[07:18:07] A long process of mine has died twice at 4 am (UTC). Is there some job killer running at that time, or some memory-intensive cronjob which might result in some OOM? [07:22:13] It was running on a large exec host and using about 6 GB of memory, a bit more than one core [07:22:23] Should probably keep running for about a week [07:35:18] Nemo_bis: do you know the exitcode of that process? [07:35:31] and the hostname [07:57:43] zhuyifei1999_: the host might have been tools-sgeexec-0919 [07:59:43] Nemo_bis: is the process update_items_fr? [08:00:28] no, oabot_prefill [08:00:59] I don't see that in the recent dmesg [08:01:15] so that's not the kernel OOM killer [08:01:59] if it's the -mem limit then you should get a SIGINT, which should have an identifiable exitcode [14:15:25] Did the SSH host key for login.tools.wmflabs.org recently change? [14:15:59] yurb: a few weeks ago [14:16:19] See the cloud-announce mailing list [14:17:16] chicocvenancio: thanks [22:17:00] hi! quick question – toolsdb user names (s12345) aren’t supposed to be secret, are they? [22:17:25] i. e. I hope I don’t need to worry about making the username of one of my tools public? [22:25:43] no lucaswerkmeister [22:25:52] okay thanks :) [22:26:24] the usernames themselves are probably not really advertised but AFAIK they're not secret [22:26:29] the passwords on the other hand. . . [22:26:41] I’m definitely not publishing those :P [22:33:50] (context: https://github.com/lucaswerkmeister/home/commit/9103a6be879ebe95859eb74dd95e2c4e50691f83)