graceful way to fail. We run the wrapper as the user and group goaccess, thanks to User= and Group=. We also use Nice= to give a low priority to the process. At this point, it's already possible to test. Just make sure that you created a directory for the GoAccess database: 
mkdir -p /var/lib/goaccess-db/SITE
chown goaccess:goaccess /var/lib/goaccess-db/SITE
Also make sure that the output directory exists: 
mkdir /tmp/goaccess-reports
chown goaccess:goaccess /tmp/goaccess-reports
Then reload systemd and fire the unit to see if it works: 
systemctl daemon-reload
systemctl start goaccess@SITE.service
journalctl   tail
And that should work already. As you can see, the argument, SITE, is passed in the systemctl start command. We just append it after the @, in the name of the unit. Now, let's create another GoAccess service file, which sole purpose is to group all the parameterized units together, so that we can start them all in one go. Note that we don't use a systemd target for that, because ultimately we want to run it once a day, and that would not be possible with a target. So instead we use a dummy oneshot service. So here it is, saved under /etc/systemd/system/goaccess.service: 
[Unit]
Description=Update GoAccess reports
Requires= \
 goaccess@SITE1.service \
 goaccess@SITE2.service
[Service]
Type=oneshot
ExecStart=true
As you can see, we simply list the sites that we want to process in the Requires= directive. In this example we have two sites named SITE1 and SITE2. Let's ensure that everything is still good: 
systemctl daemon-reload
systemctl start goaccess.service
journalctl   tail
Check the logs, both sites SITE1 and SITE2 should have been processed. And finally, let's create a timer, so that systemd runs goaccess.service once a day. Save it under /etc/systemd/system/goaccess.timer. 
[Unit]
Description=Daily update of GoAccess reports
[Timer]
OnCalendar=daily
RandomizedDelaySec=1h
Persistent=true
[Install]
WantedBy=timers.target
Finally, enable the timer: 
systemctl daemon-reload
systemctl enable --now goaccess.timer
At this point, everything should be OK. Just come back tomorrow and check the logs with something like: 
journalctl   grep goaccess
Last word: if you have only one site to process, of course you can simplify, for example you can hardcode all the paths in the file goaccess.service instead of using a parameterized unit. Up to you. Daily operations So in this part, we assume that you have GoAccess all setup and running, once a day or so. Let's just go over a few things worth noting. Serve your report Up to now in this tutorial, we created the reports in /tmp/goaccess-reports, but that was just for the sake of the example. You will probably want to save your reports in a directory that is served by your web server, so that, well, you can actually look at it in your web browser, that was the point, right? So how to do that is a bit out of scope here, and I guess that if you want to monitor your website, you already have a website, so you will have no trouble serving the GoAccess HTML report. However there's an important detail to be aware of: GoAccess shows all the IP addresses of your visitors in the report. As long as the report is private it's OK, but if ever you make your GoAccess report public, then you should definitely invoke GoAccess with the option --anonymize-ip. Keep an eye on the logs In this tutorial, the reports we create, along with the GoAccess databases, will grow bigger every day, forever. It also means that the GoAccess processing time will grow a bit each day. So maybe the first thing to do is to keep an eye on the logs, to see how long it takes to GoAccess to do its job every day. Also, maybe you'd like to keep an eye on the size of the GoAccess database with: 
du -sh /var/lib/goaccess-db/SITE
If your site has few visitors, I suspect it won't be a problem though. You could also be a bit pro-active in preventing this problem in the future, and for example you could break the reports into, say, monthly reports. Meaning that every month, you would create a new database in a new directory, and also start a new HTML report. This way you'd have monthly reports, and you make sure to limit the GoAccess processing time, by limiting the database size to a month. This can be achieved very easily, by including something like YEAR-MONTH in the database directory, and in the HTML report. You can handle that automatically in the wrapper script, for example: 
sfx=$(date +'%Y-%m')
mkdir -p $DBDIR/$sfx
goaccess \
    --db-path $DBDIR/$sfx \
    --output "$OUTDIR/$SITE-$sfx.html" \
    ...
You get the idea. Further notes Migration from older versions With the --persist option, GoAccess keeps all the information from the logs in a database, so that it can re-use it later. In prior versions, GoAccess used the Tokyo Cabinet key-value store for that. However starting from v1.4, GoAccess dropped this dependency and now uses its own database format. As a result, the previous database can't be used anymore, you will have to remove it and restart from zero. At the moment there is no way to convert the data from the old database to the new one. If you're interested, this is discussed upstream at [#1783][bug-1783]. Another thing that changed with this new version is the name for some of the command-line options. For example, --load-from-disk was dropped in favor of --restore, and --keep-db-files became --persist. So you'll have to look at the documentation a bit, and update your script(s) accordingly. Other ways to use GoAccess It's also possible to do it completely differently. You could keep GoAccess running, pretty much like a daemon, with the --real-time-html option, and have it process the logs continuously, rather than calling it on a regular basis. It's also possible to see the GoAccess report straight in the terminal, thanks to libncurses, rather than creating a HTML report. And much more, GoAccess is packed with features. Conclusion I hope that this tutorial helped some of you folks. Feel free to drop an e-mail for comments.

2 August 2020

Enrico Zini: Toxic positivity links

The Oppression of the Positivity Movement
politics privilege archive.org
2020-08-03
That which we do not bring to consciousness appears in our lives as fate. Carl Jung
The Tyrannical Culture of Positivity
health privilege archive.org
2020-08-03
Emotional support of others can take the form of surface-level consolation. But compassion means being willing to listen and feel, even when it's uncomfortable.
How Positive Thinking Can Do More Harm Than Good
health privilege archive.org
2020-08-03
Ultimately, the driving force behind the power of positive thinking meme is the word power. But what about those whose bodies are not powerful? What about those who are vulnerable? What about those who are tired, isolated, and struggling? What about those who are ill? What about those who lack
Toxic Positivity - Tanglaw Mental Health hindsight is so 2020
health privilege archive.org
2020-08-03
I have often been dismissive or unhelpful when someone close to me was dealing with painful circumstances, having learned to accentuate the positive. In the more recent past, I have recognized these behavioral patterns as part of what some mental health professionals term, toxic positivity.
Toxic Positivity: The Dark Side of Positive Vibes
health privilege archive.org
2020-08-03
Toxic positivity is the overgeneralization of a happy, optimistic state resulting in the denial & invalidation of the authentic human emotional experience.

Holger Levsen: 20200802-debconf4

DebConf4 This tshirt is 16 years old and from DebConf4. Again, I should probably wash it at 60 celcius for once... DebConf4 was my 2nd DebConf and took place in Porto Alegre, Brasil. Like many DebConfs, it was a great opportunity to meet people: I remember sitting in the lobby of the venue and some guy asked me what I did in Debian and I told him about my little involvements and then asked him what he was doing, and he told me he wanted to become involved in Debian again, after getting distracted away. His name was Ian Murdock... DebConf4 also had a very cool history session in the hallway track (IIRC, but see below) with Bdale Garbee, Ian Jackson and Ian Murdock and with a young student named Biella Coleman busy writing notes. That same hallway also saw the kickoff meeting of the Debian Women project, though sadly today http://tinc.debian.net ("there's no cabal") only shows an apache placeholder page and not a picture of that meeting. DebCon4 was also the first time I got a bit involved in preparing DebConf, together with Jonas Smedegaard I've set up some computers there, using FAI. I had no idea that this was the start of me contributing to DebConfs for text ten years. And of course I also saw some talks, including one which I really liked, which then in turn made me notice there were no people doing video recordings, which then lead to something... I missed the group picture of this one. I guess it's important to me to mention it because I've met very wonderful people at this DebConf... (some mentioned in this post, some not. You know who you are!) Afterwards some people stayed in Porto Alegre for FISL, where we saw Lawrence Lessing present Creative Commons to the world for the first time. On the flight back I sat next to a very friendly guy from Poland and we talked almost the whole flight and then we never saw each other again, until 15 years later in Asia... Oh, and then, after DebConf4, I used IRC for the first time. And stayed in the #debconf4 IRC channel for quite some years :) Finally, DebConf4 and more importantly FISL, which was really big (5000 people?) and after that, the wizard of OS conference in Berlin (which had a very nice talk about Linux in different places in the world, illustrating the different states of 'first they ignore you, then they laugh at you, then they fight you, then you win'), made me quit my job at a company supporting Windows- and Linux-setups as I realized I'd better start freelancing with Linux-only jobs. So, once again, my life would have been different if I would not have attended these events! Note: yesterdays post about DebConf3 was thankfully corrected twice. This might well happen to this post too! :)

