[00:32:55] [[Special:Log/newusers]] create 10 * Van552 * (New user account) [01:19:18] [[Special:Log/newusers]] create 10 * Gary69 * (New user account) [01:42:36] [[Account limits]] ! 10https://wiki.toolserver.org/w/index.php?diff=8096&oldid=7387&rcid=22109 * 80.239.243.139 * (+7) (/* Queries */ ) [02:29:26] @replag [08:02:50] @replag [14:51:28] myphpadmin is taking forever to load :-( [14:58:50] [[Special:Log/newusers]] create 10 * Cierra319 * (New user account) [15:06:15] mytop has trouble too, and sql request on s3 take forever [15:07:53] replication lag tool seems broken too [15:18:42] Dispenser: you can connect to database servers directly, but not to the mysql proxy [15:19:04] database servers themself are ok and rpelicating [15:19:16] thanks [15:19:34] https://jira.toolserver.org/browse/TS-1672 [15:19:37] only the proxy for s7 works for me [15:47:53] Hi all [15:50:15] Is some one there who know mysql? [15:52:47] Elph: query help or because of the currently existing connection problems? [15:53:16] Is it possible for you to write a query for me to getting the number of articles without article created by bots. [15:53:36] no [15:53:41] thanks [15:53:58] you can only write a query which estimates this value [15:54:19] are toolserver admin? [15:54:20] that's because you don't know who is a bot [15:55:03] we can know who is bot [15:55:09] no i am not a toolserver admin [15:55:26] i use this method in other queries [15:55:44] for which wiki you want to know this value? [15:56:04] ar.wiki [15:56:21] WHERE user_name NOT IN (SELECT user_name FROM user JOIN user_groups ON ug_user = user_id WHERE ug_group IN ('bot')) [15:59:21] you cannot estimate that all bot creation account are in bot group. either because the never had this flag or it was removed because of inactivity [16:02:44] oh. it's true [16:06:29] i had a query which excludes articles created by users having created more than 100 articles on that day, but i cannot find it atm [16:07:31] send it for me when you find it [16:15:58] [[Special:Log/newusers]] create 10 * Stacey58 * (New user account) [16:34:04] GRAHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH. [16:34:15] The Toolserver is falling apart. [16:52:27] Elsie: also s7? [16:52:58] then use mysql -hz-dat-s7-a arwiki_p if you want use the command line tool [16:53:10] I was trying to access enwiki_p. [16:53:12] Or metawiki_p. [16:53:17] I just killed the whole tool. [16:53:24] I don't have time/patience for this bullshit. [16:53:37] (as temporary fix until ts-admin are online and have solved problems at mysql proxy) [16:54:13] The Toolserver is dead. [16:54:22] for enwiki you can use z-dat-s1-b or rosemary [16:54:34] https://toolserver.org/~mzmcbride/watcher/ [16:54:45] Merlissimo: Thank you for trying to be helpful. :-) [16:54:58] But I've had enough of the constant brokenness and headache. [16:55:09] I'm not attaching my name to a tool that works 10% of the time. [16:55:48] no only the mysql proxy is dead, but most tools are using this to connect to replicated databases [16:56:07] Even if the proxy were working, I imagine the databases are corrupt. [16:56:11] I believe some of them are missing. [16:56:24] The Toolserver isn't dead, it's worse: it's dying. [16:56:34] It's not worth the pain. [16:56:59] why should databases corrupt? [16:57:16] There's been ongoing corruption for years. [16:57:35] Re-imports are slow and tedious. [16:58:53] yes, but not today. the failed proxy services causes much less load on server, so replication problems are more unlickely for today [16:59:10] Heh. [16:59:16] That's quite an optimistic view. [16:59:49] ;-) [17:00:16] http://munin.toolserver.org/Database/rosemary/mysql_queries.html [17:00:26] no peaks for today [17:00:28] https://jira.toolserver.org/browse/TS-997 [17:00:34] https://jira.toolserver.org/browse/TS-1427 [17:00:39] https://jira.toolserver.org/browse/TS-1414 [17:00:51] https://jira.toolserver.org/browse/TS-1431 [17:01:31] http://status.toolserver.org/ [17:01:37] All OK... [17:01:42] What nonsense. [17:02:16] Elsie: "Edit me"! :-) [17:02:28] Is the mysql-proxy in munin, or is it separate? [17:03:23] I don't have a username/password to edit the status. [17:03:23] Uh, when switched from red to blue, http://nagios.toolserver.org/ looks much better. [17:03:38] If I did, I'd mark everything as down. [17:03:41] mysql proxy runs as service on the ha-cluster [17:04:29] Elsie: Ha, you're right, that's new. [17:04:48] There's always been a password. It used to not require a name. [17:05:39] Elsie: I remember editing it once just to see if it works, but that may have been a year ago. [17:05:56] There was a small password input at the bottom of the form. [17:06:06] Now there's a username/password input after you click "modify." [17:06:11] DaB. rewrote it at some point. [17:06:17] So turnera is down? But wasn't the idea of HA that then damiana takes over? [17:06:42] Bizarrely, "sql tool server" works. [17:06:46] Bizarrely, "sql toolserver" works. [17:06:59] And "select @@hostname" says damiana. [17:07:13] But I try "mysql -hdamiana toolserver" ... nope. [17:07:15] So fucked. [17:07:28] I suppose I should just gzip my home dir. [17:07:56] the problem is that all services mentioned on status.toolserver.org are really ok [17:08:25] That's like saying "the site is up" when nobody can reach it. [17:08:29] sql and sql-toolserver does not use the proxy [17:08:43] If nobody can reach the databases, they're not okay. [17:08:45] They're broken. [17:08:52] And corrupt, if JIRA is to be believed. [17:09:00] Anyway, I won't argue. [17:09:14] [[Special:Log/newusers]] create 10 * Angella962 * (New user account) [17:09:27] I'm glad to see that the wiki can still be spammed. [17:09:37] I think Platonides suggested at some to replace the mysql-proxy with DNS round robin to remove that point of failure. [17:09:57] Why not just point toolserver.org at tools.wmflabs.org and be done with it? [17:10:14] I'm not really kidding. [17:10:29] i hacked my bots and changed db server names today, so they are currently running ;-) [17:11:10] I tried to do that, but even server names don't work. [17:11:17] $ mysql -hdamiana toolserver [17:11:17] ERROR 2003 (HY000): Can't connect to MySQL server on 'damiana' (146) [17:16:27] why damiana? [17:17:09] there is only sql-toolserver running on damiana a server but using a different routing than damiana, so mysql -hsql-toolserver will work [17:18:04] Because that's what "select @@hostname;" told me to use. [17:18:25] And hostnames are supposed to be ... accessible via -h. [17:25:35] 2013/08/03 17:21 WARN z-dat-s5-b MySQL slave SLOW_SLAVE WARNING: Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 1992 [17:28:35] 2013/08/03 17:21 WARN z-dat-s5-b wikidata replag QUERY WARNING: 'SELECT ts_rc_age()' returned 2163.000000 [17:54:36] 2013/08/03 17:53 CRIT z-dat-s5-b MySQL slave SLOW_SLAVE CRITICAL: Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 3607 [17:54:36] 2013/08/03 17:53 CRIT z-dat-s5-b wikidata replag QUERY CRITICAL: 'SELECT ts_rc_age()' returned 3607.000000 [18:23:40] 2013/08/03 18:17 WARN z-dat-s5-b s4 replag QUERY WARNING: 'SELECT ts_rc_age()' returned 2111.000000 [18:52:42] 2013/08/03 18:52 CRIT z-dat-s5-b s4 replag QUERY CRITICAL: 'SELECT ts_rc_age()' returned 3641.000000 [19:42:48] 2013/08/03 19:42 WARN z-dat-s5-b s4 replag QUERY WARNING: 'SELECT ts_rc_age()' returned 3546.000000 [20:01:49] 2013/08/03 20:01 OK z-dat-s5-b s4 replag QUERY OK: 'SELECT ts_rc_age()' returned 1792.000000 [20:06:07] I'm not sure if anyone's mentioned this - but ACC is down at the moment, due to a db issue [20:06:33] If a root could work on getting FastLizard4 access so maybe he can fix the issue, that'd be great [20:06:36] * Dusti goes back to hiding [20:09:42] Dusti: I think the problem is that no roots are around :-). [22:04:50] 2013/08/03 22:04 WARN z-dat-s5-b MySQL slave SLOW_SLAVE WARNING: Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 3460 [22:04:50] 2013/08/03 22:04 WARN z-dat-s5-b wikidata replag QUERY WARNING: 'SELECT ts_rc_age()' returned 3460.000000 [22:26:51] 2013/08/03 22:26 OK z-dat-s5-b MySQL slave Uptime: 1936923 Threads: 5 Questions: 2519455385 Slow queries: 17872 Opens: 1566392 Flush tables: 1 Open tables: 256 Queries per second avg: 1300.751 Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 1737 [22:26:51] 2013/08/03 22:26 OK z-dat-s5-b wikidata replag QUERY OK: 'SELECT ts_rc_age()' returned 1737.000000 [22:49:52] 2013/08/03 22:46 WARN z-dat-s5-b MySQL slave SLOW_SLAVE WARNING: Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 1979 [22:52:52] 2013/08/03 22:46 WARN z-dat-s5-b wikidata replag QUERY WARNING: 'SELECT ts_rc_age()' returned 2114.000000 [23:04:52] 2013/08/03 23:04 OK z-dat-s5-b MySQL slave Uptime: 1939203 Threads: 4 Questions: 2520546057 Slow queries: 17890 Opens: 1566478 Flush tables: 1 Open tables: 256 Queries per second avg: 1299.784 Slave IO: Yes Slave SQL: Yes Seconds Behind Master: 1687 [23:04:52] 2013/08/03 23:04 OK z-dat-s5-b wikidata replag QUERY OK: 'SELECT ts_rc_age()' returned 1687.000000