Search Results: "tar"

4 November 2022

Dirk Eddelbuettel: RProtoBuf 0.4.20 on CRAN: Maintenance

A new release 0.4.20 of RProtoBuf arrived on CRAN earlier today. RProtoBuf provides R with bindings for the Google Protocol Buffers ( ProtoBuf ) data encoding and serialization library used and released by Google, and deployed very widely in numerous projects as a language and operating-system agnostic protocol. This release is somewhat mechanical and in the spirit of many other recent releases. clang-15 is, as more recent compilers do, more stringent on definitions and wants to see some void in argument-less signatures. Happy to oblige. At the same time, GitHub Actions started to nag us about minimum versions of node code so an upgrade to a newer action is warranted, again as with many other affected packages. We also found another http:// URL hiding somewhere so that was cleaned. Lastly, it appears Protocol Buffers themselves moved on and now need / prefer C++17 so were happy to oblige. The following section from the NEWS.Rd file has full details.

Changes in RProtoBuf version 0.4.20 (2022-11-02)
  • Two function prototype were updated for clang-15.
  • GitHub Actions were updated to checkout version 3.
  • One more http:// URL was updated to https://
  • The package compilation is now donw with C++17 as the Protocol Buffer headers require it (and clang-15 nags otherwise).

Thanks to my CRANberries, there is a diff to the previous release. The RProtoBuf page has copies of the (older) package vignette, the quick overview vignette, and the pre-print of our JSS paper. Questions, comments etc should go to the GitHub issue tracker off the GitHub repo. If you like this or other open-source work I do, you can now sponsor me at GitHub.

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

3 November 2022

Junichi Uekawa: user namespace not enabled when running podman.

user namespace not enabled when running podman. Completely forgot about this but seems like I need to enable user namespace for podman. sudo sysctl -w kernel.unprivileged_userns_clone=1. Not quite sure when this started being required.

Russ Allbery: Review: Carpe Jugulum

Review: Carpe Jugulum, by Terry Pratchett
Series: Discworld #23
Publisher: Harper
Copyright: 1998
Printing: May 2014
ISBN: 0-06-228014-7
Format: Mass market
Pages: 409
Carpe Jugulum is the 23rd Discworld novel and the 6th witches novel. I would not recommend reading it before Maskerade, which introduces Agnes. There are some spoilers for Wyrd Sisters, Lords and Ladies, and Maskerade in the setup here and hence in the plot description below. I don't think they matter that much, but if you're avoiding all spoilers for earlier books, you may want to skip over this one. (You're unlikely to want to read it before those books anyway.) It is time to name the child of the king of Lancre, and in a gesture of good will and modernization, he has invited his neighbors in Uberwald to attend. Given that those neighbors are vampires, an open invitation was perhaps not the wisest choice. Meanwhile, Granny Weatherwax's invitation has gone missing. On the plus side, that meant she was home to be summoned to the bedside of a pregnant woman who was kicked by a cow, where she makes the type of hard decision that Granny has been making throughout the series. On the minus side, the apparent snub seems to send her into a spiral of anger at the lack of appreciation. Points off right from the start for a plot based on a misunderstanding and a subsequent refusal of people to simply talk to each other. It is partly engineered, but still, it's a cheap and irritating plot. This is an odd book. The vampires (or vampyres, as the Count wants to use) think of themselves as modern and sophisticated, making a break from the past by attempting to overcome such traditional problems as burning up in the sunlight and fear of religious symbols and garlic. The Count has put his family through rigorous training and desensitization, deciding such traditional vulnerabilities are outdated things of the past. He has, however, kept the belief that vampires are at the top of a natural chain of being, humans are essentially cattle, and vampires naturally should rule and feed on the population. Lancre is an attractive new food source. Vampires also have mind control powers, control the weather, and can put their minds into magpies. They are, in short, enemies designed for Granny Weatherwax, the witch expert in headology. A shame that Granny is apparently off sulking. Nanny and Agnes may have to handle the vampires on their own, with the help of Magrat. One of the things that makes this book odd is that it seemed like Pratchett was setting up some character growth, giving Agnes a chance to shine, and giving Nanny Ogg a challenge that she didn't want. This sort of happens, but then nothing much comes of it. Most of the book is the vampires preening about how powerful they are and easily conquering Lancre, while everyone else flails ineffectively. Pratchett does pull together an ending with some nice set pieces, but that ending doesn't deliver on any of the changes or developments it felt like the story was setting up. We do get a lot of Granny, along with an amusingly earnest priest of Om (lots of references to Small Gods here, while firmly establishing it as long-ago history). Granny is one of my favorite Discworld characters, so I don't mind that, but we've seen Granny solve a lot of problems before. I wanted to see more of Agnes, who is the interesting new character and whose dynamic with her inner voice feels like it has a great deal of unrealized potential. There is a sharp and condensed version of comparative religion from Granny, which is probably the strongest part of the book and includes one of those Discworld quotes that has been widely repeated out of context:
"And sin, young man, is when you treat people as things. Including yourself. That's what sin is." "It's a lot more complicated than that " "No. It ain't. When people say things are a lot more complicated than that, they means they're getting worried that they won t like the truth. People as things, that's where it starts."
This loses a bit in context because this book is literally about treating people as things, and thus the observation feels more obvious when it arrives in this book than when you encounter it on its own, but it's still a great quote. Sadly, I found a lot of this book annoying. One of those annoyances is a pet peeve that others may or may not share: I have very little patience for dialogue in phonetically-spelled dialect, and there are two substantial cases of that here. One is a servant named Igor who speaks with an affected lisp represented by replacing every ess sound with th, resulting in lots of this:
"No, my Uncle Igor thtill workth for him. Been thtruck by lightning three hundred timeth and thtill putth in a full night'th work."
I like Igor as a character (he's essentially a refugee from The Addams Family, which adds a good counterpoint to the malicious and arrogant evil of the vampires), but my brain stumbles over words like "thtill" every time. It's not that I can't decipher it; it's that the deciphering breaks the flow of reading in a way that I found not at all fun. It bugged me enough that I started skipping his lines if I couldn't work them out right away. The other example is the Nac Mac Feegles, who are... well, in the book, they're Pictsies and a type of fairy, but they're Scottish Smurfs, right down to only having one female (at least in this book). They're entertainingly homicidal, but they all talk like this:
"Ach, hins tak yar scaggie, yer dank yowl callyake!"
I'm from the US and bad with accents and even worse with accents reproduced in weird spellings, and I'm afraid that I found 95% of everything said by Nac Mac Feegles completely incomprehensible to the point where I gave up even trying to read it. (I'm now rather worried about the Tiffany Aching books and am hoping Pratchett toned the dialect down a lot, because I'm not sure I can deal with more of this.) But even apart from the dialect, I thought something was off about the plot structure of this book. There's a lot of focus on characters who don't seem to contribute much to the plot resolution. I wanted more of the varied strengths of Lancre coming together, rather than the focus on Granny. And the vampires are absurdly powerful, unflappable, smarmy, and contemptuous of everyone, which makes for threatening villains but also means spending a lot of narrative time with a Discworld version of Jacob Rees-Mogg. I feel like there's enough of that in the news already. Also, while I will avoid saying too much about the plot, I get very suspicious when older forms of oppression are presented as good alternatives to modernizing, rationalist spins on exploitation. I see what Pratchett was trying to do, and there is an interesting point here about everyone having personal relationships and knowing their roles (a long-standing theme of the Lancre Discworld stories). But I think the reason why there is some nostalgia for older autocracy is that we only hear about it from stories, and the process of storytelling often creates emotional distance and a patina of adventure and happy outcomes. Maybe you can make an argument that classic British imperialism is superior to smug neoliberalism, but both of them are quite bad and I don't want either of them. On a similar note, Nanny Ogg's tyranny over her entire extended clan continues to be played for laughs, but it's rather unappealing and seems more abusive the more one thinks about it. I realize the witches are not intended to be wholly good or uncomplicated moral figures, but I want to like Nanny, and Pratchett seems to be writing her as likable, even though she has an astonishing lack of respect for all the people she's related to. One might even say that she treats them like things. There are some great bits in this book, and I suspect there are many people who liked it more than I did. I wouldn't be surprised if it was someone's favorite Discworld novel. But there were enough bits that didn't work for me that I thought it averaged out to a middle-of-the-road entry. Followed by The Fifth Elephant in publication order. This is the last regular witches novel, but some of the thematic thread is picked up by The Wee Free Men, the first Tiffany Aching novel. Rating: 7 out of 10

2 November 2022

Robert McQueen: Many thanks & good luck to Neil McGovern

As President of the GNOME Foundation, I wanted to post a quick note to pass on the thanks from the Board, the Foundation staff team and membership to our outgoing Executive Director, Neil McGovern. I had the pleasure of passing on GNOME s thanks in person at the Casa Bariachi this summer at GUADEC in Guadelajara, at the most exellent mariachi celebration of GNOME s 25th Anniversary. Kindly they stopped the music and handed me the microphone for the whole place, although I think many of the other guests celebrating their own birthdays were less excited about Neil s tenure as Executive Director and the Free and Open Source desktop in general.