28 July 2020

Russ Allbery: Review: The City in the Middle of the Night

Review: The City in the Middle of the Night, by Charlie Jane Anders
Publisher: Tor
Copyright: February 2019
Printing: February 2020
ISBN: 1-4668-7113-X
Format: Kindle
Pages: 366
January is a tidally-locked planet divided between permanent night and permanent day, an unfortunate destination for a colony starship. Now, humans cling to a precarious existence along the terminator, huddling in two wildly different cities and a handful of smaller settlements, connected by a road through the treacherous cold. The novel opens with Sophie, a shy university student from the dark side of the city of Xiosphant. She has an overwhelming crush on Bianca, her high-class, self-confident roommate and one of the few people in her life to have ever treated her with compassion and attention. That crush, and her almost non-existent self-esteem, lead her to take the blame for Bianca's petty theft, resulting in what should have been a death sentence. Sophie survives only because she makes first contact with a native intelligent species of January, one that the humans have been hunting for food and sport. Sadly, I think this is enough Anders for me. I've now bounced off two of her novels, both for structural reasons that I think go deeper than execution and indicate a fundamental mismatch between what Anders wants to do as an author and what I'm looking for as a reader. I'll talk more about what this book is doing in a moment, but I have to start with Bianca and Sophie. It's difficult for me to express how much I loathed this relationship and how little I wanted to read about it. It took me about five pages to peg Bianca as a malignant narcissist and Sophie's all-consuming crush as dangerous codependency. It took the entire book for Sophie to figure out how awful Bianca is to her, during which Bianca goes through the entire abusive partner playbook of gaslighting, trivializing, contingent affection, jealous rage, and controlling behavior. And meanwhile Sophie goes back to her again, and again, and again, and again. If I hadn't been reading this book on a Kindle, I think it would have physically hit a wall after their conversation in the junkyard. This is truly a matter of personal taste and preference. This is not an unrealistic relationship; this dynamic happens in life all too often. I'm sure there is someone for whom reading about Sophie's spectacularly poor choices is affirming or cathartic. I've not personally experienced this sort of relationship, which doubtless matters. But having empathy for someone who is making awful and self-destructive life decisions and trusting someone they should not be trusting and who is awful to them in every way is difficult work. Sophie is the victim of Bianca's abuse, but she does so many stupid and ill-conceived things in support of this twisted relationship that I found it very difficult to not get angry at her. Meanwhile, Anders writes Sophie as so clearly fragile and uncertain and devoid of a support network that getting angry at her is like kicking a puppy. The result for me was spending nearly an entire book in a deeply unpleasant state of emotional dissonance. I may be willing to go through that for a close friend, but in a work of fiction it's draining and awful and entirely not fun. The other viewpoint character had the opposite problem for me. Mouth starts the book as a traveling smuggler, the sole survivor of a group of religious travelers called the Citizens. She's practical, tough, and guarded. Beneath that, I think the intent was to show her as struggling to come to terms with the loss of her family and faith community. Her first goal in the book is to recover a recording of Citizen sacred scripture to preserve it and to reconnect with her past. This sounds interesting on the surface, but none of it gelled. Mouth never felt to me like someone from a faith community. She doesn't act on Citizen beliefs to any meaningful extent, she rarely talks about them, and when she does, her attitude is nostalgia without spirituality. When Mouth isn't pursuing goals that turn out to be meaningless, she aimlessly meanders through the story. Sophie at least has agency and makes some important and meaningful decisions. Mouth is just there, even when Anders does shattering things to her understanding of her past. Between Sophie and Bianca putting my shoulders up around my ears within the first few pages of the first chapter and failing to muster any enthusiasm for Mouth, I said the eight deadly words ("I don't care what happens to these people") about a hundred pages in and the book never recovered. There are parts of the world-building I did enjoy. The alien species that Sophie bonds with is not stunningly original, but it's a good (and detailed) take on one of the alternate cognitive and social models that science fiction has dreamed up. I was comparing the strangeness and dislocation unfavorably to China Mi ville's Embassytown while I was reading it, but in retrospect Anders's treatment is more decolonialized. Xiosphant's turn to Circadianism as their manifestation of order is a nicely understated touch, a believable political overreaction to the lack of a day/night cycle. That touch is significantly enhanced by Sophie's time working in a salon whose business model is to help Xiosphant residents temporarily forget about time. And what glimmers we got of politics on the colony ship and their echoing influence on social and political structures were intriguing. Even with the world-building, though, I want the author to be interested in and willing to expand the same bits of world-building that I'm engaged with. Anders didn't seem to be. The reader gets two contrasting cities along a road, one authoritarian and one libertine, which makes concrete a metaphor for single-axis political classification. But then Anders does almost nothing with that setup; it's just the backdrop of petty warlord politics, and none of the political activism of Bianca's student group seems to have relevance or theoretical depth. It's a similar shallowness as the religion of Mouth's Citizens: We get a few fragments of culture and religion, but without narrative exploration and without engagement from any of the characters. The way the crew of the Mothership was assembled seems to have led to a factional and racial caste system based on city of origin and technical expertise, but I couldn't tell you more than that because few of the characters seem to care. And so on. In short, the world-building that I wanted to add up to a coherent universe that was meaningful to the characters and to the plot seemed to be little more than window-dressing. Anders tosses in neat ideas, but they don't add up to anything. They're just background scenery for Bianca and Sophie's drama. The one thing that The City in the Middle of the Night does well is Sophie's nervous but excited embrace of the unknown. It was delightful to see the places where a typical protagonist would have to overcome a horror reaction or talk themselves through tradeoffs and where Sophie's reaction was instead "yes, of course, let's try." It provided an emotional strength to an extended first-contact exploration scene that made it liberating and heart-warming without losing the alienness. During that part of the book (in which, not coincidentally, Bianca does not appear), I was able to let my guard down and like Sophie for the first time, and I suspect that was intentional on Anders's part. But, overall, I think the conflict between Anders's story-telling approach and my preferences as a reader are mostly irreconcilable. She likes to write about people who make bad decisions and compound their own problems. In one of the chapters of her non-fiction book about writing that's being serialized on Tor.com she says "when we watch someone do something unforgivable, we're primed to root for them as they search desperately for an impossible forgiveness." This is absolutely not true for me; when I watch a character do something unforgivable, I want to see repudiation from the protagonists and ideally some clear consequences. When that doesn't happen, I want to stop reading about them and find something more enjoyable to do with my time. I certainly don't want to watch a viewpoint character insist that the person who is doing unforgivable things is the center of her life. If your preferences on character and story arc are closer to Anders's than mine, you may like this book. Certainly lots of people did; it was nominated for multiple awards and won the Locus Award for Best Science Fiction Novel. But despite the things it did well, I had a truly miserable time reading it and am not anxious to repeat the experience. Rating: 4 out of 10

27 July 2020

Matthew Garrett: Filesystem deduplication is a sidechannel

First off - nothing I'm going to talk about in this post is novel or overly surprising, I just haven't found a clear writeup of it before. I'm not criticising any design decisions or claiming this is an important issue, just raising something that people might otherwise be unaware of.

With that out of the way: Automatic deduplication of data is a feature of modern filesystems like zfs and btrfs. It takes two forms - inline, where the filesystem detects that data being written to disk is identical to data that already exists on disk and simply references the existing copy rather than, and offline, where tooling retroactively identifies duplicated data and removes the duplicate copies (zfs supports inline deduplication, btrfs only currently supports offline). In a world where disks end up with multiple copies of cloud or container images, deduplication can free up significant amounts of disk space.

What's the security implication? The problem is that deduplication doesn't recognise ownership - if two users have copies of the same file, only one copy of the file will be stored[1]. So, if user a stores a file, the amount of free space will decrease. If user b stores another copy of the same file, the amount of free space will remain the same. If user b is able to check how much free space is available, user b can determine whether the file already exists.

This doesn't seem like a huge deal in most cases, but it is a violation of expected behaviour (if user b doesn't have permission to read user a's files, user b shouldn't be able to determine whether user a has a specific file). But we can come up with some convoluted cases where it becomes more relevant, such as law enforcement gaining unprivileged access to a system and then being able to demonstrate that a specific file already exists on that system. Perhaps more interestingly, it's been demonstrated that free space isn't the only sidechannel exposed by deduplication - deduplication has an impact on access timing, and can be used to infer the existence of data across virtual machine boundaries.

As I said, this is almost certainly not something that matters in most real world scenarios. But with so much discussion of CPU sidechannels over the past couple of years, it's interesting to think about what other features also end up leaking information in ways that may not be obvious.

(Edit to add: deduplication isn't enabled on zfs by default and is explicitly triggered on btrfs, so unless it's something you've enabled then this isn't something that affects you)

[1] Deduplication is usually done at the block level rather than the file level, but given zfs's support for variable sized blocks, identical files should be deduplicated even if they're smaller than the maximum record size

comment count unavailable comments

Russ Allbery: Review: Rise of the Warrior Cop

Review: Rise of the Warrior Cop, by Radley Balko
Publisher: PublicAffairs
Copyright: 2013
ISBN: 1-61039-212-4
Format: Kindle
Pages: 336
As the United States tries, in fits and starts, to have a meaningful discussion about long-standing police racism, brutality, overreach, corruption, and murder, I've realized that my theoretical understanding of the history of and alternative frameworks for law enforcement is woefully lacking. Starting with a book by a conservative white guy is not the most ideal of approaches, but it's what I already had on hand, and it won't be the last book I read and review on this topic. (Most of my research so far has been in podcast form. I don't review those here, but I can recommend Ezra Klein's interviews with Ta-Nehisi Coates, Paul Butler, and, most strongly, sujatha baliga.) Rise of the Warrior Cop is from 2013 and has had several moments of fame, no doubt helped by Balko's connections to the conservative and libertarian right. One of the frustrating facts of US politics is that critiques of the justice system from the right (and from white men) get more media attention than critiques from the left. That said, it's a generally well-respected book on the factual history of the topic, and police brutality and civil rights are among the points on which I have stopped-clock agreements with US libertarians. This book is very, very libertarian. In my callow youth, I was an ardent libertarian, so I've read a lot of US libertarian literature. It's a genre with its own conventions that become obvious when you read enough of it, and Rise of the Warrior Cop goes through them like a checklist. Use the Roman Republic (never the Roman Empire) as the starting point for any political discussion, check. Analyze the topic in the context of pre-revolutionary America, check. Spend considerable effort on discerning the opinions of the US founders on the topic since their opinions are always relevant to the modern world, check. Locate some point in the past (preferably before 1960) where the political issue was as good as it has ever been, check. Frame all changes since then as an erosion of rights through government overreach, check. Present your solution as a return to a previous era of respect for civil rights, check. Once you start recognizing the genre conventions, their prevalence in libertarian writing is almost comical. The framing chapters therefore leave a bit to be desired, but the meat of the book is a useful resource. Starting with the 1970s and its use as a campaigning tool by Nixon, Balko traces a useful history of the war on drugs. And starting with the 1980s, the number of cites to primary sources and the evidence of Balko's own research increases considerably. If you want to know how US police turned into military cosplayers with body armor, heavy weapons, and armored vehicles, this book provides a lot of context and history. One of the reasons why I view libertarians as allies of convenience on this specific issue is that drug legalization and disgust with the war on drugs have been libertarian issues for decades. Ideologically honest libertarians (and Balko appears to be one) are inherently skeptical of the police, so when the police overreach in an area of libertarian interest, they notice. Balko makes a solid argument, backed up with statistics, specific programs, legislation, and court cases, that the drug war and its accompanying lies about heavily-armed drug dealers and their supposed threat to police officers was the fuel for the growth of SWAT teams, no-knock search warrants, erosion of legal protections for criminal defendants, and de facto license for the police to ignore the scope and sometimes even the existence of warrants. This book is useful support for the argument that fears for the safety of officers underlying the militarization of police forces are imaginary. One telling point that Balko makes repeatedly and backs with statistical and anecdotal evidence is that the police generally do not use raid tactics on dangerous criminals. On the contrary, aggressive raids are more likely to be used on the least dangerous criminals because they're faster, they're fun for the police (they provide an adrenaline high and let them play with toys), and they're essentially risk-free. If the police believe someone is truly dangerous, they're more likely to use careful surveillance and to conduct a quiet arrest at an unexpected moment. The middle-of-the-night armed break-ins with battering rams, tear gas, and flash-bangs are, tellingly, used against the less dangerous suspects. This is part of Balko's overall argument that police equipment and tactics have become untethered from any realistic threat and have become cultural. He traces an acceleration of that trend to 9/11 and the resulting obsession with terrorism, which further opened the spigot of military hardware and "special forces" training. This became a point of competition between police departments, with small town forces that had never seen a terrorist and had almost no chance of a terrorist incident demanding their own armored vehicles. I've encountered this bizarre terrorism justification personally; one of the reasons my local police department gave in a public hearing for not having a policy against shooting at moving vehicles was "but what if terrorism?" I don't believe there has ever been a local terrorist attack. SWAT in such places didn't involve the special training or dedicated personnel of large city forces; instead, it was a part-time duty for normal police officers, and frequently they were encouraged to practice SWAT tactics by using them at random for some otherwise normal arrest or search. Balko argues that those raids were more exciting than normal police work, leading to a flood of volunteers for that duty and a tendency to use them as much as possible. That in turn normalizes disconnecting police tactics from the underlying crime or situational risk. So far, so good. But despite the information I was able to extract from it, I have mixed feelings about Rise of the Warrior Cop as a whole. At the least, it has substantial limitations. First, I don't trust the historical survey of policing in this book. Libertarian writing makes for bad history. The constraints of the genre require overusing only a few points of reference, treating every opinion of the US founders as holy writ, and tying forward progress to a return to a previous era, all of which interfere with good analysis. Balko also didn't do the research for the historical survey, as is clear from the footnotes. The citations are all to other people's histories, not to primary sources. He's summarizing other people's histories, and you'll almost certainly get better history by finding well-respected historians who cover the same ground. (That said, if you're not familiar with Peel's policing principles, this is a good introduction.) Second, and this too is unfortunately predictable in a libertarian treatment, race rarely appears in this book. If Balko published the same book today, I'm sure he would say more about race, but even in 2013 its absence is strange. I was struck while reading by how many examples of excessive police force were raids on west coast pot farms; yes, I'm sure that was traumatic, but it's not the demographic I would name as the most vulnerable to or affected by police brutality. West coast pot growers are, however, mostly white. I have no idea why Balko made that choice. Perhaps he thought his target audience would be more persuaded by his argument if he focused on white victims. Perhaps he thought it was an easier and less complicated story to tell. Perhaps, like a lot of libertarians, he doesn't believe racism has a significant impact on society because it would be a market failure. Perhaps those were the people who more readily came to mind. But to talk about police militarization, denial of civil rights, and police brutality in the United States without putting race at the center of both the history and the societal effects leaves a gaping hole in the analysis. Given that lack of engagement, I also am dubious of Balko's policy prescriptions. His reform suggestions aren't unreasonable, but they stay firmly in the centrist and incrementalist camp and would benefit white people more than black people. Transparency, accountability, and cultural changes are all fine and good, but the cultural change Balko is focused on is less aggressive arrest tactics, more use of mediation, and better physical fitness. I would not object to those things (well, maybe the last, which seemed odd), but we need to have a discussion about police white supremacist organizations, the prevalence of spousal abuse, and the police tendency to see themselves not as public servants but as embattled warriors who are misunderstood by the naive sheep they are defending. And, of course, you won't find in Rise of the Warrior Cop any thoughtful wrestling with whether there are alternative approaches to community safety, whether punitive rather than restorative justice is effective, or whether crime is a symptom of deeper societal problems we could address but refuse to. The most radical suggestion Balko has is to legalize drugs, which is both the predictable libertarian position and, as we have seen from recent events in the United States, far from the only problem of overcriminalization. I understand why this book is so frequently mentioned on-line, and its author's political views may make it more palatable to some people than a more race-centered or radical perspective. But I don't think this is the best or most useful book on police violence that one could read today. I hope to find a better one in upcoming reviews. Rating: 6 out of 10

