[13:34:25] * dcaro paged [13:34:26] looking [13:35:13] Cloudvirt node cloudvirt1048 is down. #page [13:36:37] redis seems to be readonly (according to toolschecker) [13:37:13] three tools k8s workers are down too, probably were running on that cloudvirt [13:38:32] I'm on the console, going to reboot it [13:40:31] * andrewbogott now waits forever to see if it's actually going to boot [13:41:05] ok, there it goes [13:41:14] watching too [13:42:51] host is back up [13:43:54] VMs should come back shortly [13:44:25] k8s nodes coming up again [13:44:52] redis back up again [13:45:35] everything look ok now [13:47:00] UE memory read error [13:47:12] did phault make us a bug for this or should I make one? [13:48:05] ah, T373740 [13:48:09] T373740: NodeDown - https://phabricator.wikimedia.org/T373740 [13:48:49] yep, the logs are interesting [13:48:55] Memory failure: 0x4d47b27: already hardware poisoned [13:50:56] Oh, there's more in the log, I updated that paste on the phab task (if that's where you were reading) [13:51:46] let's see if it's stable, we can gather the full logs on monday [13:52:09] yeah [13:52:26] I guess I'll take my laptop with me when I leave the house :) [13:52:36] I'll send an email about affected VMs. [13:53:30] I'll keep an eye also during EU time [13:53:34] thanks! [13:53:45] dcaro: want me to move the tools hosts off, just in case? [13:54:13] https://www.irccloud.com/pastebin/6ny9EqZb/ [13:57:48] maybe just redis, the others is ok if they go away [13:58:00] yep xd [14:00:39] tools-redis-7 is moved, email sent, going to go back to housekeeping [14:03:37] thanks! I'll be around a laptop for a few hours at least, if you want to stay disconnected for a bit [14:04:09] great, thank you