Search Results: "apollock"

20 August 2008

Andrew Pollock: [life] Snail mail spam control

John Goerzen writes an entertaining piece on his ongoing battle with Dell to stop receiving advertising from them. (The comments on the post are rather interesting in themselves) I say "good on you, John" for taking a stand. Now that I know of these prohibitory orders, I might try them out on Pennysaver so I stop getting their stupid catalogs. A while ago our junk mail situation got completely out of hand, and at the time I found 41pounds.org, which for a very reasonable $41, did a very good job of stopping pretty much all of the addressed spam. About the only thing we get now is a small wad of catalogs that USPS delivers to every resident, which isn't explicitly address to us by name, just the apartment number. That and these stupid "checks" from Washington Mutual that we can use against our credit card account with them. They give me the screaming heebie jeebies, and I shred them on sight, because I've seen how insecure cheques are.

16 August 2008

Andrew Pollock: [debian] On hypothetical alternative freeze policies

I, for one, like Lucas' idea.

Andrew Pollock: [debian] Crazy idea: Debian spends some money on snapshot.debian.net

Eddy Petri or wrote about doing some sort of user supported thing for snapshot.debian.net, which I was saddened to learn from his post had reached capacity in May. I think rather than relying on the generosity of random users, the Debian Project should just throw some money at the problem. It's got more than enough. Disk is cheap. I'd think that with a few fully loaded SR2421's and an Ethernet switch, you'd be laughing.

11 August 2008

Andrew Pollock: [tech] Amanda, tar, and Debian Etch

Somehow (thankfully) I've managed to avoid much to do with backups in my career as a sysadmin. It's always been someone else's responsibility, that's been fine by me. For some reason, I've been more paranoid lately, and with all of the crazy shit I've been doing to daedalus, I wanted to have a backup, just in case. Maybe it's old age. So a while ago, I went and had a stab at setting up Amanda, since pretty much whenever I've been near a bunch of sysadmins and someone asks for a recommendation on what to use, "Amanda" is the answer. I jumped through all of the hoops to have caesar be the backup server, and daedalus be a client. I set up a virtual tape library on caesar. I think I then lost interest for a while, until I was going to upgrade daedalus to Etch, and I wanted to have a successful backup first. I had the holding disk and the virtual tape library on the same filesystem, and it was nice and big, yet I kept running into this error about insufficient space. "dump larger than available tape space" was the complaint. I tried making the filesystem larger. I tried adding more virtual tapes. Nothing would placate it. After another round of Googling, I discovered that the problem was because Amanda would only only span a maximum of four tapes, and the filesystem I was trying to backup was way bigger than that. So I told it to use up to 10 tapes in a run (runtapes 10) and then lo and behold, I got a successful backup out of the sucker. So that was mostly all good, but occasionally a backup would fail, or fail once an succeed on a retry, because tar exited non-zero, with "file changed as we read it". I did some more Googling tonight, and discovered this Amanda/GNU tar compatibility matrix. It turns out that the version of tar and the version of Amanda shipped in Debian Etch aren't compatible with each other. Suck! Oh wait, they're maintained by the same person! Argh! So aside from the occasional backup failing due to a file being written to at the time of the backup, it's working fairly well. My DSL link isn't the fastest, so a full backup of daedalus can take over 48 hours to complete, but I'm not paying for inbound bandwidth at home, so it's not a big deal. I initially thought that to do a restore, I'd have to run amrecover on daedalus, and then I'd have to stream the contents of multiple tapes back to it at the appalling uplink speed of my DSL connection, but fortunately amrecover has a sethost command, so I can restore just the files I need on another machine local to caesar at LAN speeds, and just transfer the files I restore to daedalus. So that's all pretty neat. I quite like the FTP-client-like interface of amrecover. It's cool how you can use setdate and just go back in time. I had the opportunity to do a test restore recently when one of my users nuked his Maildir directory by accident. Unfortunately I was unable to recover the needed files in that case, and I'm not entirely sure what the problem was, so there's some more testing to be done before I'll be fully confident that I can rely on it. There's a lot of configuration options available as well, and I think I got the config I'm currently using from a template or something, so there's probably a fair bit of tweaking I can do. Oh, and I'm constantly reminded of this quote from a previous life, where technical details bubbled too far up non-technical management:
"Who is Amanda, and what is her role in the backups?"