23 July 2020

Sean Whitton: keyboardingupdates

Marks and mark rings in GNU Emacs I recently attempted to answer the question of whether experienced Emacs users should consider partially or fully disabling Transient Mark mode, which is (and should be) the default in modern GNU Emacs. That blog post was meant to be as information-dense as I could make it, but now I d like to describe the experience I have been having after switching to my custom pseudo-Transient Mark mode, which is labelled mitigation #2 in my older post. In summary: I feel like I ve uncovered a whole editing paradigm lying just beneath the surface of the editor I ve already been using for years. That is cool and enjoyable in itself, but I think it s also helped me understand other design decisions about the basics of the Emacs UI better than before in particular, the ideas behind how Emacs chooses where to display buffers, which were very frustrating to me in the past. I am now regularly using relatively obscure commands like C-x 4 C-o. I see it! It all makes sense now! I would encourage everyone who has never used Emacs without Transient Mark mode to try turning it off for a while, either fully or partially, just to see what you can learn. It s fascinating how it can come to seem more convenient and natural to pop the mark just to go back to the end of the current line after fixing up something earlier in the line, even though doing so requires pressing two modified keys instead of just C-e. Eshell I was amused to learn some years ago that someone was trying to make Emacs work as an X11 window manager. I was amazed and impressed to learn, more recently, that the project is still going and a fair number of people are using it. Kudos! I suspect that the basic motivation for such projects is that Emacs is a virtual Lisp machine, and it has a certain way of managing visible windows, and people would like to be able to bring both of those to their X11 window management. However, I am beginning to suspect that the intrinsic properties of Emacs buffers are tightly connected to the ways in which Emacs manages visible windows, and the intrinsic properties of Emacs buffers are at least as fundamental as its status as a virtual Lisp machine. Thus I am not convinced by the idea of trying to use Emacs ways of handling visible windows to handle windows which do not contain Emacs buffers. (but it s certainly nice to learn it s working out for others) The more general point is this. Emacs buffers are as fundamental to Emacs as anything else is, so it seems unlikely to be particularly fruitful to move something typically done outside of Emacs into Emacs, unless that activity fits naturally into an Emacs buffer or buffers. Being suited to run on a virtual Lisp machine is not enough. What could be more suited to an Emacs buffer, however, than a typical Unix command shell session? By this I mean things like running commands which produce text output, and piping this output between commands and into and out of files. Typically the commands one enters are sort of like tiny programs in themselves, even if there are no pipes involved, because you have to spend time determining just what options to pass to achieve what you want. It is great to have all your input and output available as ordinary buffer text, navigable just like all your other Emacs buffers. Full screen text user interfaces, like top(1), are not the sort of thing I have in mind here. These are suited to terminal emulators, and an Emacs buffer makes a poor terminal emulator what you end up with is a sort of terminal emulator emulator. Emacs buffers and terminal emulators are just different things. These sorts of thoughts lead one to Eshell, the Emacs Shell. Quoting from its documentation:
The shell s role is to make [system] functionality accessible to the user in an unformed state. Very roughly, it associates kernel functionality with textual commands, allowing the user to interact with the operating system via linguistic constructs. Process invocation is perhaps the most significant form this takes, using the kernel s fork' andexec functions. Emacs is a user application, but it does make the functionality of the kernel accessible through an interpreted language namely, Lisp. For that reason, there is little preventing Emacs from serving the same role as a modern shell. It too can manipulate the kernel in an unpredetermined way to cause system changes. All it s missing is the shell-ish linguistic model.
Eshell has been working very well for me for the past month or so, for, at least, Debian packaging work, which is very command shell-oriented (think tools like dch(1)). The other respects in which Eshell is tightly integrated with the rest of Emacs are icing on the cake. In particular, Eshell can transparently operate on remote hosts, using TRAMP. So when I need to execute commands on Debian s ftp-master server to process package removal requests, I just cd /ssh:fasolo: in Eshell. Emacs takes care of disconnecting and connecting to the server when needed there is no need to maintain a fragile SSH connection and a shell process (or anything else) running on the remote end. Or I can cd /ssh:athena\ sudo:root@athena: to run commands as root on the webserver hosting this blog, and, again, the text of the session survives on my laptop, and may be continued at my leisure, no matter whether athena reboots, or I shut my laptop and open it up again the next morning. And of course you can easily edit files on the remote host.

22 July 2020

Bits from Debian: Let's celebrate DebianDay 2020 around the world

We encourage our community to celebrate around the world the 27th Debian anniversary with organized DebianDay events. This year due to the COVID-19 pandemic we cannot organize in-person events, so we ask instead that contributors, developers, teams, groups, maintainers, and users promote The Debian Project and Debian activities online on August 16th (and/or 15th). Communities can organize a full schedule of online activities throughout the day. These activities can include talks, workshops, active participation with contributions such as translations assistance or editing, debates, BoFs, and all of this in your local language using tools such as Jitsi for capturing audio and video from presenters for later streaming to YouTube. If you are not aware of any local community organizing a full event or you don't want to join one, you can solo design your own activity using OBS and stream it to YouTube. You can watch an OBS tutorial here. Don't forget to record your activity as it will be a nice idea to upload it to Peertube later. Please add your event/activity on the DebianDay wiki page and let us know about and advertise it on Debian micronews. To share it, you have several options: PS: DebConf20 online is coming! It will be held from August 23rd to 29th, 2020. Registration is already open.

20 July 2020

Shirish Agarwal: Hearing loss, pandemic, lockdown

Sorry for not being on blog for sometime, the last few months have been brutal. While I am externally ok, because of the lockdown I sensed major hearing loss. First, I thought it may be a hallucination or something but as it persisted for days, I got myself checked and found out that I got 80% hearing loss in my right ear. How and why I don t know. Is this NIHL or some other kind of hearing loss is yet to be ascertained. I do live what is and used to be one of the busiest roads in the city, now for last few months not so much. On top of it, you have various other noises.

Tinnitus I also experienced Tinnitus which again I perceived to be a hallucination but found it s not. I have no clue if my eiplepsy has anything to do with hearing loss or both are different. I did discover that while today we know that something like Tinnitus exists, just 10-15 years back, people might mistake it for madness. In a way it is madness because you are constantly hearing sound, music etc. 24 7 , that is enough to drive anybody mad. During this brief period, did learn what an Otoscope is . I did get audiometry tests done but need to get at least a second or if possible also a third opinion but those will have to wait as the audio clinics are about 8-10 kms. away. In the open-close-open-close environment just makes it impossible to figure out the time, date and get it done. After that is done then probably get a hearing device, probably a Siemens Signia hearing aid. The hearing aids are damn expensive, almost 50k per piece and they probably have a lifetime of about 5-6 years, so it s a bit of a expensive proposition. I also need a second or/and third opinion on the audiometry profile so I know things are correct. All of these things are gonna take time.

Pandemic Situation in India and Testing Coincidentally, was talking to couple of people about this. It is sad to see that we have the third highest number of covid cases at 1/10th the tests we are doing vis-a-vis U.S.A. According to statistical site ourworldindata , we seem to be testing 0.22 per thousand people compared to 2.28 people per thousand done by United States. Sadly it doesn t give the breakup of the tests, from what I read the PCR tests are better than the antibody tests, a primer shares the difference between the two tests. IIRC, the antibody tests are far cheaper than the swab tests but swab tests are far more accurate as it looks for the virus s genetic material (RNA) . Anyways coming to the numbers, U.S. has a population of roughly 35 crores taking a little bit liberty from numbers given at popclock . India meanwhile has 135 crore or almost four times the population of U.S. and the amount of testing done is 1/10th as shared above. Just goes to share where the GOI priorities lie . We are running out of beds, ventilators and whatever else there is. Whatever resources are there are being used for covid patients and they are being charged a bomb. I have couple of hospitals near my place and the cost of a bed in an isolation ward is upward of INR 100k and if you need a ventilator then add another 50k . And in moment of rarity, the differences between charges of private and public are zero. Meaning there is immense profiteering happening it seems in the medical world. Heck, even the Govt. is on the act where they are charging 18% GST on sanitizers. If this is not looting then I dunno what is.
Example of Medical Bills people have to pay.