Neil s 6-month handover period came to an end last month and he handed over the reins to myself and Thibault Martin on the Executive Committee, and Director of Operations Rosanna Yuen has stepped up to act as Chief of Staff and interface between the Board and the staff team for the time being. Our recruitment is ongoing for a new Executive Director although the search is a little behind schedule (mostly down to me!), and we re hugely grateful to a few volunteers who have joined our search committee to help us source, screen and interview applicants.

I have really enjoyed working closely with Neil in my time on the GNOME board, and we are hugely grateful for his contributions and achievements over the past 5 years which I posted about earlier in the year. Neil is this month starting a new role as the Executive Director of Ruby Central. Our very best wishes from the GNOME community and good luck with your new role. See you soon! (also posted to Discourse if you wish to add any thanks or comments of your own)

1 November 2022

Paul Tagliamonte: Decoding LDPC: k-Bit Brute Forcing

Before you go on: I've been warned off implementing this in practice on a few counts; namely, the space tradeoff isn't worth it, and it's unlikely to correct meaningful errors. I'm going to leave this post up, but please do take the content with a very large grain of salt!
My initial efforts to build a PHY and Data Link layer from scratch using my own code have been progressing nicely since the initial BPSK based protocol I ve documented under the PACKRAT series. As part of that, I ve been diving deep into FEC, and in particular, LDPC. I won t be able to do an overview of LDPC justice in this post with any luck that ll come in a later post to come as part of the RATPACK series, so some knowledge is assumed. As such this post is less useful for those looking to learn about LDPC, and a bit more targeted to those who enjoy talking and thinking about FEC.
Hey, heads up! - This post contains extremely unvalidated and back of the napkin quality work without any effort to prove this out generally. Hopefully this work can be of help to others, but please double check anything below if you need it for your own work!
While implementing LDPC, I ve gotten an encoder and checker working, enough to use LDPC like a checksum. The next big step is to write a Decoder, which can do error correction. The two popular approaches for the actual correction that I ve seen while reading about LDPC are Belief Propagation, and some class of linear programming that I haven t dug into yet. I m not thrilled at how expensive this all is in software, so while implementing the stack I ve been exploring every shady side ally to try and learn more about how encoders and decoders work, both in theory - and in practice.

Processing an LDPC Message Checking if a message is correct is fairly straightforward with LDPC (as with encoding, I ll note). As a quick refresher given the LDPC H (check) matrix of width N, you can check your message vector (msg) of length N by multipling H and msg, and checking if the output vector is all zero. 
 // scheme contains our G (generator) and
 // H (check) matrices.
 scheme :=  G: Matrix ... , H: Matrix ... 
// msg contains our LDPC message (data and
 // check bits).
 msg := Vector ... 
// N is also the length of the encoded
 // msg vector after check bits have been
 // added.
 N := scheme.G.Width
// Now, let's generate our 'check' vector.
 ch := Multiply(scheme.H, msg)
We can now see if the message is correct or not: 
 // if the ch vector is all zeros, we know
 // that the message is valid, and we don't
 // need to do anything.
 if ch.IsZero()  
// handle the case where the message
 // is fine as-is.
 return ...
 
// Expensive decode here
This is great for getting a thumbs up / thumbs down on the message being correct, but correcting errors still requires pulling the LDPC matrix values from the g (generator) matrix out, building a bipartite graph, and iteratively reprocessing the bit values, until constraints are satisfied and the message has been corrected. This got me thinking - what is the output vector when it s not all zeros? Since 1 values in the output vector indicates consistency problems in the message bits as they relate to the check bits, I wondered if this could be used to speed up my LDPC decoder. It appears to work, so this post is half an attempt to document this technique before I put it in my hot path, and half a plea for those who do like to talk about FEC to tell me what name this technique actually is.

k-Bit Brute Forcing Given that the output Vector s non-zero bit pattern is set due to the position of errors in the message vector, let s use that fact to build up a table of k-Bit errors that we can index into. 
 // for clarity's sake, the Vector
 // type is being used as the lookup
 // key here, even though it may
 // need to be a hash or string in
 // some cases.
 idx := map[Vector]int 
for i := 0; i < N; i++  
// Create a vector of length N
 v := Vector 
v.FlipBit(i)
// Now, let's use the generator matrix to encode
 // the data with checksums, and then use the
 // check matrix on the message to figure out what
 // bit pattern results
 ev := Multiply(scheme.H, Multiply(v, scheme.G))
idx[ev] = i
This can be extended to multiple bits (hence: k-Bits), but I ve only done one here for illustration. Now that we have our idx mapping, we can now go back to the hot path on Checking the incoming message data: 
 // if the ch vector is all zeros, we know
 // that the message is valid, and we don't
 // need to do anything.
 if ch.IsZero()  
// handle the case where the message
 // is fine as-is.
 return ...
 
errIdx, ok := idx[ch]
if ok  
msg.FlipBit(errIdx)
// Verify the LDPC message using
 // H again here.
 return ...
 
// Expensive decode here
Since map lookups wind up a heck of a lot faster than message-passing bit state, the hope here is this will short-circuit easy to solve errors for k-Bits, for some value of k that the system memory can tolerate.

Does this work? Frankly I have no idea. I ve written a small program and brute forced single-bit errors in all bit positions using random data to start with, and I ve not been able to find any collisions in the 1-bit error set, using the LDPC matrix from 802.3an-2006. Even if I was to find a collision for a higher-order k-Bit value, I m tempted to continue with this approach, and treat each set of bits in the Vector s bin (like a hash-table), checking the LDPC validity after each bit set in the bin. As long as the collision rate is small enough, it should be possible to correct k-Bits of error faster than the more expensive Belief Propagation approach. That being said, I m not entirely convinced collisions will be very common, but it ll take a bit more time working through the math to say that with any confidence. Have you seen this approach called something official in publications? See an obvious flaw in the system? Send me a tip, please!

Jonathan Dowland: Halloween playlist 2022

I hope you had a nice Halloween! I've collected together some songs that I've enjoyed over the last couple of years that loosely fit a theme: ambient, instrumental, experimental, industrial, dark, disconcerting, etc. I've prepared a Spotify playlist of most of them, but not all. The list is inline below as well, with many (but not all) tracks linking to Bandcamp, if I could find them there. This is a bit late, sorry. If anyone listens to something here and has any feedback I'd love to hear it. (If you are reading this on an aggregation site, it's possible the embeds won't work. If so, click through to my main site) Spotify playlist: https://open.spotify.com/playlist/3bEvEguRnf9U1RFrNbv5fk?si=9084cbf78c364ac8; The list, with Bandcamp embeds where possible: Some sources
  1. Via Stuart Maconie's Freak Zone
  2. Via Mary Anne Hobbs
  3. Via Lose yourself with
  4. Soma FM - Doomed (Halloween Special)

