[14:31:57] hashar: hey, got a moment? [14:32:14] sure [14:33:40] sumanah: ether pad / audio / channel ? :D [14:34:02] channel, hashar [14:34:25] hashar: you've seen me mention (in wikitech-l) a volunteer named Olena who works on testcases? [14:34:40] not at all [14:34:47] is that a recent email? [14:35:13] I don't recall this either [14:35:21] And I have no unread messages in wikitech-l [14:35:59] lucky Roan [14:36:07] hashar: 13 Oct, "Re: [Wikitech-l] Improving tests post-1.18 deployment: We need acceptance tests!" [14:36:43] got it [14:36:46] still unread [14:37:09] hashar: ok. Basically you are better suited to leading/mentoring/advising Olena than I am [14:37:27] I have that emal with date 14 Oct [14:37:41] timezones... [14:37:46] Platonides: I probably sent it late in the day -- yeah [14:38:18] hashar: so can I forward some of her work to you so you can advise her a little? she loves to create test plans, but -- as robla has pointed out -- it would be better if she prioritized them a bit [14:38:20] 14 Oct 02:52 [14:38:32] Oh, it was a thread about acceptance tests [14:38:37] I kind of tldr-ed that one [14:38:44] Date: Thu, 13 Oct 2011 17:52:37 -0700 [14:38:46] hashar: a smoketest/sanity-test testplan/caseset, and then others [14:39:06] I had only read the first one in the thread :) [14:39:20] ahh https://www.mediawiki.org/wiki/MediaWiki_manual_smoke_test [14:39:22] " And I have no unread messages in wikitech-l" for some value of unread? :-) [14:39:29] Yes [14:39:35] this is great materials [14:39:41] I read some of that message, but not all of it [14:39:50] but I think it will be better suited in Selenium suites for which I am not ready yet :D [14:39:59] Oct 13, 17:52 PDT, so I must've read that during my post-flight catch-up in NOLA [14:40:16] RoanKattouw: I am working on writing shorter emails :/ you know the Pascal line I'm about to quote? [14:40:27] No? [14:40:46] hashar: true, but we may as well make use of her interest now and have her write tests that either could be followed manually post-deploy or could be Selenized [14:40:48] (I don't mind you writing long e-mails about topics I *care* about... :P ) [14:41:10] RoanKattouw: it goes something like "I'm sorry, I didn't have the time to write you a short letter about this, so I wrote a long one instead" [14:41:16] haha [14:42:41] there's a funny letter in the internets from the unalphabetized mother in the town to the child [14:42:48] it uses lots of those things [14:43:03] Platonides: liiiink! I would love to read it [14:43:06] including "I was going to send you some money, but i've already closed the enevelope" [14:43:14] heh [14:43:15] *envelope [14:43:25] sumanah, how good is your Spanish? [14:43:31] I'm not sure if there's English version [14:43:44] haha [14:43:47] Platonides: nonexistent. "Donde esta los banos" "Bueno" "Cuba si, Castro no" [14:44:00] xD [14:44:09] *RoanKattouw would have to look up the word 'banos' [14:44:17] "dos cervezas por favor" [14:44:54] RoanKattouw: I'm missing the tilde due to laziness & American imperialism [14:45:02] Oh, OK [14:45:06] That's why I'm not finding it [14:45:18] I'm guessing it means baths/bathrooms, though? [14:45:20] yes [14:45:23] welldone [14:45:26] Right [14:45:28] you get ten free airmiles [14:45:31] Then I've understood everything you said [14:45:36] sumanah: the tests are https://www.mediawiki.org/wiki/MediaWiki_manual_smoke_test are exactly what is needed for Selenium [14:45:54] hexmode told me that two people were going to work on implementing selenium tests [14:45:59] hashar: so you are a Selenium man? good to know. [14:46:03] hexmode: who are those 2 people? [14:46:05] not at all :D [14:46:23] but willing to learn. Will not happen before December at least I guess [14:47:13] sumanah: mglazer and jpos... (can't autocomplete here :P) [14:47:24] Markus Glaser & Jan Paul Posma, hexmode? [14:47:35] no... 1s [14:47:58] Jeremy Postlethwaite [14:48:17] thx mark [14:48:36] so I think it will be better to redirect Olena to Jeremy and Markus [14:48:43] hashar: okay. [14:48:46] will do [14:48:47] for now I am focused on PHPUnit tests and the integration platform [14:48:52] sumanah, you may want to try the first block of http://www.taringa.net/posts/humor/12978475/Los-Mejores-Chistes-Del-Mundo.html thought google translate [14:49:12] which is kind of the work behind the scene whereas selenium is more "customer" oriented [14:49:35] eventually, one day, we will have users send us Selenium tests. That would be great to try to reproduce bugs [14:49:36] understood, hashar. I was thinking, "at