#archlinux32 | Logs for 2023-05-15

Back
[06:06:16] -!- morriset has joined #archlinux32
[06:30:08] -!- Hobbyboy|BNC has joined #archlinux32
[06:30:30] -!- Hobbyboy has quit [Ping timeout: 260 seconds]
[06:31:05] -!- heartburn has quit [Ping timeout: 260 seconds]
[06:31:05] -!- petris has quit [Ping timeout: 260 seconds]
[06:32:40] -!- heartburn has joined #archlinux32
[06:32:56] -!- petris has joined #archlinux32
[06:34:33] Hobbyboy|BNC is now known as Hobbyboy
[08:03:04] -!- morriset has quit [Quit: Leaving]
[13:12:43] -!- T`aZ has quit [Ping timeout: 268 seconds]
[13:25:48] -!- T`aZ has joined #archlinux32
[13:41:45] -!- T`aZ has quit [Ping timeout: 240 seconds]
[13:43:57] -!- T`aZ has joined #archlinux32
[13:53:44] -!- drathir_tor has quit [Ping timeout: 240 seconds]
[13:59:54] -!- abaumann has joined #archlinux32
[14:06:04] -!- buildmaster has quit [Read error: Connection reset by peer]
[14:07:44] -!- titus_livius has joined #archlinux32
[14:08:42] -!- buildmaster has joined #archlinux32
[14:16:01] -!- drathir_tor has joined #archlinux32
[14:29:39] -!- T`aZ has quit [Ping timeout: 250 seconds]
[14:32:35] -!- T`aZ has joined #archlinux32
[15:08:57] <abaumann> so, mysql should be up and ready I again. Sorry for the delay..
[15:25:14] <abaumann> let's see what happens when I start my python rebuild thingy again.. :->
[15:28:32] <abaumann> >> ERROR 1712 (HY000) at line 1: Index dependencies is corrupted
[15:28:43] <abaumann> of course. now to the fun game of finding a non-broken dump..
[15:29:00] <abaumann> I fail to imagine that a dump is broken after export or broken after import.
[15:29:09] <abaumann> I just think mariadb is complete garbage..
[15:29:56] <abaumann> can I switch the immodb/isam database storage before importing the dump, I wonder?
[15:30:07] <girls> probably "yes"
[15:30:26] <girls> make sure, the encoding does not break, but it should all be plain ascii anyways
[15:30:41] <abaumann> so, that's the next thing I try. I will not bisect 20'000 mysql dumps just for fun. :->
[15:30:49] <abaumann> will check the encodings.
[15:31:41] <abaumann> drop and redo the index?
[15:32:59] -!- finsternis has quit [Ping timeout: 248 seconds]
[15:33:26] <abaumann> select count(*) from dependencies;
[15:33:33] <abaumann> works after an optimize table dependencies;
[15:33:39] <abaumann> I wonder for how long :->
[15:34:24] <abaumann> aha. at least the sanity check runs now and show normal errors, like diffs between the DB and the mirror.
[15:34:43] <abaumann> still: ERROR 1712 (HY000) at line 1: Index dependencies is corrupted
[15:36:54] <abaumann> | buildmaster.dependencies | optimize | note | Table does not support optimize, doing recreate + analyze instead |
[15:37:08] <abaumann> cool. I see no activity whatsoever on the buildmaster
[15:37:47] <girls> O.o
[15:38:26] <abaumann> and the slave is gone again..
[15:38:30] <abaumann> *sigh*
[16:17:50] <abaumann> So, complete rebuild on master and slave is over, corrupted indexes seem to be gone.
[16:18:00] <abaumann> Now there are some diffs to handle..
[16:18:02] <girls> \o/
[16:26:44] <abaumann> The master mirror mirror.archlinux32.org cannot be reached via sftp.
[16:26:47] <abaumann> really? what?
[16:27:05] <abaumann> fail2ban being baning too much?
[16:38:42] <abaumann> python-setuptools-1:67.7.0-1.0-any.pkg.tar.zst starts to tap on my nerves: removing it: error in sanity check, adding it : error in sanity check..