China, Nepal & Diplomacy While everybody today knows how China has intruded and captured quite a part of Ladakh, this wasn t the case when they started in April. That time Ajai Shukla had shared this with the top defence personnel but nothing came of it. Then on May 30th he broke/shared the news with the rest of the world and was immediately branded anti-national, person on Chinese payroll and what not. This is when he and Pravin Sawnhey of Force Magazine had both been warning of the same from last year itself. Pravin, has a youtube channel and had been warning India against Chinese intentions from 2015 and even before that. He had warned repeatedly that our obsession with the Pakistan border meant that we were taking eyes of the border with China which spans almost 2300 odd kms. going all the way to Arunachal Pradesh. A good map which shows the conflict can be found at dw.com which I am sharing/reproducing below
India-China Border Areas Copyright DW.com 2020
Note:- I am sharing a neutral party s rendering of the border disputes or somebody who doesn t have much at stake as the two countries have so that things could be looked at little objectively. The Prime Minister on the other hand, made the comment which made galvanising a made-up word into verb . It means to go without coming in. In fact, several news sites shared the statement told by the Prime Minister and the majority of people were shocked. In fact, there had been reports that he gave the current CDS, General Rawat, a person of his own choosing, a peace of his mind. But what lead to this confrontation in the first place ? I think many pieces are part of that puzzle, one of the pieces are surely the cutting of defense budget for the last 6 years, Even this year, if you look at the budget slashes done in the earlier part of the year when he shared how HAL had to raise loans from the market to pay salaries of its own people. Later he shared how the Govt. was planning to slash the defence budget. Interestingly, he had also shared some of the reasons which reaffirm that it is the only the Govt. which can solve some of the issues/conundrum

First, it must recognize that our firms competing for global orders are up against rivals that are being supported by their home governments with tax and export incentives and infrastructure that almost invariably surpasses India s. Our government must provide its aerospace firms with a level playing field, if not a competitive advantage. The greatest deterrent to growth our companies face is the high cost of capital and lack of access to funds. In several cases, Indian MSMEs have had to turn down offers to build components and assemblies for global OEM supply chains simply because the cost of capital to create the shop floor and train the personnel was too high. This resulted in a loss of business and a missed opportunity for creating jobs and skills. To overcome this, the government could create a sector specific A&D Fund to provide low cost capital quickly to enable our MSMEs to grab fleeting business opportunities. Ajai Shukla, blogpost 13th March 2020 . And then reporting on 11th May 2020 itself, CDS Gen. Rawat himself commented on saving the budget, they were in poor taste but still he shared what he thought about it. So, at the end of it one part of the story. The other part of the story probably lies in India s relations with its neighbors and lack of numbers in diplomats and diplomacy. So let me cover both the things one by one .

Diplomats, lack of numbers and hence the hands we are dealth with When Mr. Modi started his first term, he used the term Maximum Governance, Minimum Government but sadly cut those places where it indeed needs more people, one of which is diplomacy. A slightly dated 2012 article/opinion shared writes that India needs to engage with the rest of the world and do with higher number. Cut to 2020 and the numbers more or less remain the same . What Mr. Modi tried to do is instead of using diplomats, he tried to use his charm and hugopolicy for lack of a better term. 6 years later, here we are. After 200 trips abroad, not a single trade agreement to show what he done. I could go on but both time and energy are not on my side hence now switching to Nepal

Nepal, once friend, now enemy ? Nepal had been a friend of India for 70 odd years, what changed in the last few years that it changed from friend to enemy ? There had been two incidents in recent memory that changed the status quo. The first is the 2015 Nepal blockade . Now one could argue it either way but the truth is that Nepal understood that it is heavily dependent on India hence as any sovereign country would do in its interest it also started courting China for imports so there is some balance. The second one though is one of our own making. On December 16, 2014 RBI allowed Nepali citizens to have cash upto INR 25,000/- . Then in 2016 when demonetization was announced, they said that people could exchange only upto INR 4,500/- which was far below the limit shared above. And btw, before people start blaming just RBI for the decision, FEMA decisions are taken jointly by the finance ministry (FE) as well as ministry of external affairs (MEA) . So without them knowing the decision could not have been taken when announcing it. The result of lowering of demonetization is what made Nepal move more into Chinese hands and this has been shared by number of people in numerous articles in different websites. The wire interview with the vice-chairman of Niti Ayog is pretty interesting. The argument that Nepal show give an estimate of how much old money is there falls flat when in demonetization itself, it was thought of that around 30-40% was black money and would not be returned but by RBI s own admissions all 99.3% of the money was returned. Perhaps they should have consulted Prof. Arun Kumar of JNU who has extensively written and studied the topic before doing that fool-hardy step. It is the reason that since then, an economy which was searing at 9% has been contracting ever since, I could give a dozen articles stating that, but for the moment, just one will suffice. The slowing economy and the sharp divisions between people based on either outlook, religion or whatever also encouraged China to attack us. This year is not good for India. The only thing I hope Indians and people all over do is just maintain physical distances, masks and somehow survive till middle of next year without getting infected when probably most of the vaccine candidates have been trialed, results are in and we have a ready vaccine. I do hope that at least for once, ICMR shares data even after the vaccine is approved, whichever vaccine. Till later.

19 July 2020

Enrico Zini: More notable people

Ren Carmille - Wikipedia
history people archive.org
2020-07-20
Ren Carmille (8 January 1886 25 January 1945) was a French humanitarian, civil servant, and member of the French Resistance. During World War II, Carmille saved tens of thousands of Jews in Nazi-occupied France. In his capacity at the government's Demographics Department, Carmille sabotaged the Nazi census of France, saving tens of thousands of Jewish people from death camps.
Gino Strada - Wikipedia
health people archive.org
2020-07-20
Gino Strada (born Luigi Strada; 21 April 1948) is an Italian war surgeon and founder of Emergency, a UN-recognized international non-governmental organization.
Morbo di K - Wikipedia
history people archive.org
2020-07-20
Il morbo di K una malattia inventata nel 1943, durante la Seconda guerra mondiale, da Adriano Ossicini insieme al dottor Giovanni Borromeo per salvare alcuni italiani di religione ebraica dalle persecuzioni nazifasciste a Roma.[1][2][3][4]
Gino Bartali - Wikipedia
history people archive.org
2020-07-20
Stage races

17 July 2020

Sven Hoexter: Debian and exFAT

As some might have noticed we now have Linux 5.7 in Debian/unstable and subsequently the in kernel exFAT implementation created by Samsung available. Thus we now have two exFAT implementations, the exfat fuse driver and the Linux based one. Since some comments and mails I received showed minor confusions, especially around the available tooling, it might help to clarify a bit what is required when. Using the Samsung Linux Kernel Implementation Probably the most common use case is that you just want to use the in kernel implementation. Easy, install a Linux 5.7 kernel package for your architecture and either remove the exfat-fuse package or make sure you've version 1.3.0-2 or later installed. Then you can just run mount /dev/sdX /mnt and everything should be fine. Your result will look something like this: 
$ mount grep sdb
/dev/sdb on /mnt type exfat (rw,relatime,fmask=0022,dmask=0022,iocharset=utf8,errors=remount-ro
In the past this basic mount invocation utilized the mount.exfat helper, which was just a symlink to the helper which is shipped as /sbin/mount.exfat-fuse. The link was dropped from the package in 1.3.0-2. If you're running a not so standard setup, and would like to keep an older version of exfat-fuse installed you must invoke mount -i to prevent mount from loading any helper to mount the filesystem. See also man 8 mount. For those who care, mstone@ and myself had a brief discussion about this issue in #963752, which quickly brought me to the conclusion that it's in the best interest of a majority of users to just drop the symlink from the package. Sticking to the Fuse Driver If you would like to stick to the fuse driver you can of course just do it. I plan to continue to maintain all packages for the moment. Just keep the exfat-fuse package installed and use the mount.exfat-fuse helper directly. E.g. 
$ sudo mount.exfat-fuse /dev/sdb /mnt
FUSE exfat 1.3.0
$ mount grep sdb
/dev/sdb on /mnt type fuseblk (rw,nosuid,nodev,relatime,user_id=0,group_id=0,default_permissions,allow_other,blksize=4096)
In case this is something you would like to make permanent, I would recommend that you create yourself a mount.exfat symlink pointing at the mount.exfat-fuse helper. mkfs and fsck - exfat-utils vs exfatprogs Beside of the filesystem access itself we now also have two implementations of tooling, which support filesystem creation, fscking and label adjustment. The older one is exfat-utils by the creator of the fuse driver, which is also part of Debian since the fuse driver was packaged in 2012. New in Debian is the exfatprogs package written by the Samsung engineers. And here the situation starts to get a bit more complicated. Both packages ship a mkfs.exfat and fsck.exfat tool, so we can not co-install them. In the end both packages declare a conflict with each other at the moment. As outlined in this thread I do not plan to overcomplicate the situation by using the alternatives system. I do feel strongly that this would just create more confusion without a real benefit. Since the tools do not have matching cli options, that could also cause additional issues and confusion. I plan to keep that as is, at least for the bullseye release. Afterwards it's possible, depending on how the usage evolves, to drop the mkfs.exfat and fsck.exfat from exfat-utils, they are in fact again only symlinks. pain point might be tools interfacing with the differing implementations. Currently I see only three reverse depedencies, so that should be manageable to consolidate if required. Last but not least it might be relevant to mention that the exfat-utils package also contains a dumpexfat tool which could be helpful if you're more into forensics, or looking into other lower level analysis of an exFAT filsystem. Thus there is a bit of an interest to have those tools co-installed in some - I would say - niche cases. buster-backports Well if you use buster with a backports kernel you're a bit on your own. In case you want to keep the fuse driver installed, but would still like to mount, e.g. for testing, with the kernel exFAT driver, you must use mount -i. I do not plan any uploads to buster-backports. If you need a mkfs.exfat on buster, I would recommend to just use the one from exfat-utils for now. It has been good enough for the past years, should not get sour before the bullseye release, which ships exfatprogs for you. Kudos My sincere kudos go to:

16 July 2020

Dima Kogan: Visualizing a broken internet connection

We've all seen our ISPs crap out periodically. Things feel slow, we then go ping some server, and see lost packets or slow response times. When bitching to the ISP it's useful to have evidence so you can clearly explain exactly how they are fucking up. This happened to me again last week, so I wrote a quick oneliner to do the visualization. Here it is: 
( echo '# i dt'; ping 8.8.8.8   perl -ne 'BEGIN   $ =1  next if /PING/; s/.*seq=([0-9]+).*time=([0-9]+).*/\1 \2/ && print')   tee /tmp/ping.vnl   vnl-filter --stream -p i,dt,di='diff(i)'   vnl-filter --stream --has di   feedgnuplot  --domain --stream --with 'linespoints pt 7 palette' --tuplesizeall 3 --ymin 0 --set 'cbrange [1:]' --xlabel "ping index" --ylabel "Response time (ms)" --title "ping 8.8.8.8 response times. Consecutive indices shown as colors"
You run that, and you get a realtime-updating plot of ping times and missed packets. The one-liner also logs the data to a file, so the saved data can be re-visualized. Showing what happens on my currently-working internet connection: 
< /tmp/ping.vnl vnl-filter -p i,dt,di='diff(i)'   vnl-filter --has di   feedgnuplot  --domain --with 'linespoints pt 7 palette' --tuplesizeall 3 --ymin 0 --set 'cbrange [1:]' --xlabel "ping index" --ylabel "Response time (ms)" --title "ping 8.8.8.8 response times. Consecutive indices shown as colors" --hardcopy /tmp/isp-good.svg
good.svg
On a broken network it looks like this (I edited the log to show the kind of broken-ness I was seeing earlier):
bad.svg
The x axis is the ping index. The y axis is the response time. Misses responses are shows as gaps in the points and also as colors, for legibility. As usual, this uses the vnlog and feedgnuplot tools, so 
sudo apt install feedgnuplot vnlog
if you want to try this