Andrew Pollock: [debian] QoTD: You know the bar's been lowered when you're excited about being able to install packages like less

Spoken by a co-worker, fighting with nano, when all he wanted was less to browse the installer log to diagnose why an installation with d-i failed.

Andrew Pollock: [life] "San Francisco's Bicycle Program strives to promote safe and secure bicycle parking to complement the growing bicycle network."

One of the things I've wanted to do since I discovered that Caltrain (and the VTA light rail) are well-equipped to take bikes, was go for an outing further afield on our bikes using one of these methods of public transport. So we finally got around to doing that yesterday, when we threw our bikes on the train, and BART, and went for a ride around Golden Gate Park. The grand plan was to bike to the Caltrain station, take Caltrain to Milbrae, BART to downtown, then bike from there to the park, and then catch an afternoon showing of The Dark Knight in IMAX at the Metreon, then bike back to the BART station, and reverse the whole thing. We had a minor hiccup in that Sarah had just gotten her bike back from a friend she'd lent it to, and as we were about to head to the Caltrain station, discovered that it a broken spoke. She managed to extract the spoke and we got to the station with a couple of minutes to spare. It turned out that Mike's Bikes were fairly close to the BART station, so we stopped in there on way to the park, and they were able to fix the spoke on the spot in 20 minutes for the princely sum of $21. Unfortunately the weather was pretty typical summer San Francisco weather, and it was quite cold and foggy. It didn't get any better as we got closer to the park (and the coast), so the ride around the park itself wasn't terribly exciting. The hill we had to get over (Hayes Street) was a bit of a slog. On the way back we saw the filming of a Japanese Nissan commercial (complete with right-hand drive car). I'd done some research into bike parking beforehand, and found that all of the parking garages are required to provide bicycle parking, and there's a pretty comprehensive list. So we parked our bikes in the Moscone Center garage, which was about a block from the Metreon, and went and caught the movie. The Dark Knight was really good. Very... dark. I really liked Ledger's portrayal of the Joker, much better than the original. I wonder if they'll just keep re-imagining the Batman movies over and over? There was pretty good continuity from Batman Begins. I thought the voice of Batman was a bit ridiculous though, at least initially. You kind of got used to it after a while. Anyway, we got back to the parking garage, and some lowlife had cut through the cable lock that we'd used to lock our bikes to each other and the rack, and stolen Sarah's bike! Annoyed does not begin to describe it. We trundled off the seven or eight blocks to the Hall of Justice to file a report. The SFPD weren't the least bit interested, really. The somewhat astounding thing was the bike racks were right behind the cashier's office, and there was video surveillance, and someone still managed to just walk right in, cut the cable, and walk out with the bike. We checked with the duty manager, and he reviewed the video footage and said they have footage of him walking in and then walking out with the bike. So hopefully the police will get hold that. We have to see if we can dredge up the purchase documentation for the bike and see if the serial number is recorded. I have no idea if we'll ever see the bike again. Our renter's insurance may cover it, but I suspect the deductible won't make it worth our while to make a claim. I have to say (and it's not like it's a recent discovery or anything) that the Bay Area's public transportation is a joke. It cost us $32 to go from Mountain View to Downtown San Francisco, via Caltrain and then BART. Our Prius takes about 11 gallons at worst, and with gas prices being at say $4.30 a gallon, that's about $47 a tank. We get about 400 miles out of a tank usually, and it's vaguely 40 miles from home to the guts of Golden Gate Park. So it's cheaper, and faster to drive to San Francisco than it is to use a bicycle and public transport. Particularly when you factor in the risk of theft of your bicycle. Our next bike lock will be one of those fancy Kevlar ones.

