#archlinux32 | Logs for 2018-07-18

[00:01:24] -!- thePiGrepper has joined #archlinux32
[03:24:07] -!- oaken-source has quit [Ping timeout: 256 seconds]
[05:29:03] * buildmaster failed to execute a mysql query - can you have a look at "tmp.mysql-functions.query.stdin.2018-07-18T05:18:37.Omumau"?.
[07:00:16] -!- titus_livius has joined #archlinux32
[07:01:33] -!- isacdaavid has quit [Quit: Leaving.]
[07:16:28] -!- lukeshu has quit [*.net *.split]
[07:18:59] -!- Alina-malina has quit [Ping timeout: 260 seconds]
[07:34:28] -!- Alina-malina has joined #archlinux32
[07:35:11] -!- lukeshu has joined #archlinux32
[08:03:50] -!- deep42thought has joined #archlinux32
[08:03:59] <deep42thought> Hi buildmaster, what's up?
[08:24:20] <deep42thought> uh, "iotop" on the buildmaster gives a sefault O.o
[08:35:44] <buildmaster> Hi deep42thought!
[08:35:48] <buildmaster> up? I'm up for 6 days, 10 hours, 56 minutes, load average: 122.55, 144.14, 164.98 ... and I'm insane :-D
[08:57:55] * buildmaster resumes sanity.
[09:28:04] -!- abaumann has joined #archlinux32
[09:28:05] <buildmaster> Hi abaumann!
[09:28:13] <abaumann> morns.
[09:28:19] <abaumann> pacman-mirrorlist is ok again.
[09:29:04] <deep42thought> \o/
[09:29:15] <abaumann> iotop segfaulting. ui :-)
[09:29:18] <deep42thought> I also added the check, that the required commits exist
[09:29:30] <deep42thought> either iotop or kill
[09:29:33] <abaumann> yeah. I'm also experiencing some I/O load on the i486 VM..
[09:29:38] <deep42thought> I can't decide which would be worse
[09:29:50] <deep42thought> this was on the buildmaster
[09:29:53] <abaumann> ah.
[09:30:11] <deep42thought> the load was ceiling at ~160 :-D
[09:30:17] <deep42thought> stupid "git gc" ...
[09:30:20] <abaumann> ah. on the VM its a git fetch from community. that takes long..
[09:30:24] <abaumann> exactly.
[09:30:34] <abaumann> there is implicit 'git gc'
[09:30:56] <deep42thought> I disabled that
[09:31:03] <deep42thought> because it blocked "normal" git operations some times
[09:31:15] <deep42thought> but I'm not sure if it was a good idea
[09:31:37] <deep42thought> now we run "git gc" once every day - and usually it stalls everything :-D
[09:31:56] <deep42thought> I tried to fiddle with ionice to make it play nicely, but was not yet successful
[09:32:01] <abaumann> splitting the git repos upstream would really help here.
[09:32:31] <deep42thought> and what makes me wonder is, that the repos seem to grow in size indefinitely
[09:32:48] <deep42thought> although, they're simple mirrors of what upstream does
[09:32:52] <deep42thought> ... no commits
[09:32:54] <deep42thought> no rebase
[09:32:56] <deep42thought> no merges
[09:32:58] <abaumann> mmh.
[09:33:01] <deep42thought> no extra branches
[09:33:10] <deep42thought> git clone --mirror $upstream
[09:34:13] <deep42thought> e.g. community is 1.2GB on the buildmaster, where it is only 600MB on my box here
[09:39:52] <buildmaster> virtualbox is broken (says buildknecht2).
[10:17:48] <abaumann> deep42thought: btrfs?
[10:17:55] <deep42thought> ext4
[10:17:59] <deep42thought> both
[10:18:03] <abaumann> big cluster size?
[10:18:31] <deep42thought> should be all standard
[10:18:35] <abaumann> what is 'du' measuring: usage in blocks or file size?
[10:19:08] <deep42thought> blocks, I think
[10:19:16] <abaumann> let me measure on my build slaves..
[10:19:16] <deep42thought> but nontheless: it grows
[10:19:18] <buildmaster> skia-sharp60 is broken (says buildknecht).
[10:19:43] <deep42thought> apparent-size also gives 1.2GB
[10:21:23] <abaumann> 842 MB on one build slave 902 MB on another.
[10:21:45] <abaumann> I don't see a pattern here.
[10:36:03] * buildmaster failed to execute a mysql query - can you have a look at "tmp.mysql-functions.query.stdin.2018-07-18T10:36:02.dPW1py"?.
[10:36:05] * buildmaster failed to execute a mysql query - can you have a look at "tmp.mysql-functions.query.stdin.2018-07-18T10:36:04.bzszxs"?.