Russ Allbery: Review: What Makes This Book So Great

Review: What Makes This Book So Great, by Jo Walton
Publisher: Tor
Copyright: January 2014
ISBN: 0-7653-3193-4
Format: Hardcover
Pages: 447
Jo Walton, in addition to being an excellent science fiction and fantasy writer, is a prodigious reader and frequent participant in on-line SFF book discussion going back to the Usenet days. This book is a collection of short essays previously published on Tor.com between July 2008 and February 2011. The unifying theme is that Walton regularly re-reads her favorite books, and each essay (apart from some general essays on related topics) is about why this specific book is one that she re-reads, and (as the title says) what makes it so great. Searching for the title of one of the essays turns it up on Tor.com still, so this is one of those collections that you don't have to buy since you can read its contents on-line for free. That said, it looks like these essays were from before Tor.com started classifying posts into series, so it's going to be challenging to track them down in the huge number of other articles Walton has written for the site. (That said, you can't go far wrong by reading any of her essays at random.) I read these essays as they were originally published, so this was also a re-read for me, but it had been a while. I'm happy to report that they were just as much fun the second time. In the introduction and in the final essay of this collection, Walton draws a distinction between what she's doing, criticism, and reviewing. As someone else who writes about books (in a far more amateur fashion), I liked this distinction. The way I'd characterize it is that criticism is primarily about the work: taking it apart to see what makes it tick, looking for symbolism and hidden meanings, and comparing and contrasting other works that are tackling similar themes. I've often finished a work of criticism and still had no idea if the author enjoyed reading the work being criticized or not, since that isn't the point. Reviewing is assistance to consumers and focuses more on the reader: would you enjoy this book? Is it enjoyable to read? Does it say something new? What genre and style is it in, so that you can match that to your tastes? Talking about books is neither of those things, although it's a bit closer to reviewing. But the emphasis is on one's personal enjoyment instead of attempting to review a product for others. When I talk about books with friends, I talk primarily about what bits I liked, what bits I didn't like, where the emotional beats were for me, and what interesting things the book did that surprised me or caught my attention. One can find a review in there, and sometimes even criticism, but the focus and the formality is different. (And, to be honest, my reviews are more on the "talking about the book" side than fully proper reviews.) These essays are indeed talking about books. They're all re-reads; in some cases the first re-read, but more frequently the latest of many re-reads. There are lots of spoilers, which makes for bad reviews (the target audience of a review hasn't read the book yet) but good fodder for conversations about books. (The spoilers are mostly marked, but if you're particularly averse to spoilers, you'll need to read carefully.) Most of the essays are about a single book, but there are a few on more general topics, such as Walton's bafflement that anyone would skim a novel. Since these are re-reads, and the essays collected here are more than a decade old, the focus is on older books. Some of them are famous: Vinge's A Fire Upon the Deep and A Deepness in the Sky, early Le Guin, Samuel Delaney's SF novels, Salmon Rushdie's Midnight's Children. Some of them are more obscure. C.J. Cherryh, for example, is a writer who never seems to get much on-line attention, but who is one of Walton's favorites. Most of the essays stand alone or come in small clusters about a writer, often sprinkled through the book instead of clumped together. (The book publishes the essays in the same order they originally appeared on Tor.com.) The two largest groups of essays are re-readings of every book in Steven Brust's Vlad Taltos universe (including Brokedown Palace and the Paarfi books) up to Jhegaala, and every book in Lois McMaster Bujold's Miles Vorkosigan series up to Diplomatic Immunity. This is fitting: those are two of the great series of science fiction, but don't seem to be written about nearly as much as I would expect. There are over 130 essays in a 447 page book, so there's a lot of material here and none of them outlive their welcome. Walton has a comfortable, approachable style that bubbles with delight and appreciation for books. I think it's impossible to read this collection without wanting to read more, and without adding several more books to the ever-teetering to-read pile. This is perhaps not the best source of reading recommendations if you dislike spoilers, although it can be used for that if you read carefully. But if you love listening to conversations about the genre and talking about how books bounce off each other, and particularly if you have read most of these books already or don't mind spoilers, this collection is a delight. If you're the type of SFF reader who likes reading the reviews in Locus or is already reading Tor.com, highly recommended. Rating: 8 out of 10

Paul Wise: FLOSS Activities October 2022

Focus This month I didn't have any particular focus. I just worked on issues in my info bubble.

Changes

Issues

Review

Administration
  • Debian BTS: unarchive/reopen/triage bugs for reintroduced packages nautilus-image-converter, swift-im, runit-services
  • Debian IRC: removed 2 spammers from OFTC, disable anti-spam channel modes for some channels
  • Debian servers: restart processes due to OOM
  • Debian wiki: approve accounts

Communication
  • Initiate discussion about the apt hook protocol
  • Respond to queries from Debian users and contributors on the mailing lists and IRC

Sponsors All work was done on a volunteer basis.

Louis-Philippe V ronneau: Montreal's Debian & Stuff - October 2022

Our local Debian user group gathered on Sunday October 30th to chat, work on Debian and do other, non-Debian related hacking :) This time around, we met at EfficiOS's1 offices. As you can see from the following picture, it's a great place and the view they have is pretty awesome. Many thanks for hosting us! The view from EfficiOS' offices, overlooking the Mont-Royal This was our 4th meeting this year and once again, attendance was great: 10 people showed up to work on various things. Following our bi-monthly schedule, our next meeting should be in December, but I'm not sure it'll happen. December can be a busy month here and I will have to poke our mailing list to see if people have the spoons for an event. This time around, I was able to get a rough log of the Debian work people did: pollo: mjeanson: viashimo: lavamind: anarcat: babelouest: tvaz: As always, thanks to the Debian project for granting us a budget to buy some food!
  1. Makers of the awesome LTTng project, amongst other things.

31 October 2022

Jonathan Dowland: Cosey Fanni Tutti - Guitar/Amplitude

I haven't posted a crate digging picture for a while: Amongst other distractions, I needed to replace my turntable needle. The dead needle lasted me about 20 years, although a fair chunk of that was not spent not doing anything.
yellow 7inch record playing
I love Cosey Fanni Tutti's solo work, as well as her stuff with Chris Carter and some other collaborations: although I was familiar with Throbbing Gristle, it was the live experiment Carter Tutti Void and the recording of it pressed as Transverse that got me interested. "Tutti" was one of my top albums of 2019. Her autobiography Art, Sex, Music was fascinating. I haven't read the follow-up, Re-Sisters yet, but it's on my list. The Arena documentary she was involved with DELIA DERBYSHIRE: THE MYTHS AND THE LEGENDARY TAPES was fabulous. I've yet to hear Tutti's soundtrack album taken from it. This morning's record is a 7" from Electronic Sound magazine with two dark ambient instrumental pieces.

30 October 2022

Utkarsh Gupta: FOSS Activites in October 2022

Here s my (thirty-seventh) monthly but brief update about the activities I ve done in the F/L/OSS world.

Debian
This was my 46th month of actively contributing to Debian. I became a DM in late March 2019 and a DD on Christmas 19! \o/ There s a bunch of things I do, both, technical and non-technical. Here are the things I did this month:

Debian Uploads
  • ruby-espeak (1.1.0-1) - New upstream version, v1.1.0.

Other $things:
  • Being an AM for Arun Kumar, process #1024. Process completed. \o/
  • Sponsoring stuff for non-DDs.
  • Mentoring for newcomers.
  • Moderation of -project mailing list.

Ubuntu
This was my 21st month of actively contributing to Ubuntu. Now that I joined Canonical to work on Ubuntu full-time, there s a bunch of things I do! \o/ I mostly worked on different things, I guess. I was too lazy to maintain a list of things I worked on so there s no concrete list atm. Maybe I ll get back to this section later or will start to list stuff from the fall, as I was doing before. :D

Debian (E)LTS
Debian Long Term Support (LTS) is a project to extend the lifetime of all Debian stable releases to (at least) 5 years. Debian LTS is not handled by the Debian security team, but by a separate group of volunteers and companies interested in making it a success. And Debian Extended LTS (ELTS) is its sister project, extending support to the Jessie release (+2 years after LTS support). This was my thirty-seventh month as a Debian LTS and twenty-eighth month as a Debian ELTS paid contributor.
I worked for 35.00 hours for LTS and 25.00 hours for ELTS.

LTS CVE Fixes and Announcements:
  • Issued DLA 3146-1, fixing CVE-2022-2928 and CVE-2022-2929, for isc-dhcp.
    For Debian 10 buster, these problems have been fixed in version 4.4.1-2+deb10u2.
  • Issued DLA 3165-1, fixing CVE-2022-43680, for expat.
    For Debian 10 buster, these problems have been fixed in version 2.2.6-2+deb10u6.
  • Issued DLA 3166-1, fixing CVE-2022-29970, for ruby-sinatra.
    For Debian 10 buster, these problems have been fixed in version 2.0.5-4+deb10u1.
  • Uploaded dropbear to fix CVE-2021-36369 in buster. Waiting for ELTS upload to issue the DLA. But will do soon now.
  • src:joblib is a bit painful - having to backport patches to Py2. :/
  • Started to look at other set of packages.

ELTS CVE Fixes and Announcements:
  • Issued ELA 715-1, fixing CVE-2022-43680, for expat.
    For Debian 9 stretch, these problems have been fixed in version 2.2.0-2+deb9u7.
    For Debian 8 jessie, these problems have been fixed in version 2.1.0-6+deb8u10.
  • Issued ELA 716-1, fixing CVE-2018-25045 and CVE-2020-25626, for djangorestframework.
    For Debian 9 stretch, these problems have been fixed in version 3.4.0-2+deb9u1.
  • Uploaded dropbear to fix CVE-2021-36369 in buster. Waiting for ELTS upload, too. But some backporting problems. :/
  • src:joblib is a bit painful - having to backport patches to Py2. :/
  • Started to look at other set of packages.