8 August 2008

Andrew Pollock: [tech] Moved

Well that happened, and uneventfully as well. Now if only IntaServe actually made it possible to update glue records, everything would be done... What is it with domain registrars and glue records? Are glue records that much of an edge case these days that no one bothers having an interface for managing them?

Andrew Pollock: [tech] Moving

I've found a new colo for my server, and Brent has kindly offered to do the remote hands work for me tomorrow. Fingers crossed it'll be fairly smooth...

27 July 2008

Andrew Pollock: [geek] This almost makes me wish I had a lawn that needed mowing. Almost.

I just became aware of this electric lawn mower via Woot! I never minded mowing the lawn, when it was a decent kind of yard (i.e. level). This mower looks like it'd take the fun to a whole new level: hardly any noise, and no smoke. I just need a yard now...

Andrew Pollock: [debian] Etch and a Half no workee for me

So hot off the back of my successful upgrade from Sarge to Etch, I thought I'd try the new "Etch and a Half" 2.6.24 kernel on daedalus Let's just say that it's times like these that I'm glad I got the serial-over-LAN thing working, and have remote power. I spent many hours today rebooting, trying to get to the bottom of why it wouldn't work. At first, I thought it was hanging, but then I realised it was dropping to a shell in the initramfs, and just not making that very obvious because the serial over LAN console is a bit crappy. Once I realised what was going on, I did some poking around. It seemed like udev wasn't getting started properly, so when it went to assemble the mirrors, that failed, because the component devices couldn't be found, then it freaked out because it couldn't mount the root filesystem. If I ran the relevant bits of /scripts/init-premount/udev by hand, the SCSI devices appeared, and I could manually assemble the mirror, and mount the root filesystem, which was handy, because I also discovered that you can boot with debug on the kernel command line, and it logs the initramfs run to /tmp/initramfs.debug. So that was a convenient way of preserving the log, because there seemed to be some characters in it that made inspecting it over the serial console difficult, and there was no way to get it off the machine from the initramfs environment. As far as I can determine, it's telling udev to start, but it certainly isn't still running when it bombs out to a shell after failing to mount the root filesystem. It's not immediately clear if there's something later on that is stopping it again. I've put the log here in case anyone's interested in looking at it. This was with the addition of "x" to the options of the shebang line of /usr/share/initramfs-tools/scripts/init-premount/udev so that I could see what was happening when /scripts/init-premount/udev ran. So I don't think the kernel itself is at fault, it's some sort of weird udev/initramfs interaction. Rather than further rebooting the tripe out of daedalus, I'll have to see if I can reproduce the problem on a less important machine locally to do further debugging. In the meantime, I'll have to stick with 2.6.18.

26 July 2008

Andrew Pollock: [tech] And we're back

Well that certainly wasn't worth the amount of angst I'd pre-allocated it. The upgrade went reasonably smoothly. Lowlights included: I'm certainly glad to be getting security updates again. Given how smoothly this went, I'll definitely not procrastinate on upgrading to Lenny when it releases.

Andrew Pollock: [tech] Upgrading daedalus to Etch

I'm starting a dreaded dist-upgrade of daedalus to Etch. I'm feeling very apprehensive about it, mainly because it's a good 14,000 km away from where I am. So apologies in advance if I flood any Planets due to the Blosxom upgrade doing something weird.

23 July 2008

Andrew Pollock: [life] Three years later

Today is our third wedding anniversary. We're marking the occasion by having a night at the Tickle Pink Inn, which I can vouch for as being very nice indeed. Unfortunately I ended up on-call this week, so my day off tomorrow isn't going to be as idyllic as it could have been. It's hard to believe we've only been married three years. I think we've already been through more than a lot of married couples would have to go through in a lifetime. Here's to at least another 50 years.

