[01:11:42] !log tools.wikibugs restarted wikibugs because of apparent irc split brain (wikibugs and wikibugs____) [01:11:44] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.wikibugs/SAL [08:54:33] !log tools merged several patches by bryan for toolforge front proxy (cleanups, etc) example: https://gerrit.wikimedia.org/r/c/operations/puppet/+/622435 [08:54:35] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL [10:02:33] Hello, world! I got a message for my bot: "Someone (probably you) recently logged in to your account from a new device". Has the software configuration changed on Toolforge or has my account been hacked? [10:31:20] !log tools.wdmm deployed f58fa33946 (wuuwiki update) [10:31:22] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.wdmm/SAL [10:32:53] Iluvatar_: perhaps joakino can help you navigate that [10:50:41] Iluvatar_: have you logged out and in, or had to log in again recently, or logged in from a different browser, computer, or device like a phone? [11:00:16] if you haven't, I would strongly recommend changing the password as soon as possible [11:01:13] the message looks like the one from LoginNotify for the Wikimedia accounts, so think if you have logged in into any wikis around the time when you got the email [15:24:29] andrewbogott: could you hard reboot video-{dev,redis}-buster.video.eqiad.wmflabs? I can't ssh in and the dev one reports insane load (NFS?) and the redis one reports down [15:35:11] zhuyifei1999_: in a meeting but I'll have a look in maybe 30 minutes? [15:35:18] ok sure [16:16:27] zhuyifei1999_: want me to investigate anything first or just reboot? [16:17:10] heh, load average: 174.07, 174.02, 174.04 [16:17:31] * andrewbogott reboots [16:30:12] ikr [16:32:11] andrewbogott: can you reboot the redis instance too? [16:32:25] zhuyifei1999_: I did but it didn't come up right, trying again now [16:32:51] or possibly it was too broken to honor my reboot [16:33:00] seems ok now? [16:33:15] I'm doing some puppet repair on those hosts too while I'm here... [16:34:14] zhuyifei1999_: since you're here… do you have an intuition about whether the encoding instances are throttled on CPU, disk IO, something else? [16:34:28] CPU [16:34:30] I'm interested in cloud-vps workloads that are sensitive to disk IO [16:34:36] (apart from databases which almost certainly are) [16:34:48] ok, so slower disk access would [16:34:55] wouldn't mess with you a whole lot? [16:35:30] Do you have consistent enough performance numbers that you could tell if it got worse or better? If we created an experimental encoding node on a different system? [16:38:02] I don't think I saw a lot of D waits iirc [16:38:22] sure we can get a different node running [16:38:56] but I don't really know performance counters :/ [16:48:19] ok — I'll hold off on that for now then [16:48:30] I think your broken VMs are un-broken for the moment [16:51:44] thanks :) [17:12:48] !log admin Running 'ionice -c 3 nice -19 find /srv/tools -type f -size +100M -printf "%k KB %p\n" > tools_large_files_20200826.txt' on labstore1004 T261336 [17:12:51] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Admin/SAL [17:12:51] T261336: 2020-08-26: tools NFS share cleanup - https://phabricator.wikimedia.org/T261336 [17:41:59] !log tools.paws deleting data in the old NFS location of paws/userhomes T261338 [17:42:01] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.paws/SAL [17:42:01] T261338: Remove PAWS data in Tools volume - https://phabricator.wikimedia.org/T261338 [18:12:39] !log tools.zoomviewer Added `-o /dev/null -e /dev/null` to jsub commands generated by index.php (T248188) [18:12:42] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.zoomviewer/SAL [19:24:32] for completeness, the "308" redirect question came from https://phabricator.wikimedia.org/T256276 which said to use 308 [20:22:13] thanks for the follow up mutante :) [20:35:43] !log tools.zoomviewer Disabled crontab for "jlocal echo ${HOME}/check.sh >> /tmp/debugomat" [20:35:45] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.zoomviewer/SAL [20:36:38] !log tools.zoomviewer Changed schedule of check.sh crontab to once per 5 minutes [20:36:39] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.zoomviewer/SAL [20:39:08] !log tools.zoomviewer Deleted old access.log files (6G total size) [20:39:09] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.zoomviewer/SAL [21:08:00] !log tools Disabled puppet on tools-proxy-06 to test fixes for a bug in the new T251628 code [21:08:05] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/SAL [21:08:05] T251628: Serve some default well known files for Toolforge webservices - https://phabricator.wikimedia.org/T251628 [21:29:08] !log tools.heritage Deploy latest from Git master: 7b0a9d7 [21:29:11] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.heritage/SAL [21:29:39] !log tools.heritage jsub new categorize_images job to test 7b0a9d7 [21:29:41] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.heritage/SAL [22:16:13] !log tools.robokobot truncated file virgule.err since it had reached 97G in size and was just repeated output from a job (not actually errors) [22:16:14] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.robokobot/SAL