least hashar is all test-ish and knows more than I do about test-related things," but of course Jeremy & Markus are slightly more apropos [14:50:11] the pity is that I once again escaped one of your tasks :-\ [14:51:08] hashar: you make it sound as though we are affable enemies, like James Bond and the leader of SPECTRE [14:51:32] xDDDD [14:52:15] "Of course, sumanah, you cannot possibly expect that I would let you live" "On the contrary, hashar. One lump or two?" [14:52:44] I will have to watch some James Bond in english to understand that reference :-))) [14:53:12] hashar: none of it is quite that level of absurd [14:53:35] hashar: "one lump or two" refers to how people take tea: one or two cubes of sugar per cub [14:53:35] the 12 labours of Hashar [14:53:36] cup* [14:53:51] Platonides: oh, like Hercules? is automated testing the Augean stables? [14:54:02] hahaha [14:54:09] kind of [14:54:09] It's been compared to the Augean stables before, yes [14:54:36] hashar, go find a river to clean'em up. [14:54:39] :) [14:54:43] but he's escapign the tasks you give him [14:54:59] having a break. Will be back in roughly one hour [14:55:01] haha [14:55:02] *Platonides expects hashar to do a rm -rf / to clean the tests... [14:55:05] ahha [14:55:34] meanwhile I need to think about how to test Serbian script guessing :-D [14:56:24] and I am so going to read https://en.wikipedia.org/wiki/Serbian_language for work!! [15:35:50] surrur [15:47:11] RoanKattouw: hello [15:47:31] Hi nike [15:48:15] RoanKattouw: can you tell about test2? [15:48:25] test2wiki is a wiki we used to test het deploy [15:48:32] It's the first wiki that was switched to 1.18 [15:48:37] It doesn't have any value right now [15:48:48] Other than that it's a wiki that you can safely switch to some other MW version [15:49:07] *Danny_B|backup wonders what was the testwiki about then [15:49:31] Danny_B|backup: what do you mean? it's to test on [15:49:40] there used to be time when testwiki used to run from trunk or nearly trunk [15:50:23] sumanah: sure. i meant from devs point of view according to the roan's description it seems test2 serves as what test used to serve as [15:50:28] aww those days <3 [15:50:32] Not true [15:50:41] testwiki is still used for staging [15:50:43] That hasn't changed [15:50:58] Unless you mean "used to" in the sense of "2006", then sort of, yes [15:51:58] RoanKattouw: and how does one do configuration changes? [15:52:24] http://wikitech.wikimedia.org/view/How_to_make_a_configuration_change or something [15:52:52] Ahm [15:52:54] https://wikitech.wikimedia.org/view/How_to_do_a_configuration_change [15:54:44] emualliug: hello ;-) [15:54:49] *Nikerabbit reads [16:19:28] RoanKattouw: oh one more question, if some revision doesn't apply cleanly, can I just merge it by hand? [16:20:57] Yes [16:21:09] Ideally you would resolve the conflict but sometimes SVN goes berserk [16:21:17] Note the fact that it's manual merge in your commit summary [16:21:53] notpeter: Offending key in /etc/ssh/ssh_known_hosts:723 [16:21:58] Its SSH key is broken [16:22:06] Needs puppet run on fenari I suspect [16:22:27] RoanKattouw: want me to just fix that by hand on fenari? [16:22:54] you can just del that line from the file [16:23:05] you have the power! [16:27:57] notpeter: https://test.wikipedia.org/wiki/Special:Version 404 [16:28:51] https://test.wikipedia.org/ --> IT WORKS [16:29:24] Fixed [16:29:27] cool [16:29:28] thanks! [16:29:45] Needed a graceful to pick up the config [16:30:42] gotcha [16:30:53] cool. going to call it closed [16:30:54] thanks! [16:40:12] hello there [16:40:22] hi jpostlethwaite [16:40:46] good morning|afternoon sumanah [16:41:30] K4-713, jpostlethwaite: variable defining in $adapter->var_map. discuss. [16:41:35] hello khorn and awjr [16:41:44] ok [16:41:44] Yo. :) [16:41:50] :) [16:42:14] are we able to add anything to $adapter->var_map? [16:42:18] or only cc stuff [16:42:22] Adding extra keys to the var_map is just fine, so long as the keys are still unique. [16:42:29] i have several fields for dd [16:42:30] cool [16:42:36] just wanted to make sure [16:43:07] They don't get looked up unless they map to something in the current transaction structure, so more stuff in there is fine. [16:43:08] i have extra lines of code defined in stage_???payments() [16:43:21] i could remove them an put them in var_map [16:43:25] If you need to override a mapping, that would still go there. [16:43:40] ok, just for the overrides [16:43:40] But everything unique can just go in the var_map, no problem. [16:43:45] i put issuer_id [16:43:53] i can remove it and put in var_map [16:44:10] thanks K4-713 [16:44:16] i could not remember [16:44:33] no problem, jpostlethwaite. [16:44:33] that was all i needed to know :) [16:44:46] thanks awjr [16:44:57] np glad you got it sorted [16:46:07] one more thing K4-713, I think we need to enable IPADDRESS [16:46:10] in the request [16:46:25] there was a note about whose IPA it should be [16:46:34] server or user [16:46:55] it appears to be user_ip [16:47:04] That should be easy. Just uncomment the line in the transaction structure. [16:47:19] cool [16:47:22] i did [16:47:28] Nice. [16:47:32] thanks [16:49:38] Just double-checked: It appears to already be mapped to user_ip, so that should just work. [16:49:42] (I like when that happens) [17:01:43] RoanKattouw: hmm [17:02:05] Deploy time? [17:02:08] yeah [17:02:41] I was wondering whether I should let svn conflict the file or just fix it manually [17:03:14] Depends [17:03:22] Try the conflict route, see what kind of conflict you get [17:03:27] right [17:03:31] If it freaks you out, give up and do it manually [17:03:40] Most of the time conflicts are reasonable [17:03:43] And sometimes they're not [17:03:49] right [17:09:47] RoanKattouw: interesting, now it didn't conflict [17:10:11] hah [17:10:47] svn is so weird [17:13:00] RoanKattouw: I see local changes, what should I do to them? [17:13:19] *RoanKattouw looks [17:13:25] didn't run svn up yet [17:14:04] My fault, I'll commit them later [17:14:10] It's just two files [17:14:15] Are you deploying changes to OutputPage.php ? [17:14:20] (the other file is in WikiEditor) [17:14:30] If you're not changing the same files you'll be fine [17:14:42] I'm not [17:14:53] so I can just go ahead with svn up? [17:16:10] Yes [17:16:19] I apologize for the mess, I'll clean it up after you're done [17:18:35] it's not a problem [17:19:09] RoanKattouw: so, now I have six files, should I deploy them one by one? [17:19:54] You could try sync-dir if you have most files in one dir [17:20:15] That would reduce the number of syncs to three [17:20:24] Since you have five files and three of them are in the Narayam dir [17:20:56] five + one from the commit siebrand did [17:21:03] Oh, right [17:21:25] sync-dir extensions/Narayam 'r101126' [17:21:26] ? [17:21:33] Should work [17:21:36] I have never used that before [17:21:38] So let's find out [17:21:42] hah [17:21:46] *RoanKattouw has to eat dinner now [17:21:51] It's microwave pizza, so it'll be quick [17:21:56] okay [17:21:58] I'll wait a sec [17:22:05] But I kind of have to do it right now cause the microwave just beeped :) [17:22:08] You can go ahead if you want [17:22:11] okay [17:22:14] You have a SIP account, right? [17:22:20] If things go to hell, call 6677 [17:22:33] scary [17:22:35] *sumanah makes a note of RoanKattouw's advice [17:24:03] *Danny_B|backup dials 6677 [17:24:53] if things are going good, press 1, if things are going to hell, press 7, 3, 5, 0, 1, 8, #, 2, 8, 2, * and wait for the operator [17:32:47] You mean, if things are bad, press Left-Left-Up-Star-Down-Down-Square-Right-Up-Triangle. [17:37:03] hmph, it's not working [17:37:13] call 6677, Nikerabbit [17:37:31] Danny_B|backup: did you mean 666? [17:38:05] =) [17:38:07] that's to be called when you already are in hell [17:44:32] omg, wrong revision [18:14:28] RoanKattouw: there's one server which asks to verify key fingerprint [18:15:21] Which one, srv193? [18:15:44] yup [18:19:27] You can ignore that one [18:19:30] puppet will fix it eventually [18:22:02] RoanKattouw: okay, I just said no to it every time, does that mean there is server without changes? [18:22:13] Yes, srv193, but that's fine [18:22:24] Due to the particular nature of that server [18:22:30] It's the testwiki box and it runs off /home instead of /apache [18:22:41] uhum [18:22:43] Even though for some reason it does receive syncs to /apache [18:23:40] btw, sync-dir seems to work, when I figure to give the path relative to the correct directory [18:25:06] Yeawh [18:25:13] All the sync scripts take paths relative to /h/w/common [18:25:18] good to know [18:29:22] Nikerabbit: svn st php-1.18 is clean now, thanks for noticing [18:33:14] RoanKattouw: oki [18:33:24] I'm just being extra careful [18:33:33] Always appreciated [22:53:09] if i run scap on fenari, will it also deploy config changes or do i need to deploy those separately? [22:53:47] RoanKattouw ^ [22:54:05] It will deploy those too [22:54:19] scap means sync all of /h/w/common [22:54:26] In addition to rebuilding things and clearing caches [22:54:29] cool thanks