Other (E)LTS Work:
  • Triaged joblib, dropbear, ruby-sinatra, djangorestframework, isc-dhcp, and expat.
  • Reverted Mark freerdp CVEs wrongly affecting freerdp <2.0.0 in the ELTS tracker.
  • Helped and assisted new contributors joining Freexian (LTS/ELTS).
  • Answered questions (& discussions) on IRC (#debian-lts and #debian-elts) and Matrix.
  • Participated and helped fellow members with their queries via private mail and chat.
  • General and other discussions on LTS private and public mailing list.
  • Attended the monthly meeting held on Jitsi on October 27th.
Until next time.
:wq for today.

29 October 2022

Dirk Eddelbuettel: littler 0.3.17 on CRAN: Maintenance

max-heap image The eighteenth release of littler as a CRAN package just landed, following in the now sixteen year history (!!) as a package started by Jeff in 2006, and joined by me a few weeks later. littler is the first command-line interface for R as it predates Rscript. It allows for piping as well for shebang scripting via #!, uses command-line arguments more consistently and still starts faster. It also always loaded the methods package which Rscript only started to do in recent years. littler lives on Linux and Unix, has its difficulties on macOS due to yet-another-braindeadedness there (who ever thought case-insensitive filesystems as a default were a good idea?) and simply does not exist on Windows (yet the build system could be extended see RInside for an existence proof, and volunteers are welcome!). See the FAQ vignette on how to add it to your PATH. A few examples are highlighted at the Github repo, as well as in the examples vignette. This release, coming just a few weeks since the last release in August, heeds to clang-15 a updates one signature to a proper interface. It also contains one kindly contributed patch updating install2.r (and installBioc.r) to cope with a change in R-devel. The full change description follows.

Changes in littler version 0.3.17 (2022-10-29)
  • Changes in package
    • An internal function prototype was updated for clang-15.
  • Changes in examples
    • The install2.r and installBioc. were updated for an update in R-devel (Tatsuya Shima and Dirk in #104).

My CRANberries service provides a comparison to the previous release. Full details for the littler release are provided as usual at the ChangeLog page, and also on the package docs website. The code is available via the GitHub repo, from tarballs and now of course also from its CRAN page and via install.packages("littler"). Binary packages are available directly in Debian as well as soon via Ubuntu binaries at CRAN thanks to the tireless Michael Rutter. Comments and suggestions are welcome at the GitHub repo. If you like this or other open-source work I do, you can now sponsor me at GitHub.

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

Russ Allbery: California general election

As usual with these every-two-year posts, probably of direct interest only to California residents. Maybe the more obscure things we're voting on will be a minor curiosity to people elsewhere. Apologies to Planet Debian readers for the explicitly political post because I'm too lazy to change my blog software to do more fine-grained post classification. For what it's worth, most of the discussion here will be about the more fiddly and nuanced things we vote on, not on the major hot-button proposition. As in 2020, I'm only going to cover the ballot propositions, as all of the state-wide and most of the district races are both obvious to me and boring to talk about. The hyperlocal races are more interesting this year, but the number of people who would care and who are also reading this blog is essentially nonexistent, so I won't bother writing them up. This year, everything except Proposition 1 is an initiative (not put on the ballot by the legislature), which means I default to voting against them because they're usually poorly-written. Proposition 1: YES. Adds reproductive rights to the California state constitution. I'm fairly sure everyone reading this has already made up their mind on this topic and certainly nothing will ever change my mind, so I'll leave it at that. Proposition 26: YES. This mushes two different things together in an unhelpful way: allowing sports betting at some racetracks, and allowing a wider variety of gambling on tribal lands. I have no strong opinion about the former (I'll get into that more with the next proposition). For the latter, my starting point is that Native American tribes are and should be treated like independent governments with their own laws (which is what we promised them by treaty and then have systematically and maliciously betrayed ever since). I am not a citizen of any of the tribes and therefore fundamentally I should not get a say on this. I'm not a big fan of gambling or of the companies they're likely to hire to run casinos, but it should be their land and their decision. Proposition 27: NO. This, on the other hand, is about on-line sports betting outside of tribal lands, and looks to be a lot more about corruption and corporate greed. I am fairly dubious that outlawing gambling in general is that good of an idea. I think the harms are overstated given the existence of even wilder forms of gambling (crypto and financial derivatives) that are perfectly legal, and I'm always suspicious of attempting to solve social problems with police and prohibition systems. If there were a ballot proposition to simply legalize gambling in California, I'd have to think hard about that. But this is not that. This requires companies that want to offer on-line gambling to pay substantial up-front costs (which will restrict this to only huge gambling companies). In return, they are allowed entry into what is essentially a state-constructed partial monopoly. As usual, there's a typical vice tax deal attached where those companies are taxed to fund some program (in this case, homeless services and mental health treatment), but these sorts of taxes tend to be regressive in effect. We could just tax richer people like me to pay for those services instead. I'm also dubious that the money for homelessness will be used to build housing, which is what we need to do to address the problem. Proposition 28: YES. Sets aside money for art and music funding in public and charter schools. This is a reluctant yes because this sort of law should not be done via proposition; it should be done through a normal legislative process that balances all of the priorities for school funding. But despite the broken process by which this was put forward, it seems like a reasonable law and no one is opposing it, so okay, fine. Proposition 29: NO. The attempt to force all dialysis clinics to have licensed doctors on site is back again. Everything about the way dialysis health care is provided in California makes me angry. We should have a state health care system similar to the NHS. We should open dialysis clinics based on the number of people requiring dialysis in that area. Every one of them should be unionized. We absolutely should not allow for-profit companies to have primary responsibility for basic life-saving medical care like dialysis. But this proposition does not solve any of those problems, and what it claims to do is false. It claims that by setting credential requirements on who has to be on-site at a dialysis clinic, the clinics will become safer. This is simply not true, for all of the reasons discussed in Still Not Safe. This is not how safety works. The safest person to do dialysis is someone with extensive experience in performing dialysis, who has seen all the problems and has an intuition for what to watch out for. That has less to do with credentials than with good training specifically in dialysis, apprenticeship, and practice, not to mention reasonable hours and good pay so that the workers are not stressed. Do I think the private dialysis clinics are likely doing a good job with this? Hah. (Do I think dialysis clinics run by large medical non-profits would do a good job with this? Also hah.) But this would enshrine into law a fundamentally incorrect solution to the problem that makes dialysis more expensive without addressing any of the other problems with the system. It's the same tactic that was used on abortion clinics, with the same bogus argument that having people with specific credentials on-site would make them safer. It was false then and it's still false now. I would agree with better regulation of dialysis clinics, but this specific regulation is entirely wrong-headed. Also, while this isn't an overriding factor, I get annoyed when the same proposition shows up again without substantial changes. For matters of fundamental rights, okay, sure. But for technical regulation fixes like this one, the proponents should consider taking no for an answer and trying a different approach. Like going to the legislature, which is where this kind of regulation should be designed anyway. Proposition 30: YES. Raises taxes on the personal income of extremely rich Californians (over $2 million in income in one year) to fund various climate change mitigation programs. This is another reluctant yes vote, because once again this shouldn't be done by initiative and should be written properly by the legislature. I also don't like restricting tax revenue to particular programs, which reduce budget flexibilty to no real purpose. It's not important to me that these revenues go to these specific programs, although the programs seem like good ones to fund. But the reality remains that wealthy Silicon Valley executives are undertaxed and the only way we can ever manage to raise taxes is through voting for things like this, so fine. Proposition 31: NO. The Calfornia legislature banned the sale of flavored vape products. If NO beats YES on this proposition, that ban will be overturned. Drug prohibition has never, ever worked, and yet we keep trying it over and over again in the hope that this time we'll get a different outcome. As usual, the pitch in favor of this is all about the children, specifically the claim that flavored tobacco products are only about increasing their appeal to kids because... kids like candy? Or something? I am extremely dubious of this argument; it's obvious to me from walking around city streets that adults prefer the flavored products as well and sale to kids is already prohibited and unchanged by this proposition. I don't like vaping. I wish people would stop, at least around me, because the scent is obnoxious and the flavored stuff is even more obnoxious, even apart from whatever health problems it causes. But I'm never going to vote for drug prohibition because drug prohibition doesn't work. It just creates a black market and organized crime and makes society overall worse. Yes, the tobacco companies are some of the worst corporations on the planet, and I hope they get sued into oblivion (and ideally prosecuted) for all the lying they do, but I'm still not going to vote for prohibition. Even the best kind of prohibition that only outlaws sale and not possession. Also, secondarily but still significant, bans like this just frustrate a bunch of people and burn good will and political capital, which we should be trying to preserve to tackle far more important problems. The politics of outlawing people's pleasures for their own good are not great. We have a lot of serious problems to deal with; maybe let's not pick fights we don't have to.