[10:36:10] <deep42thought> whoopsie
[10:36:53] -!- buildmaster has quit [Remote host closed the connection]
[10:43:57] -!- buildmaster has joined #archlinux32
[11:17:57] -!- buildmaster has quit [Remote host closed the connection]
[11:50:51] -!- buildmaster has joined #archlinux32
[11:51:06] -!- buildmaster has quit [Remote host closed the connection]
[11:51:19] -!- buildmaster has joined #archlinux32
[12:08:11] -!- AndrevS has joined #archlinux32
[12:13:47] * buildmaster resumes sanity.
[13:04:25] <buildmaster> virtualbox-modules-arch is broken (says buildknecht).
[13:08:23] -!- AndrevS has quit [Quit: umount /dev/irc]
[14:05:50] -!- abaumann has quit [Quit: Page closed]
[14:54:10] -!- thePiGrepper has quit [Quit: leaving]
[14:55:14] -!- thePiGrepper has joined #archlinux32
[15:58:19] <buildmaster> gitlab is broken (says buildknecht2).
[16:36:05] -!- deep42thought has quit [Quit: Leaving.]
[17:37:04] <buildmaster> sagemath is broken (says rechenknecht).
[18:08:56] -!- isacdaavid has joined #archlinux32
[18:41:42] -!- deep42thought has joined #archlinux32
[18:41:43] <buildmaster> Hi deep42thought!
[18:41:52] <deep42thought> Hi buildmaster, what's up?
[18:41:53] <buildmaster> up? I'm up for 7 hours, 16 minutes, load average: 0.73, 1.35, 1.76
[18:47:25] <buildmaster> hexchat is broken (says tyzoid-srv0-bs0).
[18:52:13] <deep42thought> O.o
[18:53:29] <tyzoid> broken packages are broken
[18:53:40] <deep42thought> git cherry-pick in prepare()
[18:53:47] <deep42thought> requires a set-up git
[18:54:01] <tyzoid> lol
[18:54:05] <tyzoid> might need that added to build deps
[18:54:57] <deep42thought> no git is not the issue
[18:55:04] <deep42thought> having set up name and email
[18:55:07] <deep42thought> so you can commit
[18:55:10] <deep42thought> that's the problem
[18:55:35] <deep42thought> !bug 59377
[18:55:36] <phrik> https://bugs.archlinux.org
[18:56:00] -!- abaumann has joined #archlinux32
[18:56:01] <buildmaster> Hi abaumann!
[18:56:02] <abaumann> hi.
[18:56:05] <deep42thought> Hi abaumann!
[18:56:10] <abaumann> my build slave hangs in linux: Salsa20 stream cipher algorithm (i586) (CRYPTO_SALSA20_586) [N/m/y/?] (NEW) ^C
[18:56:27] <deep42thought> umm
[18:56:33] <deep42thought> maybe I messed something up :-/
[18:56:51] <deep42thought> I'll look into this after dinner
[18:56:56] <abaumann> I had this too today with the i486 kernel config. A make menuconfig helped.
[18:57:05] <deep42thought> make oldconfig
[18:57:09] <deep42thought> ^ better
[18:57:11] <abaumann> or so. yes.
[19:12:04] <buildmaster> krita is broken (says tyzoid-srv0-bs0).
[19:14:58] -!- oaken-source has joined #archlinux32
[19:45:08] -!- abaumann has quit [Quit: leaving]
[19:54:06] <tyzoid> deep42thought: You think something's up with my build slave?
[19:55:56] <deep42thought> hmm?
[19:56:01] <deep42thought> no, why should I?
[19:56:20] <deep42thought> I just stopped it yesterday morning when git.archlinux32.org was overloaded
[19:56:29] <deep42thought> and some hours ago I started it again
[19:57:16] <deep42thought> ls: cannot access '/mnt/archlinux32/pool/vc-*.xz': No such file or directory
[19:57:24] <deep42thought> we don't have that package (yet?)
[20:00:27] -!- thePiGrepper has quit [Ping timeout: 240 seconds]
[20:01:28] <tyzoid> don't you just love when ssh connections start lagging?
[20:01:43] <tyzoid> how long does it take to send a bit of text across a network :/
[20:03:14] <deep42thought> well, as long as it's faster than a post card ;-)
[20:10:22] -!- abaumann has joined #archlinux32
[20:10:22] <buildmaster> Hi abaumann!
[20:10:42] <tyzoid> wb
[20:10:46] <abaumann> hi. :-)
[20:11:31] <abaumann> some trouble with my machines: not one booted without commenting out something in /etc/fstab
[20:12:03] <tyzoid> weird
[20:12:09] <abaumann> the weirdest was a cycle in fsck on a partition, as soon as I put a 0 there (it's 4 TB anyway, so I don't want automatic fsck on boot anyways), things went ok.
[20:12:20] <abaumann> no, improvement, future, systemd and parallelism :-)
[20:12:55] <abaumann> another wird thing was bind mounts depending on a non-fsck partition (yet fscked), similar issues.