22 July 2008

Andrew Pollock: [tech] Looking for a new colo for my server

Unfortunately the free ride I've had for the last oh, gee, it must be something like four years, is coming to an end, so I have to find somewhere else to co-locate daedalus. Thanks Ben, you've saved me a packet! The first step is to figure out roughly how much traffic this thing is doing. Fortunately I've been running Argus on it for donkey's ages, so I should be able to figure that part out. Then I have to find a colo facility, (preferably in Brisbane, to avoid too much dicking about with moving the hardware), which won't charge me an arm and a leg for traffic. I'd prefer a flat monthly fee if possible. I also need to do some soul searching to decide if it's even worthwhile continuing to keep a physical server. I could move my mail to Google Apps, the photos to Picasa, but I kind of like my blog the way it is. I spent a non-trivial amount of money of daedalus about 3 years ago. It's a grunty server. It seems something of a shame to get rid of it, but I could potentially get a virtual server a lot more cheaply. The other option would be to try and generate sufficient income with it to cover its costs, but that has all sorts of tax implications, not to mention having to maintain a service level agreement. I really don't have the time or inclination to be doing all of that on the side. All stuff to think about. I have an amorphous month to move my server, so if anyone out there has suggestions on a colo provider in Brisbane, please drop me a line.

11 July 2008

Andrew Pollock: [life] She's unstoppable

Sarah just got her University results for semester 1. A distinction and a credit. That was starting the semester recovering from heart surgery. I'm very proud of her. She's doing three subjects next semester, and then she's finished first year. Pretty impressive progress really. I know I wouldn't stand a snowflake's chance in hell of doing well studying externally.

10 July 2008

Andrew Pollock: [tech] What is a good RAID reconstruction speed?

I'm in the process of converting my poor-man's SAN from IDE disks attached to the host by IDE-USB adapters to SATA disks attached by eSATA to SATA cables. I had some spare time tonight, so I thought I'd prepare the new RAID-10 on brutus. The intent is to then take the 4-port USB 2.0 card out of minotaur during a lull in recordings on Saturday, and put it in brutus as well, and do a giant pvmove, then put the 4-port eSATA card back in minotaur with the new disks attached. So I've got the 4-port eSATA card (a Silicon Image SiI 3124) in brutus, and I've built the RAID-10 with mdadm and it's busily reconstructing. I'm getting a reconstruction speed of around 60Mb per second (/proc/mdstat says 60254K/sec right now). I have no idea if that's good, bad, or indifferent. I do know that output of hdparm -tT is very nice by comparison:
apollock@brutus:~$ sudo hdparm -tT /dev/sda
/dev/sda:
 Timing cached reads:   260 MB in  2.01 seconds = 129.06 MB/sec
 Timing buffered disk reads:  222 MB in  3.01 seconds =  73.77 MB/sec
compared to
apollock@minotaur:~$ sudo hdparm -tT /dev/sda
/dev/sda:
 Timing cached reads:   162 MB in  2.02 seconds =  80.20 MB/sec
 Timing buffered disk reads:   48 MB in  3.04 seconds =  15.78 MB/sec
Those times are with a reconstruction running on brutus and a recording being written to the array on minotaur, so not the most controlled benchmarking environment. If I can sit on my hands long enough, I'll try some benchmarks from the MythTV machine, comparing the performance of the SAN before and after switching disks.

7 July 2008

Andrew Pollock: [tech] On tab completion in Python

Bastian Venthur has discovered how to enable tab completion in the standard Python interpreter. Good to know, but I'll still stick with ipython. It offers tab completion by default, and a lot more.