Fran ois Marier: Making the mounting of an encrypted /home optional on a home server

I have a computer that serves as a home server as well as a desktop machine. It has an encrypted home directory to protect user files and, in the default configuration, that unfortunately interferes with unattended reboots since someone needs to be present to enter the encryption password. Here's how I added a timeout and made /home optional on that machine. I started by adding a one-minute timeout on the password prompt by adding timeout=60 in my /etc/crypttab: 
crypt  UUID=7e12c123-abcd-5555-8c40-900d1f8cc281  none  luks,timeout=60
then I made /home optional by adding nofail to the appropriate mount point in /etc/fstab: 
/dev/mapper/crypt  /home  ext4  nodev,noatime,nosuid,nofail  0  2
Before that, the password prompt would timeout but the system would be unable to boot since one of the required partitions had failed to mount. Now, to ensure that I don't accidentally re-create home directories for users when the system is mounted without a /home, I made the /home directory on the non-encrypted drive read-only: 
umount /home
cd /home
chmod a-w .
Finally, with all of this in place, I was now happy to configure the machine to automatically reboot after a kernel panic by putting the following in /etc/sysctl.d/local.conf: 
# Automatic reboot 10 seconds after a kernel panic
kernel.panic = 10
since I know that the machine will come back up just fine and that all services will be running. I simply won't be able to log into that machine as any other user than root until I manually unlock and mount /home.

Russ Allbery: Review: The Last Continent

Review: The Last Continent, by Terry Pratchett
Series: Discworld #22
Publisher: Harper
Copyright: 1998
Printing: May 2014
ISBN: 0-06-228019-8
Format: Mass market
Pages: 392
This is the 22nd Discworld novel and follows Interesting Times in internal continuity. Like some of the other Rincewind novels, it stands alone well enough that you could arguably start reading here, but I have no idea why you'd want to. When we last saw Rincewind, he was being magically yanked out of the Agatean Empire. The intent was to swap him with a cannon and land him back in Ankh-Morpork, but an unfortunate expansion of the spell to three targets instead of two meant that a kangaroo had a very bad day. Ever since then, Rincewind has been trying to survive the highly inhospitable land of FourEcks (XXXX), so called because no one in Ankh-Morpork knows where it is. The faculty at the Unseen University didn't care enough about Rincewind to bother finding him until the Librarian fell sick. He's feverish and miserable, but worse, he's lost control of his morphic function, which means that he's randomly turning into other things and is unable to take care of the books. When those books are magical, this is dangerous. One possible solution is to stabilize the Librarian's form with a spell, but to do that they need his real name. The only person who might know it is the former assistant librarian: Rincewind. I am increasingly convinced that one of the difficulties in getting people hooked on Discworld is that the series starts with two Rincewind books, and the Rincewind books just aren't very good. The fundamental problem is that Rincewind isn't a character, he's a gag. Discworld starts out as mostly gags, but then the characterization elsewhere gets deeper, the character interactions become more complex, and Pratchett adds more and more philosophy. That, not the humor, is what I think makes these books worth reading. But none of this applies to Rincewind. By this point, he's been the protagonist of six novels, and still the only thing I know about him is that he runs away from everything. Other than that, he's just sort of... there. In the better Rincewind novels, some of the gap is filled by Twoflower, the Luggage, Cohen the barbarian, the Librarian (who sadly is out of commission for most of this book), or the Unseen University faculty. But they're all supporting characters. Most of them are also built around a single (if better) gag. As a result, the Rincewind books tend more towards joke collections than the rest of Discworld. There isn't a philosophical or characterization through line to hold them together. The Last Continent is, as you might have guessed, a parody of Australia. And by that I mean it's a mash-up of Crocodile Dundee, Mad Max, The Adventures of Priscilla, Queen of the Desert, and every dad joke about Australia that you've heard. Pratchett loves movie references and I do not love movie references, so there's always part of his books that doesn't click for me, but this one was just Too Much. Yes, everything in Australia is poisonous. Yes, Australians talk funny. Oh look, there's another twist on a Crocodile Dundee quote. Yes, yes, that's a knife. Gah. The Rincewind sections were either confusing (there's some sort of drug-trip kangaroo god because reasons) or cliched and boring. Sometimes both. The second plot, following the Unseen University faculty in their inept attempts to locate Rincewind, is better. Their bickering is still a bit one-trick and works better in the background of stronger characters (such as Death and Susan), but Pratchett does make their oblivious overconfidence entertaining. It's enough to sustain half of the book, but not enough to make up for the annoyances of the Rincewind plot. To his credit, I think Pratchett was really trying to say something interesting in this novel about Discworld metaphysics. There are bits in the Australian plot that clearly are references to Aboriginal beliefs, which I didn't entirely follow but which I'm glad were in there. The Unseen University faculty showing up in the middle of a creation myth and completely misunderstanding it was a good scene. But the overall story annoyed me and failed to hold my interest. I don't feel qualified to comment on the Priscilla scenes, since I've never seen the movie and have only a vague understanding of its role in trans history. I'm not sure his twists on the story quite worked, but I'm glad that Pratchett is exploring gender; that wasn't as common when these books were written. Overall, though, this was forgettable and often annoying. There are a few great lines and a few memorable bits in any Pratchett book, including this one, but the Rincewind books just aren't... good. Not like the rest of the series, at least. I will be very happy to get back to the witches in the next book. Followed in publication order by Carpe Jugulum, and later thematically by The Last Hero. Rating: 5 out of 10

Louis-Philippe V ronneau: Extruded Schiit Stack

I've been a fan of the products manufactured by Schiit Audio for a while now. They are affordable (for high-end audio gear), sound great, are made in the USA1 and I think their industrial design looks great. I first started with one of their classic "Schiit Stack"2, but eventually upgraded to the Modi Multibit (I wanted the TOSLINK input), added a physical EQ (the Loki) and eventually got a Sys when I bought a Vidar speaker amp. The original Schiit Stack being 2 devices high was pretty manageable as-is. With my current 4-high stack though, things became unstable and I had to resort to finding a way to bolt them together. Mooching from a friend with a 3D printer, I printed this clever mount from Thingiverse. It worked well enough, but was somewhat imperfect for multiple reasons:
  1. The plastic tabs had a tendency of breaking in two when the screws where tight enough for the stack to feel solid.
  2. The plastic wasn't really rigid enough to support the 4 devices properly and the stack, being back-heavy from the cables, was unstable and tipped over easily.
  3. Due to the plastic tabs being fragile, it was pretty much impossible to disassemble the stack.
This last issue was what killed this solution for me. When I tried to replace my Modi 2 by the Modi Multibit, the mount pretty much crumbled away. Sadly, my friend warped a bunch of pieces on his 3D printer while trying to print ABS and I couldn't have him print me replacement parts either. After a while, I grew tired of having these four devices laying around my desk and wasting valuable space. I had tasted the 4-stack and knew how better things could be! That's when I realised the solution was to ditch 3D printing altogether, use aluminum framing extrusions and build my own stack out of metal. The 4 different Schiit devices with the hardware needed to build the extruded frame This was my first time working with aluminium frame extrusions and I had tons of fun! I specced the first version using 10mm x 10mm rails from McMaster-Carr, but discovered they do not ship to residential addresses in Canada... After looking at local options, I then decided to use 15mm x 15mm rails from Misumi. I went with this option since the rails are still small enough not to be an eyesore, but also because this system uses M3 screws, which the Schiit mini series also uses, making assembly much easier. I choose to make the assembled stack quite a bit taller than the previous one made with 3D printed plastic, as I found the headphone amp got pretty hot during the summer and I wanted to provide better airflow. If you are interested in replicating this stack, here are the parts I used, all from Misumi: I didn't order any since I had some already, but you'll also need M3 screws, namely: You can also cheap out and use only M3-10 screws (as I did), but you'll have to use the extra nuts you ordered as spacers. The assembled stack, complete with my lucky cat For the curious ones, the cabling is done this way: 
                            
                                                              
          Magni (hp amp)       Vidar (sp amp)  
                                                              
                            
                                      
                            
                                             
                              Sys (switch)   
                                                      
                                     
                                                      
                                                      
                            
                                                              
                 Modi (DAC)        Loki (EQ)
The Vidar is not part of the actual stack, as it's a 600W amp that weights 10kg :D. The last thing I think I want to change in this setup is the cables. The ones I have are too long for the stack. Shorter ones would reduce the wasted space in the back and make the whole thing more elegant.
  1. As in, designed, manufactured and assembled in the USA, from parts, transformers and boards made in the USA. I find this pretty impressive.
  2. A USB DAC and a headphone amp you can stack one of top of the other.

