[09:17:49] 10DBA, 13Patch-For-Review: Reimage & upgrade db1064 - https://phabricator.wikimedia.org/T144723#2614004 (10Marostegui) All looks good after the deployment, so this is done [09:18:09] 10DBA, 13Patch-For-Review: Reimage & upgrade db1064 - https://phabricator.wikimedia.org/T144723#2614005 (10Marostegui) 05Open>03Resolved [12:58:23] "I am a packager and I am going to break your mysql installation" https://www.reddit.com/r/mysql/comments/51fgnt/installation_problems_on_ubuntu_1604_lts/ [13:05:07] Hopefully he is trying in a non critical environment ;-) [13:07:04] and that is why we do not use official packages [13:19:02] that's really not a very productive thread (or comment) [13:19:27] this person should file a bug against the package [13:19:41] and if you have your own list of package bugs or shortcomings, you should do the same too [13:20:09] the disconnect between the DBA community and the packaging community is real but both sides are at fault IMHO [13:49:33] ha ha [13:50:41] I think we discussed this many times, paravoid [13:50:54] yes, this is a bug [13:51:02] but there is no disconnect [13:51:30] it is just separate goals, there is not "your philosophy is wrong and this is what you should do" [13:51:46] it's not really though [13:51:48] packagers are doing what they should [13:51:55] the goal of Debian is to cover all potential use cases [13:52:03] it's not always successful obviously [13:52:23] I disagree, "universal operating system" [13:52:26] but I doubt that you'd hear a "we're not interested about your use case, make your own build" from any packager [13:52:48] <> have absolutly covered 100% of the uses you can do [13:53:16] and debian covers all my needs, just its packaging does not [13:53:39] and I am less than 1% of the use cases [13:53:47] they sould not listen to me at all [13:54:08] specially, because I can solve what I need on my own [13:54:44] they should go for the 99% of the cases where they cannot / do not know what to do [13:55:15] that's fine and that may be the case, but you're kinda making this judgement for them instead of letting them judge it for themselves [13:55:53] I'd hate it if on one of my own packages someone that feels like they have a minor use case doesn't report it to me because they can cope themselves [13:56:12] this issue asside, which is clearly a bug [13:56:33] (of a one time issue of the user) [13:57:04] I do not have an issue with them, I am just not the audience [13:57:22] why do you say so? :) [13:57:33] jynus: don't agree, everyone is the audience in theory [13:57:39] did any packager ever tell you "you're not our audience"? [13:57:43] if so, I'll just shut up :P [13:57:50] and there are a lot of medium companies that use mysql and have similar needs [13:58:12] well, "I cannot reproduce" is indeed a common closing reason [13:58:21] the other is doing nothing about it and let it rot [13:58:38] but just to prove you wrong, I will create a couple of bugs [13:58:42] and see what happens [13:59:18] I'm not saying that these will get fixed or that they won't bitrot [13:59:36] I'm saying that trying to engage with that community can't hurt [13:59:57] paravoid, why do you assume I do not talk regularly with packagers [14:00:03] I did with oracle [14:00:13] and I think redhat/fedora? [14:00:23] they kind of had the same position [14:00:24] and I haven't been pushing strongly towards that direction (neither you, nor Sean), because contributing back is low in priority in the grand scheme of things [14:00:56] and our DBA people are scarce and overloaded :) [14:01:02] "lightly agreeing that the use case is there" [14:01:51] I regularly send patches to oracle and mariadb [14:02:11] I am in their position, and I understand why it takes a lot to respond [14:02:22] it would take me a lot to me too [14:02:45] with so many things open [14:03:10] I would prefer if actual time was spent on real bugs, that in the end I can benefit too [14:03:30] things that affect 99% or 100% of the people [14:04:22] remember that my plan is to migrate to debian packages any time soon [14:04:45] I just need an automatic way to apply patches from its original source [16:42:21] 10DBA, 10Wikidata: [Task] Avoid inconsistencies in Wikidata editing due to slave lag - https://phabricator.wikimedia.org/T88986#2615419 (10hoo) [16:44:03] 10DBA, 10Wikidata: [Task] Avoid inconsistencies in Wikidata editing due to slave lag - https://phabricator.wikimedia.org/T88986#1024803 (10hoo) 05Open>03Resolved a:03hoo I'm closing this and suggest that we rather great individual actionable bugs in case we actually encounter any issues. I don't think th... [16:47:12] 10DBA, 10MediaWiki-Maintenance-scripts, 06Operations, 06Release-Engineering-Team, and 2 others: Add section for long-running tasks on the Deployment page (specially for database maintenance) - https://phabricator.wikimedia.org/T144661#2615428 (10greg) [21:20:21] 10DBA, 10MediaWiki-Maintenance-scripts, 06Operations, 06Release-Engineering-Team, and 2 others: Add section for long-running tasks on the Deployment page (specially for database maintenance) - https://phabricator.wikimedia.org/T144661#2616900 (10greg) I'm also a big +1 on having those long running maint sc...