Enrico Zini: Build Qt5 cross-builder with raspbian sysroot: compiling with the sysroot

Whack-A-Mole machines from <https://commons.wikimedia.org/wiki/File:Whac-A-Mole_Cedar_Point.jpg> This is part of a series of posts on compiling a custom version of Qt5 in order to develop for both amd64 and a Raspberry Pi. Now that I have a sysroot, I try to use it to build Qt5 with QtWebEngine. Nothing seems to work straightforwardly with Qt5's build system, and hit an endless series of significant blockers to try and work around.
Problem in wayland code QtWayland's source currently does not compile: 
../../../hardwareintegration/client/brcm-egl/qwaylandbrcmeglwindow.cpp: In constructor  QtWaylandClient::QWaylandBrcmEglWindow::QWaylandBrcmEglWindow(QWindow*) :
../../../hardwareintegration/client/brcm-egl/qwaylandbrcmeglwindow.cpp:131:67: error: no matching function for call to  QtWaylandClient::QWaylandWindow::QWaylandWindow(QWindow*&) 
     , m_eventQueue(wl_display_create_queue(mDisplay->wl_display()))
                                                                   ^
In file included from ../../../../include/QtWaylandClient/5.15.0/QtWaylandClient/private/qwaylandwindow_p.h:1,
                 from ../../../hardwareintegration/client/brcm-egl/qwaylandbrcmeglwindow.h:43,
                 from ../../../hardwareintegration/client/brcm-egl/qwaylandbrcmeglwindow.cpp:40:
../../../../include/QtWaylandClient/5.15.0/QtWaylandClient/private/../../../../../src/client/qwaylandwindow_p.h:97:5: note: candidate:  QtWaylandClient::QWaylandWindow::QWaylandWindow(QWindow*, QtWayland
Client::QWaylandDisplay*) 
     QWaylandWindow(QWindow *window, QWaylandDisplay *display);
     ^~~~~~~~~~~~~~
../../../../include/QtWaylandClient/5.15.0/QtWaylandClient/private/../../../../../src/client/qwaylandwindow_p.h:97:5: note:   candidate expects 2 arguments, 1 provided
make[5]: Leaving directory '/home/build/armhf/qt-everywhere-src-5.15.0/qttools/src/qdoc'
I am not trying to debug here. I understand that Wayland support is not a requirement, and I'm adding -skip wayland to Qt5's configure options. Next round. nss not found Qt5 embeds Chrome's sources. Chrome's sources require libnss3-dev to be available for both host and target architectures. Although I now have it installed both on the build system and in the sysroot, the pkg-config wrapper that Qt5 hooks into its Chrome's sources, failes to find it: 
Command: /usr/bin/python2 /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/3rdparty/chromium/build/config/linux/pkg-config.py -s /home/build/sysroot/ -a arm -p /usr/bin/arm-linux-gnueabihf-pkg-config --system_libdir lib nss -v -lssl3
Returned 1.
stderr:
Package nss was not found in the pkg-config search path.
Perhaps you should add the directory containing  nss.pc'
to the PKG_CONFIG_PATH environment variable
No package 'nss' found
Traceback (most recent call last):
  File "/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/3rdparty/chromium/build/config/linux/pkg-config.py", line 248, in <module>
    sys.exit(main())
  File "/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/3rdparty/chromium/build/config/linux/pkg-config.py", line 143, in main
    prefix = GetPkgConfigPrefixToStrip(options, args)
  File "/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/3rdparty/chromium/build/config/linux/pkg-config.py", line 82, in GetPkgConfigPrefixToStrip
    "--variable=prefix"] + args, env=os.environ).decode('utf-8')
  File "/usr/lib/python2.7/subprocess.py", line 223, in check_output
    raise CalledProcessError(retcode, cmd, output=output)
subprocess.CalledProcessError: Command '['/usr/bin/arm-linux-gnueabihf-pkg-config', '--variable=prefix', 'nss']' returned non-zero exit status 1
See //build/config/linux/nss/BUILD.gn:15:3: whence it was called.
  pkg_config("system_nss_no_ssl_config")  
  ^---------------------------------------
See //crypto/BUILD.gn:218:25: which caused the file to be included.
    public_configs += [ "//build/config/linux/nss:system_nss_no_ssl_config" ]
                        ^--------------------------------------------------
Project ERROR: GN run error!
It's trying to look into $SYSROOT/usr/lib/pkgconfig, while it should be $SYSROOT//usr/lib/arm-linux-gnueabihf/pkgconfig. I worked around this this patch to qtwebengine/src/3rdparty/chromium/build/config/linux/pkg-config.py: 
--- pkg-config.py.orig  2020-07-16 11:46:21.005373002 +0200
+++ pkg-config.py   2020-07-16 11:46:02.605296967 +0200
@@ -61,6 +61,7 @@
   libdir = sysroot + '/usr/' + options.system_libdir + '/pkgconfig'
   libdir += ':' + sysroot + '/usr/share/pkgconfig'
+  libdir += ':' + sysroot + '/usr/lib/arm-linux-gnueabihf/pkgconfig'
   os.environ['PKG_CONFIG_LIBDIR'] = libdir
   return libdir
Next round. g++ 8.3.0 Internal Compiler Error Qt5's sources embed Chrome's sources that embed the skia library sources. One of the skia library sources, when cross-compiled to ARM with -O1 or -O2 with g++ 8.3.0, produces an Internal Compiler Error: 
/usr/bin/arm-linux-gnueabihf-g++ -MMD -MF obj/skia/skcms/skcms.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -DTOOLKIT_QT -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DCR_SYSROOT_HASH=76e6068f9f6954e2ab1ff98ce5fa236d3d85bcbd -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -I../../3rdparty/chromium/third_party/skia/include/third_party/skcms -Igen -I../../3rdparty/chromium -w -std=c11 -mfp16-format=ieee -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -march=armv7-a -mfloat-abi=hard -mtune=generic-armv7-a -mfpu=vfpv3-d16 -mthumb -Wall -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -Wno-psabi -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -fno-delete-null-pointer-checks -Wno-comments -Wno-packed-not-aligned -Wno-dangling-else -Wno-missing-field-initializers -Wno-unused-parameter -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fvisibility=hidden -std=gnu++14 -Wno-narrowing -Wno-class-memaccess -Wno-attributes -Wno-class-memaccess -Wno-subobject-linkage -Wno-invalid-offsetof -Wno-return-type -Wno-deprecated-copy -fno-exceptions -fno-rtti --sysroot=../../../../../../sysroot/ -fvisibility-inlines-hidden -c ../../3rdparty/chromium/third_party/skia/third_party/skcms/skcms.cc -o obj/skia/skcms/skcms.o
during RTL pass: expand
In file included from ../../3rdparty/chromium/third_party/skia/third_party/skcms/skcms.cc:2053:
../../3rdparty/chromium/third_party/skia/third_party/skcms/src/Transform_inl.h: In function  void baseline::exec_ops(const Op*, const void**, const char*, char*, int) :
../../3rdparty/chromium/third_party/skia/third_party/skcms/src/Transform_inl.h:766:13: internal compiler error: in convert_move, at expr.c:218
 static void exec_ops(const Op* ops, const void** args,
             ^~~~~~~~
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-8/README.Bugs> for instructions.
I reported the bug at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96206 Since this source compiles with -O0, I attempted to fix this by editing qtwebkit/src/3rdparty/chromium/build/config/compiler/BUILD.gn and replacing instances of -O1 and -O2 with -O0. Spoiler: wrong attempt. We'll see it in the next round. Impossible constraint in asm Qt5's sources embed Chrome's sources that embed the ffmpeg library sources. Even if ffmpeg's development libraries are present both in the host and in the target system, the build system insists in compiling and using the bundled version. Unfortunately, using -O0 breaks the build of ffmpeg: 
/usr/bin/arm-linux-gnueabihf-gcc -MMD -MF obj/third_party/ffmpeg/ffmpeg_internal/opus.o.d -DHAVE_AV_CONFIG_H -D_POSIX_C_SOURCE=200112 -D_XOPEN_SOURCE=600 -DPIC -DFFMPEG_CONFIGURATION=NULL -DCHROMIUM_NO_LOGGING -D_ISOC99_SOURCE -D_LARGEFILE_SOURCE -DUSE_UDEV -DUSE_AURA=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -DTOOLKIT_QT -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -DCR_SYSROOT_HASH=76e6068f9f6954e2ab1ff98ce5fa236d3d85bcbd -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DOPUS_FIXED_POINT -I../../3rdparty/chromium/third_party/ffmpeg/chromium/config/Chromium/linux/arm -I../../3rdparty/chromium/third_party/ffmpeg -I../../3rdparty/chromium/third_party/ffmpeg/compat/atomics/gcc -Igen -I../../3rdparty/chromium -I../../3rdparty/chromium/third_party/opus/src/include -fPIC -Wno-deprecated-declarations -fomit-frame-pointer -w -std=c99 -pthread -fno-math-errno -fno-signed-zeros -fno-tree-vectorize -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -march=armv7-a -mfloat-abi=hard -mtune=generic-armv7-a -mfpu=vfpv3-d16 -mthumb -g0 -fvisibility=hidden -Wno-psabi -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -fno-delete-null-pointer-checks -Wno-comments -Wno-packed-not-aligned -Wno-dangling-else -Wno-missing-field-initializers -Wno-unused-parameter -O0 -fno-ident -fdata-sections -ffunction-sections -std=gnu11 --sysroot=../../../../../../sysroot/ -c ../../3rdparty/chromium/third_party/ffmpeg/libavcodec/opus.c -o obj/third_party/ffmpeg/ffmpeg_internal/opus.o
In file included from ../../3rdparty/chromium/third_party/ffmpeg/libavutil/intmath.h:30,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavutil/common.h:106,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavutil/avutil.h:296,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavutil/audio_fifo.h:30,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavcodec/opus.h:28,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavcodec/opus_celt.h:29,
                 from ../../3rdparty/chromium/third_party/ffmpeg/libavcodec/opus.c:32:
../../3rdparty/chromium/third_party/ffmpeg/libavcodec/opus.c: In function  ff_celt_quant_bands :
../../3rdparty/chromium/third_party/ffmpeg/libavutil/arm/intmath.h:77:5: error: impossible constraint in  asm 
     __asm__ ("usat %0, %2, %1" : "=r"(x) : "r"(a), "i"(p));
     ^~~~~~~
The same source compiles with using -O2 instead of -O0. I worked around this by undoing the previous change, and limiting -O0 to just the source that causes the Internal Compiler Error. I edited qtwebengine/src/3rdparty/chromium/third_party/skia/third_party/skcms/skcms.cc to prepend: 
#pragma GCC push_options
#pragma GCC optimize ("O0")