28 October 2022

Antoine Beaupr : Debating VPN options

In my home lab(s), I have a handful of machines spread around a few points of presence, with mostly residential/commercial cable/DSL uplinks, which means, generally, NAT. This makes monitoring those devices kind of impossible. While I do punch holes for SSH, using jump hosts gets old quick, so I'm considering adding a virtual private network (a "VPN", not a VPN service) so that all machines can be reachable from everywhere. I see three ways this can work:
  1. a home-made Wireguard VPN, deployed with Puppet
  2. a Wireguard VPN overlay, with Tailscale or equivalent
  3. IPv6, native or with tunnels
So which one will it be?

Wireguard Puppet modules As is (unfortunately) typical with Puppet, I found multiple different modules to talk with Wireguard.
module score downloads release stars watch forks license docs contrib issue PR notes
halyard 3.1 1,807 2022-10-14 0 0 0 MIT no requires firewall and Configvault_Write modules?
voxpupuli 5.0 4,201 2022-10-01 2 23 7 AGPLv3 good 1/9 1/4 1/61 optionnally configures ferm, uses systemd-networkd, recommends systemd module with manage_systemd to true, purges unknown keys
abaranov 4.7 17,017 2021-08-20 9 3 38 MIT okay 1/17 4/7 4/28 requires pre-generated private keys
arrnorets 3.1 16,646 2020-12-28 1 2 1 Apache-2 okay 1 0 0 requires pre-generated private keys?
The voxpupuli module seems to be the most promising. The abaranov module is more popular and has more contributors, but it has more open issues and PRs. More critically, the voxpupuli module was written after the abaranov author didn't respond to a PR from the voxpupuli author trying to add more automation (namely private key management). It looks like setting up a wireguard network would be as simple as this on node A: 
wireguard::interface   'wg0':
  source_addresses => ['2003:4f8:c17:4cf::1', '149.9.255.4'],
  public_key       => $facts['wireguard_pubkeys']['nodeB'],
  endpoint         => 'nodeB.example.com:53668',
  addresses        => [ 'Address' => '192.168.123.6/30', , 'Address' => 'fe80::beef:1/64' ,],
This configuration come from this pull request I sent to the module to document how to use that fact. Note that the addresses used here are examples that shouldn't be reused and do not confirm to RFC5737 ("IPv4 Address Blocks Reserved for Documentation", 192.0.2.0/24 (TEST-NET-1), 198.51.100.0/24 (TEST-NET-2), and 203.0.113.0/24 (TEST-NET-3)) or RFC3849 ("IPv6 Address Prefix Reserved for Documentation", 2001:DB8::/32), but that's another story. (To avoid boostrapping problems, the resubmit-facts configuration could be used so that other nodes facts are more immediately available.) One problem with the above approach is that you explicitly need to take care of routing, network topology, and addressing. This can get complicated quickly, especially if you have lots of devices, behind NAT, in multiple locations (which is basically my life at home, unfortunately). Concretely, basic Wireguard only support one peer behind NAT. There are some workarounds for this, but they generally imply a relay server of some sort, or some custom registry, it's kind of a mess. And this is where overlay networks like Tailscale come in.

Tailscale Tailscale is basically designed to deal with this problem. It's not fully opensource, but pretty close, and they have an interesting philosophy behind that. The client is opensource, and there is an opensource version of the server side, called headscale. They have recently (late 2022) hired the main headscale developer while promising to keep supporting it, which is pretty amazing. Tailscale provides an overlay network based on Wireguard, where each peer basically has a peer-to-peer encrypted connexion, with automatic key rotation. They also ship a multitude of applications and features on top of that like file sharing, keyless SSH access, and so on. The authentication layer is based on an existing SSO provider, you don't just register with Tailscale with new account, you login with Google, Microsoft, or GitHub (which, really, is still Microsoft). The Headscale server ships with many features out of that:
  • Full "base" support of Tailscale's features
  • Configurable DNS
    • Split DNS
    • MagicDNS (each user gets a name)
  • Node registration
    • Single-Sign-On (via Open ID Connect)
    • Pre authenticated key
  • Taildrop (File Sharing)
  • Access control lists
  • Support for multiple IP ranges in the tailnet
  • Dual stack (IPv4 and IPv6)
  • Routing advertising (including exit nodes)
  • Ephemeral nodes
  • Embedded DERP server (AKA NAT-to-NAT traversal)
Neither project (client or server) is in Debian (RFP 972439 for the client, none filed yet for the server), which makes deploying this for my use case rather problematic. Their install instructions are basically a curl bash but they also provide packages for various platforms. Their Debian install instructions are surprisingly good, and check most of the third party checklist we're trying to establish. (It's missing a pin.) There's also a Puppet module for tailscale, naturally. What I find a little disturbing with Tailscale is that you not only need to trust Tailscale with authorizing your devices, you also basically delegate that trust also to the SSO provider. So, in my case, GitHub (or anyone who compromises my account there) can penetrate the VPN. A little scary. Tailscale is also kind of an "all or nothing" thing. They have MagicDNS, file transfers, all sorts of things, but those things require you to hook up your resolver with Tailscale. In fact, Tailscale kind of assumes you will use their nameservers, and have suffered great lengths to figure out how to do that. And naturally, here, it doesn't seem to work reliably; my resolv.conf somehow gets replaced and the magic resolution of the ts.net domain fails. (I wonder why we can't opt in to just publicly resolve the ts.net domain. I don't care if someone can enumerate the private IP addreses or machines in use in my VPN, at least I don't care as much as fighting with resolv.conf everywhere.) Because I mostly have access to the routers on the networks I'm on, I don't think I'll be using tailscale in the long term. But it's pretty impressive stuff: in the time it took me to even review the Puppet modules to configure Wireguard (which is what I'll probably end up doing), I was up and running with Tailscale (but with a broken DNS, naturally). (And yes, basic Wireguard won't bring me DNS either, but at least I won't have to trust Tailscale's Debian packages, and Tailscale, and Microsoft, and GitHub with this thing.)

IPv6 IPv6 is actually what is supposed to solve this. Not NAT port forwarding crap, just real IPs everywhere. The problem is: even though IPv6 adoption is still growing, it's kind of reaching a plateau at around 40% world-wide, with Canada lagging behind at 34%. It doesn't help that major ISPs in Canada (e.g. Bell Canada, Videotron) don't care at all about IPv6 (e.g. Videotron in beta since 2011). So we can't rely on those companies to do the right thing here. The typical solution here is often to use a tunnel like HE's tunnelbroker.net. It's kind of tricky to configure, but once it's done, it works. You get end-to-end connectivity as long as everyone on the network is on IPv6. And that's really where the problem lies here; the second one of your nodes can't setup such a tunnel, you're kind of stuck and that tool completely breaks down. IPv6 tunnels also don't give you the kind of security a VPN provides as well, naturally. The other downside of a tunnel is you don't really get peer-to-peer connectivity: you go through the tunnel. So you can expect higher latencies and possibly lower bandwidth as well. Also, HE.net doesn't currently charge for this service (and they've been doing this for a long time), but this could change in the future (just like Tailscale, that said). Concretely, the latency difference is rather minimal, Google: 
--- ipv6.l.google.com ping statistics ---
10 packets transmitted, 10 received, 0,00% packet loss, time 136,8ms
RTT[ms]: min = 13, median = 14, p(90) = 14, max = 15
--- google.com ping statistics ---
10 packets transmitted, 10 received, 0,00% packet loss, time 136,0ms
RTT[ms]: min = 13, median = 13, p(90) = 14, max = 14
In the case of GitHub, latency is actually lower, interestingly: 
--- ipv6.github.com ping statistics ---
10 packets transmitted, 10 received, 0,00% packet loss, time 134,6ms
RTT[ms]: min = 13, median = 13, p(90) = 14, max = 14
--- github.com ping statistics ---
10 packets transmitted, 10 received, 0,00% packet loss, time 293,1ms
RTT[ms]: min = 29, median = 29, p(90) = 29, max = 30
That is because HE.net peers directly with my ISP and Fastly (which is behind GitHub.com's IPv6, apparently?), so it's only 6 hops away. While over IPv4, the ping goes over New York, before landing AWS's Ashburn, Virginia datacenters, for a whopping 13 hops... I managed setup a HE.net tunnel at home, because I also need IPv6 for other reasons (namely debugging at work). My first attempt at setting this up in the office failed, but now that I found the openwrt.org guide, it worked... for a while, and I was able to produce the above, encouraging, mini benchmarks. Unfortunately, a few minutes later, IPv6 just went down again. And the problem with that is that many programs (and especially OpenSSH) do not respect the Happy Eyeballs protocol (RFC 8305), which means various mysterious "hangs" at random times on random applications. It's kind of a terrible user experience, on top of breaking the one thing it's supposed to do, of course, which is to give me transparent access to all the nodes I maintain. Even worse, it would still be a problem for other remote nodes I might setup where I might not have acess to the router to setup the tunnel. It's also not absolutely clear what happens if you setup the same tunnel in two places... Presumably, something is smart enough to distribute only a part of the /48 block selectively, but I don't really feel like going that far, considering how flaky the setup is already.