[16:38:51] <abaumann> python-setuptools-1:67.7.0-1.0-any.pkg.tar.zst
[16:38:55] <abaumann> https://buildmaster.archlinux32.org
[16:38:57] <phrik> Title: result of archlinux32 build master's sanity checksanity of the buildmaster's mysql database (at buildmaster.archlinux32.org)
[16:39:30] <abaumann> "..,or vice versa" it would be nice to know, which way round it is :->
[16:40:28] <abaumann> "The replication slave is currently not running. The database might be outdated"
[16:40:33] <abaumann> yeah, I could start over again now..
[16:41:38] <abaumann> May 15 19:41:32 archlinux32 mariadbd[658]: 2023-05-15 19:41:32 513 [ERROR] Slave SQL: Could not execute Write_rows_v1 event on table buildmaster.upstream_packages; Duplicate entry 'zziplib-0-0.13.72-2' for key 'content', Error_code: 1062; handler error HA_ERR_FOUND_DUPP_KEY; the event's master log mysql-bin.000001, end_log_pos 1315861, Gtid 0-1-154, Internal MariaDB error code: 1062
[16:44:11] <abaumann> *sigh*
[16:44:27] <abaumann> so, fix the buildmaster, make a backup, restore the slave, restart the slave..
[16:44:44] * abaumann start some music on his headphones, fitting to the task at hand
[16:50:01] <abaumann> "from the database" does this mean the pacman database or the buildmaster mysql database?
[16:53:42] <abaumann> ok, database ok. now for duplicity..
[16:54:02] * abaumann makes a note that once when he has time he will improve the error messages of the sanity check
[16:54:19] <abaumann> maintainance incoming..
[17:00:53] -!- bill-auger has quit [Ping timeout: 268 seconds]
[17:05:24] -!- bill-auger has joined #archlinux32
[19:03:59] -!- bill-auger has quit [Ping timeout: 240 seconds]
[19:04:33] -!- bill-auger has joined #archlinux32
[19:39:47] -!- heartburn has quit [Ping timeout: 264 seconds]
[19:40:20] <abaumann> mysql>2 ERROR 1064 (42000) at line 23: You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near 'OR `architectures`.`id`=`build_assignments`.`architecture`,1,0) DESC,IF(EXIST...' at line 1
[19:40:29] <abaumann> jippie, error is back.. ok then..
[19:42:17] <abaumann> this needs a proper fix.. but tomorrow. :-)
[19:42:19] -!- abaumann has quit [Quit: leaving]
[19:47:33] -!- heartburn has joined #archlinux32
[20:00:01] <girls> abaumann: upstream trouble ahead: https://archlinux.org
[20:00:01] <phrik> Title: Arch Linux - News: Git migration announcement (at archlinux.org)
[20:01:09] <girls> luckily, it *should* only be some database magic, no need to change our scripts - and luckily (if you want to look at it that way), I broke my foot and have some free time, now ...
[20:11:27] <girls> in the broken sql, there's the architecture missing
[20:11:50] <girls> do you have any idea, when that happened?
[20:12:49] <girls> looks like get-assignment
[20:16:41] <girls> mysql_determine_majority_build_slave_architecture_id returned empty, apparently - maybe just a hickup, let me check
[20:17:33] <girls> aha, this fails, if there are no recent ssh logs - lemme fix that
[20:19:36] <Foxboron> girls: please reach out if you need any information from us :)
[20:20:08] <girls> Foxboron: thanks for the offer, I believe, I understand what's happening :) I will reach out, if I find out, that I was wrong ;)
[20:32:32] <girls> ah, maybe one question ahead for you, Foxboron: although it says "git-migration" it is really "just" a movement of packages inside the git repository - in other words: the git repository stays at the same location and also the directory structure of the packages stays the same (besides renaming testing -> core-testing/extra-testing and so on, of course). Right?
[21:18:04] <Foxboron> girls: we dont have any git repository?
[21:18:18] <Foxboron> it's all svn behind the scenes. All of this are going moved, so each package has their own gitlab repository