4 July 2008

Andrew Pollock: [life] It's official: I am married to the crazy cat lady

We currently have fifteen cats in the house. Only three of them are ours. When Sarah was contracting at Google, she started an informal, but company-endorsed trap/neuter/release program, because the basement of buildings 40-43 had a bit of a stray/feral cat infestation. Since she left Google, she's continued doing trappings on an on and off basis. She got a call a couple of days ago about a litter of kittens behind one of the other buildings, so she headed off last night with one of her crazy cat lady friends, and they proceeded to catch one of the kittens. She brought that home, and I went back with her, and the mother, which seems fairly feral, had been caught in a trap. The rest of the kittens seemed to have managed to secrete themselves away inside a generator behind the building. This morning, with some help from various facilities types, she managed to extract the other four kittens from the generator. So that put the mother plus five kittens in the spare room (in a large cage). At more or less the same time, Sarah got news of a litter of six kittens that need rescuing from down Gilroy way (some woman in a trailer park didn't seem to realise that boy cat + girl cat = kittens), so after grabbing the other four kittens this morning, she headed off to Gilroy to grab them. So the current plan is to get the mother cat of what we'll call the Google litter desexed on Tuesday and then release her back where she came from. Her five kittens are young enough (~8 weeks) that they can be socialised and won't be feral. The six Gilroy kittens are from non-feral parents, and are about 11 weeks old. All of the kittens are old enough to be desexed as well, but Sarah can't get them into Palo Alto Animal Services for desexing for a couple of weeks, so I suspect we're stuck with them in the interim, unless Sarah can find foster homes for them. To cap it off, we're going to Sacramento for the weekend, so we have to get one of our friends in the complex to take care of them all. The sad thing is the Gilroy Six seem to be in worse condition than the Google Five. They've got pretty poor coats, and seem to have some diarrhoea. Hopefully with a better diet they'll pick up. They're all very cute. The first one of the Google Five that Sarah caught last night is particularly cute. I expect Sarah will get some photos of them all up soon. If anyone is in the Bay Area is interested in a kitten, we've got them by the near-dozen.

3 July 2008

Andrew Pollock: [debian] Is autofs maintained upstream?

The current autofs package in Debian unstable has twenty-nine patches applied to it. The Ubuntu autofs package in Hardy Heron LTS has thirty patches applied to it (twenty-nine of them being the Debian ones). We discovered a bug today in Hardy's autofs, which will undoubtedly also affect Debian's autofs as well, which is apparently fixed in the likes of Red Hat Enterprise Linux for something like four years. So either Red Hat is doing a shoddy job of feeding patches back upstream, or upstream is doing a shoddy job of accepting them. The fact that Debian has to carry twenty-nine patches suggests something is similarly wrong in our camp. We've been finding various problems in autofs in our environment recently, and have accumulated maybe half a dozen patches or so in the last month. I'll have to make sure that they get fed back into Debian and Ubuntu at the very least. Given the nature of one of the bugs we found today, I have grave doubts about the code quality of autofs. I have to wonder if it's worth rewriting from scratch in Python or something.

2 July 2008

Andrew Pollock: [tech] On microblogging

Now that Tim Connors has taken the lid off it, I feel I should opine on the matter. Microblogging is just not terribly exciting for the reader. I don't see the point in aggregating it. I don't particularly find Mikal's blatherings terribly exciting reading. I don't find Stewart's twitterings much fun either. I think Facebook's status updates are a better way to poll such things, myself, and it's where I confine my "microblogging" to. I think one of the things that used to make reading blogs more worthwhile than reading mailing lists was the high signal-to-noise ratio. Back in the day, there was a bit of a "cost" or barrier to entry for writing a blog post. For most blogs, it wasn't as easy as writing an email. I still hand-craft my blog posts in HTML, so for me to write something, I have to have enough inclination to fire up an editor and write the words. Often I run out of motivation half way through a blog post, and just end up scrapping it. I'm sitting on a few more lengthy posts that I haven't managed to summon up the energy to write at all yet. I think some of the "push-button publishing" out there is rapidly commoditising blogging, to the point where it becomes as easy or easier than writing an email. And we all know what happened to email... Blogging for me is essentially a journal. I like to use it to refer back to. To share what I'm doing with others. Occasionally it can be cathartic. Often it's a substitute for going to work and telling my co-workers about what crazy geeky thing I've done in my spare time. Mostly it's for me. The fact that I choose to share it with the world is not really a primary motivating factor for me, it just makes it more accessible for me, and sometimes helpful for others.

Next.

Previous.