Other options If this post sounds a little biased towards IPv6 and Wireguard, it's because it is. I would like everyone to migrate to IPv6 already, and Wireguard seems like a simple and sound system. I'm aware of many other options to make VPNs. So before anyone jumps in and says "but what about...", do know that I have personnally experimented with:
  • tinc: nice, automatic meshing, used for the Montreal mesh, serious design flaws in the crypto that make it generally unsafe to use; supposedly, v1.1 (or 2.0?) will fix this, but that's been promised for over a decade by now
  • ipsec, specifically strongswan: hard to configure (especially configure correctly!), harder even to debug, otherwise really nice because transparent (e.g. no need for special subnets), used at work, but also considering a replacement there because it's a major barrier to entry to train new staff
  • OpenVPN: mostly used as a client for [VPN service][]s like Riseup VPN or Mullvad, mostly relevant for client-server configurations, not really peer-to-peer, shared secrets or TLS, kind of an hassle to maintain, see also SoftEther for an alternative implementation
All of those solutions have significant problems and I do not wish to use any of those for this project. Also note that Tailscale is only one of many projects laid over Wireguard to do that kind of thing, see this LWN review for others (basically NetbBird, Firezone, and Netmaker).

Future work Those are options that came up after writing this post, and might warrant further examination in the future.
  • Meshbird, a "distributed private networking" with little information about how it actually works other than "encrypted with strong AES-256"
  • Nebula, "A scalable overlay networking tool with a focus on performance, simplicity and security", written by Slack people to replace IPsec, docs, runs as an overlay for Slack's 50k node network, only packaged in Debian experimental, lagging behind upstream (1.4.0, from May 2021 vs upstream's 1.6.1 from September 2022), requires a central CA, Golang, I'm in "wait and see" mode for now
  • n2n: "layer two VPN", seems packaged in Debian but inactive
  • ouroboros: "peer-to-peer packet network prototype", sounds and seems complicated
  • QuickTUN is interesting because it's just a small wrapper around NaCL, and it's in Debian... but maybe too obscure for my own good
  • unetd: Wireguard-based full mesh networking from OpenWRT, not in Debian
  • vpncloud: "high performance peer-to-peer mesh VPN over UDP supporting strong encryption, NAT traversal and a simple configuration", sounds interesting, not in Debian
  • Yggdrasil: actually a pretty good match for my use case, but I didn't think of it when starting the experiments here; packaged in Debian, with the Golang version planned, Puppet module; major caveat: nodes exposed publicly inside the global mesh unless configured otherwise (firewall suggested), requires port forwards, alpha status

Conclusion Right now, I'm going to deploy Wireguard tunnels with Puppet. It seems like kind of a pain in the back, but it's something I will be able to reuse for work, possibly completely replacing strongswan. I have another Puppet module for IPsec which I was planning to publish, but now I'm thinking I should just abort that and replace everything with Wireguard, assuming we still need VPNs at work in the future. (I have a number of reasons to believe we might not need any in the near future anyways...)

Shirish Agarwal: Shantaram, The Pyramid, Japan s Hikikomori & Backpack

Shantaram I know I have been quite behind in review of books but then that s life. First up is actually not as much as a shocker but somewhat of a pleasant surprise. So, a bit of background before I share the news. If you have been living under a rock, then about 10-12 years ago a book called Shantaram was released. While the book is said to have been released in 2003/4 I got it in my hand around 2008/09 or somewhere around that. The book is like a good meal, a buffet. To share the synopsis, Lin a 20 something Australian guy gets involved with a girl, she encourages him to get into heroin, he becomes a heroin user. And drugs, especially hard drugs need constant replenishment, it is a chemical thing. So, to fund those cravings, he starts to steal, rising to rob a bank and while getting away shoots a cop who becomes dead. Now either he surrenders or is caught is unclear, but he is tortured in the jail. So one day, he escapes from prison, lands up at home of somebody who owes him a favor, gets some money, gets a fake passport and lands up in Mumbai/Bombay as it was then known. This is from where the actual story starts. And how a 6 foot something Australian guy relying on his street smartness and know how the transformation happens from Lin to Shantaram. Now what I have shared is perhaps just 5% of the synopsis, as shared the real story starts here. Now the good news, last week 4 episodes of Shantaram were screened by Apple TV. Interestingly, I have seen quite a number people turning up to buy or get this book and also sharing it on Goodreads. Now there seems to have been some differences from the book to TV. Now I m relying on 10-12 year back memory but IIRC Khaderbhai, one of the main characters who sort of takes Lin/Shantaram under his wing is an Indian. In the series, he is a western or at least looks western/Middle Eastern to me. Also, they have tried to reproduce 1980s in Mumbai/Bombay but dunno how accurate they were My impression of that city from couple of visits at that point in time where they were still more tongas (horse-ridden carriages), an occasional two wheelers and not many three wheelers. Although, it was one of the more turbulent times as lot of agitation for worker rights were happening around that time and a lot of industrial action. Later that led to lot of closure of manufacturing in Bombay and it became more commercial. It would be interesting to know whether they shot it in actual India or just made a set somewhere in Australia, where it possibly might have been shot. The chawl of the book needs a bit of arid land and Australia has lots of it. It is also interesting as this was a project that had who s who interested in it for a long time but somehow none of them was able to bring the project to fruition, the project seems to largely have an Australian cast as well as second generations of Indians growing in Australia. To take names, Amitabh Bacchan, Johnny Depp, Russel Crowe each of them wanted to make it into a feature film. In retrospect, it is good it was not into a movie, otherwise they would have to cut a lot of material and that perhaps wouldn t have been sufficient. Making it into a web series made sure they could have it in multiple seasons if people like it. There is a lot between now and 12 episodes to even guess till where it would leave you then. So, if you have not read the book and have some holidays coming up, can recommend it. The writing IIRC is easy and just flows. There is a bit of action but much more nuance in the book while in the web series they are naturally more about action. There is also quite a bit of philosophy between him and Kaderbhai and while the series touches upon it, it doesn t do justice but then again it is being commercially made. Read the book, see the series and share your thoughts on what you think. It is possible that the series might go up or down but am sharing from where I see it, may do another at the end of the season, depending on where they leave it and my impressions. Update A slight update from the last blog post. Seems Rishi Sunak seems would be made PM of UK. With Hunt as chancellor and Rishi Sunak, Austerity 2.0 seems complete. There have been numerous articles which share how austerity gives rises to fascism and vice-versa. History gives lot of lessons about the same. In Germany, when the economy was not good, it was all blamed on the Jews for number of years. This was the reason for rise of Hitler, and while it did go up by a bit, propaganda by him and his loyalists did the rest. And we know and have read about the Holocaust. Today quite a few Germans deny it or deny parts of it but that s how misinformation spreads. Also Hitler is looked now more as an aberration rather than something to do with the German soul. I am not gonna talk more as there is still lots to share and that actually perhaps requires its own blog post to do justice for the same.

