[09:33:12] I found more tables/databases to be deleted. Sigh [09:34:11] that's good :-) [09:34:37] better keeping track of them than not at all [09:37:06] Yeah. I noticed them on a semi random wikitech page as a TODO at the bottom [09:37:12] So moved the TODO to a task [09:37:49] thanks! [09:38:06] there is tracking task, make sure to add them there too [09:38:16] Yeah, I did [11:04:53] jynus: Any reason you can see for https://phabricator.wikimedia.org/T87716 to be a private/security bug? :/ [11:05:09] Seems to be corruption/not replication etc [11:05:09] Not leaking of data [11:06:57] no [11:07:02] software error [11:07:16] see springle comment+ [11:07:29] it was the same state on the master [11:07:35] so no replication problem [11:07:40] Ok [11:07:48] So it's definitely not a security bug either? [11:08:08] that should be asked to mediawiki-core devels [11:08:45] I should count as that, I've been doing it long enough :P [11:08:52] It's talking about missing category link rows etc [11:08:57] my opinion is not security bug, but I am just an operator, so you should not listen to me [11:08:58] * Reedy unclassifies [11:09:16] (or that is what some people say) [11:09:52] Ugh, it won't let me remove the security policies [11:10:36] ah, missed a combo obx [11:11:48] in fact. I cannot reproduce [11:13:46] probably a lost quejob [11:13:59] I would say, priority low [11:15:55] https://phabricator.wikimedia.org/T119841 was just merged into it [11:16:09] "Looks like labsdb1001 and labsdb1003 have different data." [11:16:33] that has nothing to do [11:16:42] yours is a production issue [11:16:45] lol [11:17:00] the last one is a labs replication issue [15:19:10] Hi Reedy, thanks for your comment on my patch. That's my first patch, so any help is appreciated. How can I set the password on the puppet template? [15:19:39] is there any user/pass I can use to send the stats? [15:20:27] Reedy, be kind, this is a new "cult" member I am trying to adquire :-) [15:20:41] If it's to go into WMF production, there's a seperate repo used for password and alike (which isn't public) [15:21:16] and how can I reference this user/pass from my script? [15:21:23] So you need to get someone to add a username/password to there, and then you should be able to use it using the format I put on the comment [15:21:32] https://github.com/wikimedia/operations-puppet/blob/production/modules/cassandra/templates/cqlshrc.erb#L3 [15:22:10] you doesn't have access, so isart, suppose you have those in a variable and just use them on a template [15:22:14] Ok, anyone kind enough to add a user/pass for me? :) [15:22:24] So then, when puppet has run and done it's magic, your MysqlPSCollector.conf.erb will become a MysqlPSCollector.conf and then you can just load it and use it like normal [15:24:00] You shouldn't need to worry about that too much, the opsen that reviews (and then merges and deploys it) should setup the login details for you [15:24:22] yep, probably the best thing for now is to assume those get sent as class parameters to you [15:24:33] Just give it descriptive names [15:24:35] Ok, so you mean I should use "<%= @password_variable %>" (with that exact name?) [15:24:36] and they are magically available [15:24:41] ok [15:24:53] mysqlps_user and mysqlps_pass or similar [15:25:25] ok ok, got it [15:25:28] https://github.com/wikimedia/operations-puppet/search?utf8=%E2%9C%93&q=password&type=Code [15:28:26] the eassiest thing is to "commit --amend" the change [15:28:43] the review branch is not used except for reviewing [15:29:00] mmm ok, sorry [15:29:12] so you can freely amend (and actually it is expected) [15:29:59] do not worry! we are very special in some ways- in others it is necessary due to the size of contributors, etc. [15:31:48] and then `git review` again? [15:31:52] yes [15:32:03] and you can abandon the new commit [15:32:30] do not worry, you will get accostum to it on the n'th ammend :-) [15:32:38] ok, I have to go now, but I will work on it later [15:32:48] no prob [20:54:27] Ok, I think it's now done :)