and append: 
#pragma GCC pop_options
Next round. Missing build-deps for i386 code Qt5's sources embed Chrome's sources that embed the V8 library sources. For some reason, torque, that is part of V8, wants to build some of its sources into 32 bit code with -m32, and I did not have i386 cross-compilation libraries installed: 
/usr/bin/g++ -MMD -MF v8_snapshot/obj/v8/torque_base/csa-generator.o.d -DUSE_UDEV -DUSE_AURA=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -DTOOLKIT_QT -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DV8_TYPED_ARRAY_MAX_SIZE_IN_HEAP=64 -DENABLE_MINOR_MC -DV8_INTL_SUPPORT -DV8_CONCURRENT_MARKING -DV8_ENABLE_LAZY_SOURCE_POSITIONS -DV8_EMBEDDED_BUILTINS -DV8_SHARED_RO_HEAP -DV8_WIN64_UNWINDING_INFO -DV8_ENABLE_REGEXP_INTERPRETER_THREADED_DISPATCH -DV8_31BIT_SMIS_ON_64BIT_ARCH -DV8_DEPRECATION_WARNINGS -DV8_TARGET_ARCH_ARM -DCAN_USE_ARMV7_INSTRUCTIONS -DCAN_USE_VFP3_INSTRUCTIONS -DUSE_EABI_HARDFLOAT=1 -DV8_HAVE_TARGET_OS -DV8_TARGET_OS_LINUX -DDISABLE_UNTRUSTED_CODE_MITIGATIONS -DV8_31BIT_SMIS_ON_64BIT_ARCH -DV8_DEPRECATION_WARNINGS -Iv8_snapshot/gen -I../../3rdparty/chromium -I../../3rdparty/chromium/v8 -Iv8_snapshot/gen/v8 -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -m32 -msse2 -mfpmath=sse -mmmx -Wall -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -fno-delete-null-pointer-checks -Wno-comments -Wno-packed-not-aligned -Wno-dangling-else -Wno-missing-field-initializers -Wno-unused-parameter -fno-omit-frame-pointer -g0 -fvisibility=hidden -Wno-strict-overflow -Wno-return-type -O3 -fno-ident -fdata-sections -ffunction-sections -std=gnu++14 -Wno-narrowing -Wno-class-memaccess -Wno-attributes -Wno-class-memaccess -Wno-subobject-linkage -Wno-invalid-offsetof -Wno-return-type -Wno-deprecated-copy -fvisibility-inlines-hidden -fexceptions -frtti -c ../../3rdparty/chromium/v8/src/torque/csa-generator.cc -o v8_snapshot/obj/v8/torque_base/csa-generator.o
In file included from ../../3rdparty/chromium/v8/src/torque/csa-generator.h:8,
                 from ../../3rdparty/chromium/v8/src/torque/csa-generator.cc:5:
/usr/include/c++/8/iostream:38:10: fatal error: bits/c++config.h: No such file or directory
 #include <bits/c++config.h>
          ^~~~~~~~~~~~~~~~~~
compilation terminated.
New build dependencies needed: 
apt install lib32stdc++-8-dev
apt install libc6-dev-i386
dpkg --add-architecture i386
apt install linux-libc-dev:i386
Next round. OpenGL build issues Next bump are OpenGL related compiler issues: 
/usr/bin/arm-linux-gnueabihf-g++ -MMD -MF obj/QtWebEngineCore/gl_ozone_glx_qt.o.d -DCHROMIUM_VERSION=\"80.0.3987.163\" -DUSE_UDEV -DUSE_AURA=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -DTOOLKIT_QT -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DCR_SYSROOT_HASH=76e6068f9f6954e2ab1ff98ce5fa236d3d85bcbd -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DQT_NO_LINKED_LIST -DQT_NO_KEYWORDS -DQT_USE_QSTRINGBUILDER -DQ_FORWARD_DECLARE_OBJC_CLASS=QT_FORWARD_DECLARE_CLASS -DQTWEBENGINECORE_VERSION_STR=\"5.15.0\" -DQTWEBENGINEPROCESS_NAME=\"QtWebEngineProcess\" -DBUILDING_CHROMIUM -DQTWEBENGINE_EMBEDDED_SWITCHES -DQT_NO_EXCEPTIONS -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DQT_QUICK_LIB -DQT_GUI_LIB -DQT_QMLMODELS_LIB -DQT_WEBCHANNEL_LIB -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_POSITIONING_LIB -DQT_CORE_LIB -DQT_WEBENGINECOREHEADERS_LIB -DVK_NO_PROTOTYPES -DGL_GLEXT_PROTOTYPES -DUSE_GLX -DUSE_EGL -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -DU_USING_ICU_NAMESPACE=0 -DU_ENABLE_DYLOAD=0 -DUSE_CHROMIUM_ICU=1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL=ICU_UTIL_DATA_FILE -DUCHAR_TYPE=uint16_t -DWEBRTC_NON_STATIC_TRACE_EVENT_HANDLERS=0 -DWEBRTC_CHROMIUM_BUILD -DWEBRTC_POSIX -DWEBRTC_LINUX -DABSL_ALLOCATOR_NOTHROW=1 -DWEBRTC_USE_BUILTIN_ISAC_FIX=1 -DWEBRTC_USE_BUILTIN_ISAC_FLOAT=0 -DHAVE_SCTP -DNO_MAIN_THREAD_WRAPPING -DSK_HAS_PNG_LIBRARY -DSK_HAS_WEBP_LIBRARY -DSK_USER_CONFIG_HEADER=\"../../skia/config/SkUserConfig.h\" -DSK_GL -DSK_HAS_JPEG_LIBRARY -DSK_USE_LIBGIFCODEC -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSK_SUPPORT_GPU=1 -DSK_GPU_WORKAROUNDS_HEADER=\"gpu/config/gpu_driver_bug_workaround_autogen.h\" -DVK_NO_PROTOTYPES -DLEVELDB_PLATFORM_CHROMIUM=1 -DLEVELDB_PLATFORM_CHROMIUM=1 -DV8_31BIT_SMIS_ON_64BIT_ARCH -DV8_DEPRECATION_WARNINGS -I../../3rdparty/chromium/skia/config -I../../3rdparty/chromium/third_party -I../../3rdparty/chromium/third_party/boringssl/src/include -I../../3rdparty/chromium/third_party/skia/include/core -Igen -I../../3rdparty/chromium -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/api -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick/5.15.0/QtQuick -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui/5.15.0/QtGui -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels/5.15.0/QtQmlModels -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml/5.15.0/QtQml -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore/5.15.0/QtCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebchannel/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebchannel/include/QtWebChannel -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtNetwork -I/home/build/armhf/qt-everywhere-src-5.15.0/qtlocation/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtlocation/include/QtPositioning -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore/5.15.0/QtWebEngineCore -I.moc -I/home/build/sysroot/opt/vc/include -I/home/build/sysroot/opt/vc/include/interface/vcos/pthreads -I/home/build/sysroot/opt/vc/include/interface/vmcs_host/linux -Igen/.moc -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/mkspecs/devices/linux-rasp-pi2-g++ -Igen -Igen -I../../3rdparty/chromium/third_party/libyuv/include -Igen -I../../3rdparty/chromium/third_party/jsoncpp/source/include -I../../3rdparty/chromium/third_party/jsoncpp/generated -Igen -Igen -I../../3rdparty/chromium/third_party/khronos -I../../3rdparty/chromium/gpu -I../../3rdparty/chromium/third_party/vulkan/include -I../../3rdparty/chromium/third_party/perfetto/include -Igen/third_party/perfetto/build_config -Igen -Igen -Igen/third_party/dawn/src/include -I../../3rdparty/chromium/third_party/dawn/src/include -Igen -I../../3rdparty/chromium/third_party/boringssl/src/include -I../../3rdparty/chromium/third_party/protobuf/src -Igen/protoc_out -I../../3rdparty/chromium/third_party/protobuf/src -I../../3rdparty/chromium/third_party/ced/src -I../../3rdparty/chromium/third_party/icu/source/common -I../../3rdparty/chromium/third_party/icu/source/i18n -I../../3rdparty/chromium/third_party/webrtc_overrides -I../../3rdparty/chromium/third_party/webrtc -Igen/third_party/webrtc -I../../3rdparty/chromium/third_party/abseil-cpp -I../../3rdparty/chromium/third_party/skia -I../../3rdparty/chromium/third_party/libgifcodec -I../../3rdparty/chromium/third_party/vulkan/include -I../../3rdparty/chromium/third_party/skia/third_party/vulkanmemoryallocator -I../../3rdparty/chromium/third_party/vulkan/include -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -I../../3rdparty/chromium/third_party/crashpad/crashpad -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/non_mac -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/linux -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/non_win -I../../3rdparty/chromium/third_party/libwebm/source -I../../3rdparty/chromium/third_party/leveldatabase -I../../3rdparty/chromium/third_party/leveldatabase/src -I../../3rdparty/chromium/third_party/leveldatabase/src/include -I../../3rdparty/chromium/v8/include -Igen/v8/include -I../../3rdparty/chromium/third_party/mesa_headers -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -march=armv7-a -mfloat-abi=hard -mtune=generic-armv7-a -mfpu=vfpv3-d16 -mthumb -Wall -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -Wno-psabi -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -fno-delete-null-pointer-checks -Wno-comments -Wno-packed-not-aligned -Wno-dangling-else -Wno-missing-field-initializers -Wno-unused-parameter -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fvisibility=hidden -g -O2 -fdebug-prefix-map=/home/build/armhf/qt-everywhere-src-5.15.0=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -O2 -fno-exceptions -Wall -Wextra -D_REENTRANT -I/home/build/sysroot/usr/include/nss -I/home/build/sysroot/usr/include/nspr -std=gnu++14 -Wno-narrowing -Wno-class-memaccess -Wno-attributes -Wno-class-memaccess -Wno-subobject-linkage -Wno-invalid-offsetof -Wno-return-type -Wno-deprecated-copy -fno-exceptions -fno-rtti --sysroot=../../../../../../sysroot/ -fvisibility-inlines-hidden -g -O2 -fdebug-prefix-map=/home/build/armhf/qt-everywhere-src-5.15.0=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -O2 -std=gnu++1y -fno-exceptions -Wall -Wextra -D_REENTRANT -Wno-unused-parameter -Wno-unused-variable -Wno-deprecated-declarations -c /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/ozone/gl_ozone_glx_qt.cpp -o obj/QtWebEngineCore/gl_ozone_glx_qt.o
In file included from ../../3rdparty/chromium/ui/gl/gl_bindings.h:497,
                 from ../../3rdparty/chromium/ui/gl/gl_gl_api_implementation.h:12,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/ozone/gl_ozone_glx_qt.cpp:49:
../../3rdparty/chromium/ui/gl/gl_bindings_autogen_egl.h:227:5: error:  EGLSetBlobFuncANDROID  has not been declared
     EGLSetBlobFuncANDROID set,
     ^~~~~~~~~~~~~~~~~~~~~
../../3rdparty/chromium/ui/gl/gl_bindings_autogen_egl.h:228:5: error:  EGLGetBlobFuncANDROID  has not been declared
     EGLGetBlobFuncANDROID get);
     ^~~~~~~~~~~~~~~~~~~~~
../../3rdparty/chromium/ui/gl/gl_bindings_autogen_egl.h:571:46: error:  EGLSetBlobFuncANDROID  has not been declared
                                              EGLSetBlobFuncANDROID set,
                                              ^~~~~~~~~~~~~~~~~~~~~
../../3rdparty/chromium/ui/gl/gl_bindings_autogen_egl.h:572:46: error:  EGLGetBlobFuncANDROID  has not been declared
                                              EGLGetBlobFuncANDROID get) = 0;
                                              ^~~~~~~~~~~~~~~~~~~~~
cc1plus: warning: unrecognized command line option  -Wno-deprecated-copy 