The Pyramid by Henning Mankell I had actually wanted to review this book but then the bomb called Shantaram appeared and I had to post it above. I had read two-three books before it, but most of them were about multiple beheadings and serial killers. Enough to put anybody into depression. I do not know if modern crime needs to show crime and desperation of and to such a level. Why I and most loved and continue to love Sherlock Holmes as most stories were not about gross violence but rather a homage to the art of deduction, which pretty much seems to be missing in modern crime thrillers rather than grotesque stuff. In that, like a sort of fresh air I read/am reading the Pyramid by Henning Mankell. The book is about a character made by Monsieur Henning Mankell named Kurt Wallender. I am aware of the series called Wallender but haven t yet seen it. The book starts with Wallender as a beat cop around age 20 and on his first case. He is ambitious, wants to become a detective and has a narrow escape with death. I wouldn t go much into it as it basically gives you an idea of the character and how he thinks and what he does. He is more intuitive by nature and somewhat of a loner. Probably most detectives IRL are, dunno, have no clue. At least in the literary world it makes sense, in real world think there would be much irony for sure. This is speculation on my part, who knows. Back to the book though. The book has 5 stories a sort of prequel one could say but also not entirely true. The first case starts when he is a beat cop in 1969 and he is just a beat cop. It is a kind of a prequel and a kind of an anthology as he covers from the first case to the 1990s where he is ending his career sort of. Before I start sharing about the stories in the book, I found the foreword also quite interesting. It asks questions about the interplay of the role of welfare state and the Swedish democracy. Incidentally did watch couple of videos about a sort of mixed sort of political representation that happens in Sweden. It uses what is known as proportional representation. Ironically, Sweden made a turn to the far right this election season. The book was originally in Swedish and were translated to English by Ebba Segerberg and Laurie Thompson. While all the stories are interesting, would share the last three or at least ask the questions of intrigue. Of course, to answer them you would need to read the book So the last three stories I found the most intriguing. The first one is titled Man on the Beach. Apparently, a gentleman goes to one of the beaches, a sort of lonely beach, hails a taxi and while returning suddenly dies. The Taxi driver showing good presence of mind takes it to hospital where the gentleman is declared dead on arrival. Unlike in India, he doesn t run away but goes to the cafeteria and waits there for the cops to arrive and take his statement. Now the man is in his early 40s and looks to be fit. Upon searching his pockets he is found to relatively well-off and later it turns out he owns a couple of shops. So then here are the questions ? What was the man doing on a beach, in summer that beach is somewhat popular but other times not so much, so what was he doing there? How did he die, was it a simple heart attack or something more? If he had been drugged or something then when and how? These and more questions can be answered by reading the story Man on the Beach . 2. The death of a photographer Apparently, Kurt lives in a small town where almost all the residents have been served one way or the other by the town photographer. The man was polite and had worked for something like 40 odd years before he is killed/murdered. Apparently, he is murdered late at night. So here come the questions a. The shop doesn t even stock any cameras and his cash box has cash. Further investigation reveals it is approximate to his average takeout for the day. So if it s not for cash, then what is the motive ? b. The body was discovered by his cleaning staff who has worked for almost 20 years, 3 days a week. She has her own set of keys to come and clean the office? Did she give the keys to someone, if yes why? c. Even after investigation, there is no scandal about the man, no other woman or any vices like gambling etc. that could rack up loans. Also, nobody seems to know him and yet take him for granted till he dies. The whole thing appears to be quite strange. Again, the answers lie in the book. 3. The Pyramid Kurt is sleeping one night when the telephone rings. The scene starts with a Piper Cherokee, a single piston aircraft flying low and dropping something somewhere or getting somebody from/on the coast of Sweden. It turns and after a while crashes. Kurt is called to investigate it. Turns out, the plane was supposed to be destroyed. On crash, both the pilot and the passenger are into pieces so only dental records can prove who they are. Same day or a day or two later, two seemingly ordinary somewhat elderly women, spinsters, by all accounts, live above the shop where they sell buttons and all kinds of sewing needs of the town. They seem middle-class. Later the charred bodies of the two sisters are found :(. So here come the questions a.Did the plane drop something or pick something somebody up ? The Cherokee is a small plane so any plane field or something it could have landed up or if a place was somehow marked then could be dropped or picked up without actually landing. b. The firefighter suspects arson started at multiple places with the use of petrol? The question is why would somebody wanna do that? The sisters don t seem to be wealthy and practically everybody has bought stuff from them. They weren t popular but weren t also unpopular. c. Are the two crimes connected or unconnected? If connected, then how? d. Most important question, why the title Pyramid is given to the story. Why does the author share the name Pyramid. Does he mean the same or the original thing? He could have named it triangle. Again, answers to all the above can be found in the book. One thing I also became very aware of during reading the book that it is difficult to understand people s behavior and what they do. And this is without even any criminality involved in. Let s say for e.g. I die in some mysterious circumstances, the possibility of the police finding my actions in last days would be limited and this is when I have hearing loss. And this probably is more to do with how our minds are wired. And most people I know are much more privacy conscious/aware than I am.

Japan s Hikikomori Japan has been a curious country. It was more or less a colonizer and somewhat of a feared power till it dragged the U.S. unnecessarily in World War 2. The result of the two atom bombs and the restitution meant that Japan had to build again from the ground up. It is also in a seismically unstable place as they have frequent earthquakes although the buildings are hardened/balanced to make sure that vibrations don t tear buildings apart. Had seen years ago on Natgeo a documentary that explains all that. Apart from that, Japan was helped by the Americans and there was good kinship between them till the 1980s till it signed the Plaza Accord which enhanced asset price bubbles that eventually burst. Something from which they are smarting even today. Japan has a constitutional monarchy. A somewhat history lesson or why it exists even today can be found here. Asset price bubbles of the 1980s, more than 50 percent of the population on zero hour contracts and the rest tend to suffer from overwork. There is a term called Karoshi that explains all. An Indian pig-pen would be two, two and a half times larger than a typical Japanese home. Most Japanese live in micro-apartments called konbachiku . All of the above stresses meant that lately many young Japanese people have become Hikikomori. Bloomberg featured about the same a couple of years back. I came to know about it as many Indians are given the idea of Japan being a successful country without knowing the ills and issues it faces. Even in that most women get the wrong end of the short stick i.e. even it they manage to find jobs, it would be most back-breaking menial work. The employment statistics of Japan s internal ministry tells its own story.

If you look at the data above, it seems that the between 2002 and 2019, the share of zero hour contracts has increased while regular work has decreased. This also means that those on the bottom of the ladder can no longer afford a home. There is and was a viral video called Lost in Manboo that went viral few years ago. It is a perfect set of storms. Add to that the Fukushima nuclear incident about which I had shared a few years ago. While the workers are blamed but all design decisions are taken by the management. And as was shown in numerous movies, documentaries etc. Interestingly, and somewhat ironically, the line workers knew the correct things to do and correct decisions to take unlike the management. The shut-ins story is almost a decade or two decades old. It is similar story in South Korea but not as depressive as the in Japan. It is somewhat depressive story but needed to be shared. The stories shared in the bloomberg article makes your heart ache

Backpacks In and around 2015, I had bought a Targus backpack, very much similar to the Targus TSB194US-70 Motor 16-inch Backpack. That bag has given me a lot of comfort over the years but now has become frayed the zip sometimes work and sometimes doesn t. Unlike those days there are a bunch of companies now operating in India. There are eight different companies that I came to know about, Aircase, Harrisons Sirius, HP Oddyssey, Mokobara, Artic Hunter, Dell Pro Hybrid, Dell Roller Backpack and lastly the Decathlon Quechua Hiking backpack 32L NH Escape 500 . Now of all the above, two backpacks seem the best, the first one is Harrisons Sirius, with 45L capacity, I don t think I would need another bag at all. The runner-up is the Decathlon Quecha Hiking Backpack 32L. One of the better things in all the bags is that all have hidden pockets for easy taking in and out of passport while having being ant-theft. I do not have to stress how stressful it is to take out the passport and put it back in. Almost all the vendors have made sure that it is not a stress point anymore. The good thing about the Quecha is that they are giving 10 years warranty, the point to be asked is if that is does the warranty cover the zip. Zips are the first thing that goes out in bags.That actually has what happened to my current bag. Decathlon has a store in Wakad, Pune while I have reached out to the gentleman in charge of Harrisons India to see if they have a reseller in Pune. So hopefully, in next one week I should have a backpack that isn t spilling with things all over the place, whichever I m able to figure out.

27 October 2022

Emmanuel Kasper: Convert a root filesystem to a bootable disk image

The year is 2022, and it is still that complicated to install GRUB2 externally onto a disk image. But using the wonders of libguestfs, you can create a bootable diskimage using a qemu VM abstraction very easily. The steps here imply we want to create a disk with a single partition containing the root filesystem. Create an empty disk image, partition it 
$ truncate --size 40G target.img
$ virt-format --add target.img --partition=mbr --filesystem=ext4
copy the root file system into a partition 
cd path/to/root/fs
sudo tar --numeric-owner -cvf - .   guestfish --rw --add ../target.img --mount /dev/sda1:/ -- tar-in - /
install grub using guestfish 
$ guestfish --add target.img --inspector
and in the guestfish prompt: 
>> command 'grub-install /dev/sda'
>> command 'update-grub'
# also make sure init can mount our root partition
>> write /etc/fstab '/dev/sda1 / ext4 defaults 0 1'
>> exit
test boot the disk image 
$ kvm -m 1024 -drive file=target.img,format=raw

26 October 2022

Scarlett Gately Moore: KDE Gear snaps round 3!

While trying to stay warm in our first snow of the year, I got several apps tested and released in round 3 of https://www.scarlettgatelymoore.dev/kde-gear-snaps-round-2/ ! All of these are being ( re ) tested on both arm64 and amd64. Hence, the release is going slower than it will in the future. Thank you for your patience. More rounds of bug gardening has been done, if you find bugs, or know one of your bugs is related to it being a snap, please assign to me. https://bugs.kde.org Please consider a donation so I may bring many more apps your way. https://www.patreon.com/sgmoore I am working on one time donations.

Next.

Previous.