#archlinux32 | Logs for 2023-05-14

Back
[05:46:31] -!- abaumann has joined #archlinux32
[05:46:32] <buildmaster> Hi abaumann!
[05:46:32] <buildmaster> !rq abaumann
[05:46:32] <phrik> buildmaster: <abaumann> I start to doubt people still know how Unix permission are supposed to be used
[06:32:46] <abaumann> mmh. the buildmaster spits out interesting error messages on the build clients.
[06:33:06] <abaumann> something arojnd considered_build_assignments
[06:49:07] <abaumann> yeah, some syntax errors around build assignment calculation.
[07:00:24] <abaumann> IF(`architectures`.`id`= OR
[07:00:33] <abaumann> this looks like a bug in the architecture table?
[07:02:08] <abaumann> as if it queries with an unknown architecture, then 'id' is empty and the concatenated SQL statement gets illegal..
[07:06:03] <abaumann> printf '`architectures`.`id`=%s' \
[07:06:03] <abaumann> "$(mysql_determine_majority_build_slave_architecture_id)"
[07:06:08] <abaumann> looks promising
[07:07:54] <abaumann> SELECT COUNT(DISTINCT `ssh_log`.`id`) AS `count`,`architectures`.`id`FROM `ssh_log` JOIN `architectures` ON `ssh_log`.`parameters`=CONCAT(`architectures`.`name`," ") WHERE `action`="get-assignment" AND `ssh_log`.`date`>ADDTIME(NOW(),"-1 00:00:00") GROUP BY `ssh_log`.`parameters` ORDER BY `count` DESC LIMIT 1;
[07:08:00] <abaumann> returns an empty set.
[07:08:38] <abaumann> maybe a bogus entry in the ssh_log causing the join to do silly things.. :-)
[07:09:12] <abaumann> ADDTIME(NOW(),"-2 00:00:00") works
[07:09:19] <abaumann> aha.
[07:19:41] <abaumann> ok, something is building, not python-setuptools, but this is progress.. :-)
[08:13:18] <abaumann> mmh, the forced files on the buildmaster have python-setuptools in them if I force building, then they disappear and the force build clients are building something else
[08:13:27] <abaumann> as if python-setuptools was on a black list
[08:28:58] -!- zxrom has quit [Remote host closed the connection]
[08:31:13] <abaumann> ok, I force python-wheel, just to see if there is a problem with the package or with the mechanism
[08:37:19] -!- zxrom has joined #archlinux32
[08:46:32] -!- drathir_tor has quit [Ping timeout: 240 seconds]
[09:05:52] -!- drathir_tor has joined #archlinux32
[09:43:26] <abaumann> ok, I could force python-wheel, but not python-setuptools. weird.
[10:42:06] <abaumann> don't know, where I have to fiddle now, to kick python-setuptools..
[10:42:08] -!- abaumann has quit [Quit: leaving]
[11:37:22] <girls> abaumann: is it on the build list? Did you put it there with the correct architecture?
[11:37:40] <girls> forcing something, that's not on the build-list with that architecture, will simply do nothing
[13:32:44] -!- abaumann has joined #archlinux32
[13:32:44] <buildmaster> Hi abaumann!
[13:32:45] <buildmaster> !rq abaumann
[13:32:45] <phrik> buildmaster: <abaumann> there is a reason why I do things without reading any documentation: you find out many more mistakes people can make.
[13:32:55] <abaumann> so, refurbishment of eurobuild6 is over..
[13:33:02] <abaumann> ..I hope.
[13:33:15] <abaumann> aeh. yes, I thing I'm using a ~/force script since ages..
[13:42:01] <abaumann> ..it just does a seed-build-list then a prioritize-build-list. adding entries to forced.<arch>.* files
[14:07:02] <abaumann> let me try (now that all build machines are back), if they all behave the same way.
[14:43:49] -!- abaumann has quit [Quit: Lost terminal]
[14:54:46] <girls> yes, but still you need to provide the correct architecture - if you force "i686 archlinux32-keyring", nothing will happen, because it's an "any" build
[15:16:30] -!- zxrom_ has joined #archlinux32
[15:18:36] -!- zxrom has quit [Ping timeout: 268 seconds]
[15:23:33] -!- abaumann has joined #archlinux32
[15:23:33] <buildmaster> Hi abaumann!
[15:23:34] <buildmaster> !rq abaumann
[15:23:34] <phrik> buildmaster: <abaumann> you can not take the blame all the time.. I also want my share. ;-)
[15:36:04] <abaumann> mmh, python-wheel and python-setuptools are both any.
[15:38:14] <girls> and I guess, you didn't schedule them as such ;)
[16:05:13] zxrom_ is now known as zxrom
[17:26:24] <abaumann> aha, forcing any python-setuptools worked now. But I could swear it woked before without that.. but maybe just by accident. :-)
[17:26:44] <abaumann> -rw-r--r-- 1 http http 787054 May 14 18:34 python-setuptools-1:67.7.0-1.0-any.pkg.tar.zst
[17:27:04] <abaumann> ok, much better. I could trigger now simply all pyton modules to be rebuild..
[17:40:54] -!- abaumann has quit [Quit: leaving]
[18:35:53] -!- abaumann has joined #archlinux32
[18:35:54] <buildmaster> Hi abaumann!
[18:35:54] <buildmaster> !rq abaumann
[18:35:55] <phrik> buildmaster: <abaumann> who would have thought digital books burn too. ;-)
[18:35:56] <abaumann> mysql>2 ERROR 1712 (HY000) at line 1: Index dependencies is corrupted
[18:35:58] <abaumann> cool
[18:36:03] <girls> :(
[18:36:25] <girls> this is not a database, this is a joke
[18:36:34] <abaumann> this is my joke :-)
[18:36:43] <girls> you wrote mariadb?
[18:36:56] <abaumann> jokes about mysql being a database..
[18:37:13] <girls> ah, sry, didn't mean to offend ;)
[18:37:21] <abaumann> none taken :-)
[18:37:44] <abaumann> "Use OPTIMIZE TABLE to rebuid the InnoDB table that have corrupt indexes"
[18:37:52] <abaumann> marketing wrote this manual? ;-)
[18:38:00] <girls> :D
[18:40:08] <abaumann> mmh. actually. this could explain why things are slow.. it just ignores the indexes and executes the SQL statements without the indexes..
[18:40:23] <girls> :-/
[18:41:09] <abaumann> https://www.vultr.com causes of corruption, half of them is exacly the use case what databases have been invented for :-)
[18:45:06] <abaumann> "The storage engine for the table doesn't support repair" whut!?
[18:50:37] <abaumann> mysqldump: Error 1712: Index dependencies is corrupted when dumping table `dependencies` at row: 0
[18:50:56] <abaumann> Well, you can not repair it, and when you want to export import it as SQL, this also fails.
[18:51:15] <girls> we should have a recent dump, don't we?
[18:51:30] <abaumann> yeah, but this creates other troubles..
[18:52:22] <abaumann> -rw-r--r-- 1 root root 148333312 May 14 18:39 database-2023-05-14-18:23:35.xz
[18:52:34] <abaumann> recent enough (I hope)
[18:52:47] <abaumann> and with my usual luck, it will also be corrupt.
[18:52:48] <abaumann> :-)
[19:19:23] <abaumann> mmh. the database should not fail just like that. Either the machine or the memory has issues..
[19:19:53] <abaumann> ..or runnig bleeding edge mariadb :-)
[19:22:50] -!- abaumann has quit [Quit: leaving]
[19:26:23] -!- abaumann has joined #archlinux32
[19:48:49] -!- abaumann has quit [Remote host closed the connection]