[03:41:04] https://grafana.wikimedia.org/d/000000106/parser-cache?viewPanel=5&orgId=1&from=now-7d&to=now [03:41:04] https://grafana.wikimedia.org/d/000000106/parser-cache?viewPanel=5&orgId=1&from=now-7d&to=now [05:33:19] wow [05:33:19] wow [05:33:35] that's so nice [05:33:36] that's so nice [08:35:15] I guess we could add a column manually_reviewed to the backups table, which defaults to false? [08:35:16] I guess we could add a column manually_reviewed to the backups table, which defaults to false? [08:36:05] We don't use it no? [08:36:05] We don't use it no? [08:36:17] what do you mean? [08:36:17] what do you mean? [08:36:30] What is that column for? [08:36:30] What is that column for? [08:36:35] We use it at the moment? [08:36:35] We use it at the moment? [08:36:55] to answer riccardo's comment about what to do when a backup size change is expected [08:36:55] to answer riccardo's comment about what to do when a backup size change is expected [08:37:13] eh? [08:37:13] eh? [08:37:26] there is no such a column, it would be a new feature of backup checks- when manually_reviewed = ignore alerts [08:37:26] there is no such a column, it would be a new feature of backup checks- when manually_reviewed = ignore alerts [08:37:26] which comment? [08:37:27] which comment? [08:37:45] I assumed so yes, but is there a way to tell it that it's ok and move on? [08:37:45] I assumed so yes, but is there a way to tell it that it's ok and move on? [08:37:47] in -ops, I was asking if there is a way to tell the backup/alerting system that it's ok that m5 backup is so smaller [08:37:47] in -ops, I was asking if there is a way to tell the backup/alerting system that it's ok that m5 backup is so smaller [08:37:49] Ah on the other channel [08:37:49] Ah on the other channel [08:38:06] we have a similar feature for regular backups [08:38:06] we have a similar feature for regular backups [08:38:18] which is job_ignorelist [08:38:18] which is job_ignorelist [08:38:34] but I am not fond of using it, as it leads to people forgetting to revert [08:38:34] but I am not fond of using it, as it leads to people forgetting to revert [08:38:55] in this case it would be for only 1 run, so it would be less impacting [08:38:55] in this case it would be for only 1 run, so it would be less impacting [08:39:47] in general, I would like to see backup monitoring out of icinga, but there is not a lot of alternatives right now for "alerts that are not that prioritary" [08:39:47] in general, I would like to see backup monitoring out of icinga, but there is not a lot of alternatives right now for "alerts that are not that prioritary" [10:45:51] Parsercache data usage is still going down :D [10:45:51] Parsercache data usage is still going down :D [10:56:37] did you see https://github.blog/2021-09-27-partitioning-githubs-relational-databases-scale/ ? [10:56:37] did you see https://github.blog/2021-09-27-partitioning-githubs-relational-databases-scale/ ? [10:57:18] I have it marked to be read! [10:57:18] I have it marked to be read! [14:40:11] jynus: you might get alerts on backups size due to this: https://phabricator.wikimedia.org/T290340 [14:40:11] jynus: you might get alerts on backups size due to this: https://phabricator.wikimedia.org/T290340 [14:40:23] that table is huge on some wikis (ie: dewiki is 115GB) [14:40:23] that table is huge on some wikis (ie: dewiki is 115GB) [14:41:02] ok, thanks. less size == faster backups! [14:41:02] ok, thanks. less size == faster backups! [14:50:08] I was about to come and make the same comment about that ticket :D [14:50:08] I was about to come and make the same comment about that ticket :D