[20:13:00] <abaumann> *weird
[20:13:32] <abaumann> * abaumann shrugs and goes dabbling with a DOS GRUB..
[20:21:05] -!- abaumann has quit [Remote host closed the connection]
[20:25:56] <deep42thought> abaumann: I cannot reproduce the kernel issue
[20:26:13] <deep42thought> there is "# CONFIG_CRYPTO_SALSA20_586 is not set" in config
[20:26:24] <deep42thought> and it seems to compile fine
[20:26:30] <deep42thought> and does not change the config
[20:26:55] <deep42thought> if I run "make config"
[20:32:22] <tyzoid> hey abaumann: when you hop on again, I'm testing a varnish config in front of my repos
[20:33:40] <deep42thought> where do you host the varnish?
[20:33:58] <tyzoid> Cloud vps's through DigitalOcean
[20:34:08] <tyzoid> I'm looking to test a global cdn network
[20:34:37] <tyzoid> The network is at http://32arlm.cdn.tyzoid.com
[20:34:38] <phrik> Title:Index of / (at 32arlm.cdn.tyzoid.com)
[20:34:47] <tyzoid> currently ony one node in NYC right now
[20:36:52] <deep42thought> abaumann: is your build slave working with the latest source?
[20:37:11] <deep42thought> maybe it does not have the fail-if-commit-unavailable fix
[20:37:17] <deep42thought> yet
[20:43:33] <tyzoid> deep42thought: Ok, hosts in NYC, London, and Singapore are up
[20:43:41] <deep42thought> :-)
[20:43:45] <tyzoid> That host is set for round robin DNS
[20:43:56] <tyzoid> Need to experiment to see if I can't always route locall
[20:43:58] <tyzoid> locally*
[20:46:34] <tyzoid> deep42thought: Ok, added http://us.32arlm.cdn.tyzoid.com http://gb.32arlm.cdn.tyzoid.com and http://apac.32arlm.cdn.tyzoid.com
[20:46:36] <phrik> Title:Index of / (at us.32arlm.cdn.tyzoid.com)
[20:47:06] <deep42thought> should I put them on the mirrorlist?
[20:47:06] <tyzoid> Note that the above are for testing only, and are subject to going down at any moment ^
[20:47:12] <tyzoid> lol
[20:47:13] <deep42thought> ok :-)
[20:48:28] <tyzoid> deep42thought: If you wouldn't mind, can you run some package download speed tests against the gb one?
[20:52:30] <deep42thought> peak: ~15MB/s, but it seems to take quite a while to initiate connections
[20:52:40] <deep42thought> thus it drops in between to below 1MB/s
[20:52:53] <tyzoid> right, since it has to backhaul traffic from Atlanta, Georgia
[20:53:27] <deep42thought> I shortly had ~30MB/s
[20:53:29] <tyzoid> deep42thought: Can you try again on the same packages?
[20:53:49] <deep42thought> looks similar
[20:53:54] <deep42thought> ah, well
[20:54:00] <deep42thought> maybe somewhat faster
[20:54:05] <deep42thought> yeah, probably
[20:54:06] <tyzoid> I just enabled caching to disk, so it might take another try to get it cached
[20:54:20] <tyzoid> since before it only had 512M of memory cache
[20:54:22] <deep42thought> yeah, much faster now
[20:54:28] <deep42thought> almost instantly at 20MB/s
[20:54:38] <tyzoid> sweet
[20:54:45] <tyzoid> I'll enable that in the other regions
[20:54:47] <deep42thought> 43MB/s :-D
[20:55:01] <deep42thought> I'll stop now
[20:59:01] <tyzoid> alright, disk cache should be enabled in all three regions
[21:00:47] <tyzoid> deep42thought: Last one is spinning up in Amsterdam, but I've got a meeting, so I'll finish configuring that later.
[21:03:12] <tyzoid> actually, finished already
[21:03:23] <tyzoid> it's in rotation, and at http://nl.32arlm.cdn.tyzoid.com
[21:03:28] <phrik> Title:Index of / (at nl.32arlm.cdn.tyzoid.com)
[21:09:55] -!- deep42thought has quit [Quit: Leaving.]
[21:18:18] -!- davor_ has joined #archlinux32
[21:19:20] -!- davor has quit [Ping timeout: 268 seconds]
[21:19:20] davor_ is now known as davor
[22:08:37] -!- abaumann has joined #archlinux32
[22:08:37] <buildmaster> Hi abaumann!
[22:09:22] <abaumann> deep42thought: regarding the linux kernel config, I run make oldconfig again and checked in a new version, which seems to build now.
[22:09:43] <abaumann> * abaumann hopps of the 'Swiss way' :-)
[22:09:44] -!- abaumann has quit [Client Quit]
[22:18:14] <tyzoid> abaumann: Darn, didn't have any time to see you on xD
[22:18:38] <tyzoid> regardless, varnish seems like it's working pretty well, if you wanted me to test a config in front of your system