/usr/bin/arm-linux-gnueabihf-g++ -MMD -MF obj/QtWebEngineCore/display_gl_output_surface.o.d -DCHROMIUM_VERSION=\"80.0.3987.163\" -DUSE_UDEV -DUSE_AURA=1 -DUSE_NSS_CERTS=1 -DUSE_OZONE=1 -DOFFICIAL_BUILD -DTOOLKIT_QT -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DNO_UNWIND_TABLES -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -DCR_SYSROOT_HASH=76e6068f9f6954e2ab1ff98ce5fa236d3d85bcbd -DNDEBUG -DNVALGRIND -DDYNAMIC_ANNOTATIONS_ENABLED=0 -DQT_NO_LINKED_LIST -DQT_NO_KEYWORDS -DQT_USE_QSTRINGBUILDER -DQ_FORWARD_DECLARE_OBJC_CLASS=QT_FORWARD_DECLARE_CLASS -DQTWEBENGINECORE_VERSION_STR=\"5.15.0\" -DQTWEBENGINEPROCESS_NAME=\"QtWebEngineProcess\" -DBUILDING_CHROMIUM -DQTWEBENGINE_EMBEDDED_SWITCHES -DQT_NO_EXCEPTIONS -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DQT_QUICK_LIB -DQT_GUI_LIB -DQT_QMLMODELS_LIB -DQT_WEBCHANNEL_LIB -DQT_QML_LIB -DQT_NETWORK_LIB -DQT_POSITIONING_LIB -DQT_CORE_LIB -DQT_WEBENGINECOREHEADERS_LIB -DVK_NO_PROTOTYPES -DGL_GLEXT_PROTOTYPES -DUSE_GLX -DUSE_EGL -DGOOGLE_PROTOBUF_NO_RTTI -DGOOGLE_PROTOBUF_NO_STATIC_INITIALIZER -DHAVE_PTHREAD -DU_USING_ICU_NAMESPACE=0 -DU_ENABLE_DYLOAD=0 -DUSE_CHROMIUM_ICU=1 -DU_STATIC_IMPLEMENTATION -DICU_UTIL_DATA_IMPL=ICU_UTIL_DATA_FILE -DUCHAR_TYPE=uint16_t -DWEBRTC_NON_STATIC_TRACE_EVENT_HANDLERS=0 -DWEBRTC_CHROMIUM_BUILD -DWEBRTC_POSIX -DWEBRTC_LINUX -DABSL_ALLOCATOR_NOTHROW=1 -DWEBRTC_USE_BUILTIN_ISAC_FIX=1 -DWEBRTC_USE_BUILTIN_ISAC_FLOAT=0 -DHAVE_SCTP -DNO_MAIN_THREAD_WRAPPING -DSK_HAS_PNG_LIBRARY -DSK_HAS_WEBP_LIBRARY -DSK_USER_CONFIG_HEADER=\"../../skia/config/SkUserConfig.h\" -DSK_GL -DSK_HAS_JPEG_LIBRARY -DSK_USE_LIBGIFCODEC -DSK_VULKAN_HEADER=\"../../skia/config/SkVulkanConfig.h\" -DSK_VULKAN=1 -DSK_SUPPORT_GPU=1 -DSK_GPU_WORKAROUNDS_HEADER=\"gpu/config/gpu_driver_bug_workaround_autogen.h\" -DVK_NO_PROTOTYPES -DLEVELDB_PLATFORM_CHROMIUM=1 -DLEVELDB_PLATFORM_CHROMIUM=1 -DV8_31BIT_SMIS_ON_64BIT_ARCH -DV8_DEPRECATION_WARNINGS -I../../3rdparty/chromium/skia/config -I../../3rdparty/chromium/third_party -I../../3rdparty/chromium/third_party/boringssl/src/include -I../../3rdparty/chromium/third_party/skia/include/core -Igen -I../../3rdparty/chromium -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/api -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick/5.15.0/QtQuick -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui/5.15.0/QtGui -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQuick -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtGui -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels/5.15.0/QtQmlModels -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml/5.15.0/QtQml -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore/5.15.0/QtCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQmlModels -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebchannel/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebchannel/include/QtWebChannel -I/home/build/armhf/qt-everywhere-src-5.15.0/qtdeclarative/include/QtQml -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtNetwork -I/home/build/armhf/qt-everywhere-src-5.15.0/qtlocation/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtlocation/include/QtPositioning -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/include/QtCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore/5.15.0 -I/home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/include/QtWebEngineCore/5.15.0/QtWebEngineCore -I.moc -I/home/build/sysroot/opt/vc/include -I/home/build/sysroot/opt/vc/include/interface/vcos/pthreads -I/home/build/sysroot/opt/vc/include/interface/vmcs_host/linux -Igen/.moc -I/home/build/armhf/qt-everywhere-src-5.15.0/qtbase/mkspecs/devices/linux-rasp-pi2-g++ -Igen -Igen -I../../3rdparty/chromium/third_party/libyuv/include -Igen -I../../3rdparty/chromium/third_party/jsoncpp/source/include -I../../3rdparty/chromium/third_party/jsoncpp/generated -Igen -Igen -I../../3rdparty/chromium/third_party/khronos -I../../3rdparty/chromium/gpu -I../../3rdparty/chromium/third_party/vulkan/include -I../../3rdparty/chromium/third_party/perfetto/include -Igen/third_party/perfetto/build_config -Igen -Igen -Igen/third_party/dawn/src/include -I../../3rdparty/chromium/third_party/dawn/src/include -Igen -I../../3rdparty/chromium/third_party/boringssl/src/include -I../../3rdparty/chromium/third_party/protobuf/src -Igen/protoc_out -I../../3rdparty/chromium/third_party/protobuf/src -I../../3rdparty/chromium/third_party/ced/src -I../../3rdparty/chromium/third_party/icu/source/common -I../../3rdparty/chromium/third_party/icu/source/i18n -I../../3rdparty/chromium/third_party/webrtc_overrides -I../../3rdparty/chromium/third_party/webrtc -Igen/third_party/webrtc -I../../3rdparty/chromium/third_party/abseil-cpp -I../../3rdparty/chromium/third_party/skia -I../../3rdparty/chromium/third_party/libgifcodec -I../../3rdparty/chromium/third_party/vulkan/include -I../../3rdparty/chromium/third_party/skia/third_party/vulkanmemoryallocator -I../../3rdparty/chromium/third_party/vulkan/include -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -Igen/third_party/perfetto -I../../3rdparty/chromium/third_party/crashpad/crashpad -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/non_mac -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/linux -I../../3rdparty/chromium/third_party/crashpad/crashpad/compat/non_win -I../../3rdparty/chromium/third_party/libwebm/source -I../../3rdparty/chromium/third_party/leveldatabase -I../../3rdparty/chromium/third_party/leveldatabase/src -I../../3rdparty/chromium/third_party/leveldatabase/src/include -I../../3rdparty/chromium/v8/include -Igen/v8/include -I../../3rdparty/chromium/third_party/mesa_headers -fno-strict-aliasing --param=ssp-buffer-size=4 -fstack-protector -fno-unwind-tables -fno-asynchronous-unwind-tables -fPIC -pipe -pthread -march=armv7-a -mfloat-abi=hard -mtune=generic-armv7-a -mfpu=vfpv3-d16 -mthumb -Wall -U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2 -Wno-psabi -Wno-unused-local-typedefs -Wno-maybe-uninitialized -Wno-deprecated-declarations -fno-delete-null-pointer-checks -Wno-comments -Wno-packed-not-aligned -Wno-dangling-else -Wno-missing-field-initializers -Wno-unused-parameter -O2 -fno-ident -fdata-sections -ffunction-sections -fno-omit-frame-pointer -g0 -fvisibility=hidden -g -O2 -fdebug-prefix-map=/home/build/armhf/qt-everywhere-src-5.15.0=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -O2 -fno-exceptions -Wall -Wextra -D_REENTRANT -I/home/build/sysroot/usr/include/nss -I/home/build/sysroot/usr/include/nspr -std=gnu++14 -Wno-narrowing -Wno-class-memaccess -Wno-attributes -Wno-class-memaccess -Wno-subobject-linkage -Wno-invalid-offsetof -Wno-return-type -Wno-deprecated-copy -fno-exceptions -fno-rtti --sysroot=../../../../../../sysroot/ -fvisibility-inlines-hidden -g -O2 -fdebug-prefix-map=/home/build/armhf/qt-everywhere-src-5.15.0=. -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -O2 -std=gnu++1y -fno-exceptions -Wall -Wextra -D_REENTRANT -Wno-unused-parameter -Wno-unused-variable -Wno-deprecated-declarations -c /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp -o obj/QtWebEngineCore/display_gl_output_surface.o
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_interface.h:8,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/client_transfer_cache.h:15,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:28,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
/home/build/sysroot/opt/vc/include/GLES2/gl2.h:78: warning: "GL_FALSE" redefined
 #define GL_FALSE                          (GLboolean)0
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/client_context_state.h:10,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:27,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
../../3rdparty/chromium/third_party/khronos/GLES3/gl3.h:85: note: this is the location of the previous definition
 #define GL_FALSE                          0
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_interface.h:8,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/client_transfer_cache.h:15,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:28,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
/home/build/sysroot/opt/vc/include/GLES2/gl2.h:79: warning: "GL_TRUE" redefined
 #define GL_TRUE                           (GLboolean)1
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/client_context_state.h:10,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:27,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
../../3rdparty/chromium/third_party/khronos/GLES3/gl3.h:86: note: this is the location of the previous definition
 #define GL_TRUE                           1
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_interface.h:8,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/client_transfer_cache.h:15,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:28,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
/home/build/sysroot/opt/vc/include/GLES2/gl2.h:600:37: error: conflicting declaration of C function  void glShaderSource(GLuint, GLsizei, const GLchar**, const GLint*) 
 GL_APICALL void         GL_APIENTRY glShaderSource (GLuint shader, GLsizei count, const GLchar** string, const GLint* length);
                                     ^~~~~~~~~~~~~~
In file included from ../../3rdparty/chromium/gpu/command_buffer/client/client_context_state.h:10,
                 from ../../3rdparty/chromium/gpu/command_buffer/client/gles2_implementation.h:27,
                 from /home/build/armhf/qt-everywhere-src-5.15.0/qtwebengine/src/core/compositor/display_gl_output_surface.cpp:47:
../../3rdparty/chromium/third_party/khronos/GLES3/gl3.h:624:29: note: previous declaration  void glShaderSource(GLuint, GLsizei, const GLchar* const*, const GLint*) 
 GL_APICALL void GL_APIENTRY glShaderSource (GLuint shader, GLsizei count, const GLchar *const*string, const GLint *length);
                             ^~~~~~~~~~~~~~
cc1plus: warning: unrecognized command line option  -Wno-deprecated-copy
I'm out of the allocated hour budget, and I'll stop here for now. Building Qt5 has been providing some of the most nightmarish work time in my entire professional life. If my daily job became being required to deal with this kind of insanity, I would strongly invest in a change of career. Update Andreas Gruber wrote:
Long story short, a fast solution for the issue with EGLSetBlobFuncANDROID is to remove libraspberrypi-dev from your sysroot and do a full rebuild. There will be some changes to the configure results, so please review them - if they are relevant for you - before proceeding with your work.
And thanks to Andreas, the story can continue...

14 July 2020

Russell Coker: Debian PPC64EL Emulation

In my post on Debian S390X Emulation [1] I mentioned having problems booting a Debian PPC64EL kernel under QEMU. Giovanni commented that they had PPC64EL working and gave a link to their site with Debian QEMU images for various architectures [2]. I tried their image which worked then tried mine again which also worked it seemed that a recent update in Debian/Unstable fixed the bug that made QEMU not work with the PPC64EL kernel. Here are the instructions on how to do it. First you need to create a filesystem in an an image file with commands like the following: 
truncate -s 4g /vmstore/ppc
mkfs.ext4 /vmstore/ppc
mount -o loop /vmstore/ppc /mnt/tmp
Then visit the Debian Netinst page [3] to download the PPC64EL net install ISO. Then loopback mount it somewhere convenient like /mnt/tmp2. The package qemu-system-ppc has the program for emulating a PPC64LE system, the qemu-user-static package has the program for emulating PPC64LE for a single program (IE a statically linked program or a chroot environment), you need this to run debootstrap. The following commands should be most of what you need. 
apt install qemu-system-ppc qemu-user-static
update-binfmts --display
# qemu ppc64 needs exec stack to solve "Could not allocate dynamic translator buffer"
# so enable that on SE Linux systems
setsebool -P allow_execstack 1
debootstrap --foreign --arch=ppc64el --no-check-gpg buster /mnt/tmp file:///mnt/tmp2
chroot /mnt/tmp /debootstrap/debootstrap --second-stage
cat << END > /mnt/tmp/etc/apt/sources.list
deb http://mirror.internode.on.net/pub/debian/ buster main
deb http://security.debian.org/ buster/updates main
END
echo "APT::Install-Recommends False;" > /mnt/tmp/etc/apt/apt.conf
echo ppc64 > /mnt/tmp/etc/hostname
# /usr/bin/awk: error while loading shared libraries: cannot restore segment prot after reloc: Permission denied
# only needed for chroot
setsebool allow_execmod 1
chroot /mnt/tmp apt update
# why aren't they in the default install?
chroot /mnt/tmp apt install perl dialog
chroot /mnt/tmp apt dist-upgrade
chroot /mnt/tmp apt install bash-completion locales man-db openssh-server build-essential systemd-sysv ifupdown vim ca-certificates gnupg
# install kernel last because systemd install rebuilds initrd
chroot /mnt/tmp apt install linux-image-ppc64el
chroot /mnt/tmp dpkg-reconfigure locales
chroot /mnt/tmp passwd
cat << END > /mnt/tmp/etc/fstab
/dev/vda / ext4 noatime 0 0
#/dev/vdb none swap defaults 0 0
END
mkdir /mnt/tmp/root/.ssh
chmod 700 /mnt/tmp/root/.ssh
cp ~/.ssh/id_rsa.pub /mnt/tmp/root/.ssh/authorized_keys
chmod 600 /mnt/tmp/root/.ssh/authorized_keys
rm /mnt/tmp/vmlinux* /mnt/tmp/initrd*
mkdir /boot/ppc64
cp /mnt/tmp/boot/[vi]* /boot/ppc64
# clean up
umount /mnt/tmp
umount /mnt/tmp2
# setcap binary for starting bridged networking
setcap cap_net_admin+ep /usr/lib/qemu/qemu-bridge-helper
# afterwards set the access on /etc/qemu/bridge.conf so it can only
# be read by the user/group permitted to start qemu/kvm
echo "allow all" > /etc/qemu/bridge.conf
Here is an example script for starting kvm. It can be run by any user that can read /etc/qemu/bridge.conf. 
#!/bin/bash
set -e
KERN="kernel /boot/ppc64/vmlinux-4.19.0-9-powerpc64le -initrd /boot/ppc64/initrd.img-4.19.0-9-powerpc64le"
# single network device, can have multiple
NET="-device e1000,netdev=net0,mac=02:02:00:00:01:04 -netdev tap,id=net0,helper=/usr/lib/qemu/qemu-bridge-helper"
# random number generator for fast start of sshd etc
RNG="-object rng-random,filename=/dev/urandom,id=rng0 -device virtio-rng-pci,rng=rng0"
# I have lockdown because it does no harm now and is good for future kernels
# I enable SE Linux everywhere
KERNCMD="net.ifnames=0 noresume security=selinux root=/dev/vda ro lockdown=confidentiality"
kvm -drive format=raw,file=/vmstore/ppc64,if=virtio $RNG -nographic -m 1024 -smp 2 $KERN -curses -append "$KERNCMD" $NET
Related posts:
  1. Debian S390X Emulation I decided to setup some virtual machines for different architectures....
  2. installing Xen domU on Debian Etch I have just been installing a Xen domU on Debian...
  3. Installing a Red Hat based DomU on a Debian Dom0 The first step is to copy /images/xen/vmlinuz and /images/xen/initrd.img from...

26 June 2020

Chris Lamb: On the pleasure of hating

People love to tell you that they "don't watch sports" but the story of Lance Armstrong provides a fascinating lens through which to observe our culture at large. For example, even granting all that he did and all the context in which he did it, why do sports cheats act like a lightning rod for such an instinctive hatred? After all, the sheer level of distaste directed at people such as Lance eludes countless other criminals in our society, many of whom have taken a lot more with far fewer scruples. The question is not one of logic or rationality, but of proportionality. In some ways it should be unsurprising. In all areas of life, we instinctively prefer binary judgements to moral ambiguities and the sports cheat is a clich of moral bankruptcy cheating at something so seemingly trivial as a sport actually makes it more, not less, offensive to us. But we then find ourselves strangely enthralled by them, drawn together in admiration of their outlaw-like tenacity, placing them strangely close to criminal folk heroes. Clearly, sport is not as unimportant as we like to claim it is. In Lance's case in particular though, there is undeniably a Shakespearean quality to the story and we are forced to let go of our strict ideas of right and wrong and appreciate all the nuance.

There is a lot of this nuance in Marina Zenovich's new documentary. In fact, there's a lot of everything. At just under four hours, ESPN's Lance combines the duration of a Tour de France stage with the depth of the peloton an endurance event compared to the bite-sized hagiography of Michael Jordan's The Last Dance. Even for those who follow Armstrong's story like a mini-sport in itself, Lance reveals new sides to this man for all seasons. For me, not only was this captured in his clumsy approximations at being a father figure but also in him being asked something I had not read in countless tell-all books: did his earlier experiments in drug-taking contribute to his cancer? But even in 2020 there are questions that remain unanswered. By needlessly returning to the sport in 2009, did Lance subconsciously want to get caught? Why does he not admit he confessed to Betsy Andreu back in 1999 but will happily apologise to her today for slurring her publicly on this very point? And why does he remain so vindictive towards former-teammate Floyd Landis? In all of Armstrong's evasions and masterful control of the narrative, there is the gnawing feeling that we don't even know what questions we should be even asking. As ever, the questions are more interesting than the answers.

Lance also reminded me of how professional cycling's obsession with national identity. Although I was intuitively aware of it to some degree, I had not fully grasped how much this kind of stereotyping runs through the veins of the sport itself, just like the drugs themselves. Journalist Daniel Friebe first offers us the portrait of:
Spaniards tend to be modest, very humble. Very unpretentious. And the Italians are loud, vain and outrageous showmen.
Former directeur sportif Johan Bruyneel then asserts that "Belgians are hard workers... they are ambitious to a certain point, but not overly ambitious", and cyclist J rg Jaksche concludes with:
The Germans are very organised and very structured. And then the French, now I have to be very careful because I am German, but the French are slightly superior.
This kind of lazy caricature is nothing new, especially for those brought up on a solid diet of Tintin and Asterix, but although all these examples are seemingly harmless, why does the underlying idea of ascribing moral, social or political significance to genetic lineage remain so durable in today's age of anti-racism? To be sure, culture is not quite the same thing as race, but being judged by the character of one's ancestors rather than the actions of an individual is, at its core, one of the many conflations at the heart of racism. There is certainly a large amount of cognitive dissonance at work, especially when Friebe elaborates:
East German athletes were like incredible robotic figures, fallen off a production line somewhere behind the Iron Curtain...
... but then bermensch Jan Ullrich is immediately described as "emotional" and "struggled to live the life of a professional cyclist 365 days a year". We see the habit to stereotype is so ingrained that even in the face of this obvious contradiction, Friebe unironically excuses Ullrich's failure to live up his German roots due to him actually being "Mediterranean".

I mention all this as I am known within my circles for remarking on these national characters, even collecting stereotypical examples of Italians 'being Italian' and the French 'being French' at times. Contrary to evidence, I don't believe in this kind of innate quality but what I do suspect is that people generally behave how they think they ought to behave, perhaps out of sheer imitation or the simple pleasure of conformity. As the novelist Will Self put it:
It's quite a complicated collective imposture, people pretending to be British and people pretending to be French, and then they get really angry with each other over what they're pretending to be.
The really remarkable thing about this tendency is that even if we consciously notice it there is no seemingly no escape even I could not smirk when I considered that a brash Texan winning the Tour de France actually combines two of America's cherished obsessions: winning... and annoying the French.

22 June 2020

Dirk Eddelbuettel: #27: R and CRAN Binaries for Ubuntu

Welcome to the 27th post in the rationally regularized R revelations series, or R4 for short. This is a edited / updated version of yesterday s T^4 post #7 as it really fits the R4 series as well as it fits the T4 series. A new video in both our T^4 series of video lightning talks with tips, tricks, tools, and toys is also a video in the R^4 series as it revisits a topic previously covered in the latter: how to (more easily) get (binary) packages onto your Ubuntu system. In fact, we show it in three different ways.

The slides are here. This repo at GitHub support the series: use it to open issues for comments, criticism, suggestions, or feedback. Thanks to I aki Ucar who followed up on twitter with a Fedora version. We exchanged some more message, and concluded that complete comparison (from an empty Ubuntu or Fedora container) to a full system takes about the same time on either system. If you like this or other open-source work I do, you can now sponsor me at GitHub. For the first year, GitHub will match your contributions.

This post by Dirk Eddelbuettel originated on his Thinking inside the box blog. Please report excessive re-aggregation in third-party for-profit settings.

Next.

Previous.