| [10:07:29] | * cweagans_afk is now known as cweagans |
| [10:32:57] | * freiheit has quit (Quit: Leaving.) |
| [10:48:08] | * cweagans is now known as cweagans_afk |
| [10:52:47] | * gandhiano has quit (Ping timeout: 246 seconds) |
| [11:09:31] | * fatguylaughing has joined #aegir |
| [11:41:47] | * gusaus has quit (Quit: gusaus) |
| [11:49:42] | * cweagans_afk is now known as cweagans |
| [11:51:37] | * gusaus has joined #aegir |
| [12:10:08] | * Egyptian[Home] has quit (Remote host closed the connection) |
| [12:11:48] | * Egyptian[Home] has joined #aegir |
| [13:03:53] | * Egyptian[Home] has quit (Quit: Leaving.) |
| [13:18:48] | * gusaus has quit (Quit: gusaus) |
| [13:21:58] | * gusaus has joined #aegir |
| [13:32:19] | * Yaazkal has quit () |
| [13:41:21] | * cweagans is now known as cweagans_afk |
| [14:36:06] | * cweagans_afk is now known as cweagans |
| [14:36:35] | * cweagans is now known as cweagans_afk |
| [14:50:55] | * gusaus has quit (Quit: gusaus) |
| [14:57:56] | <ergonlogic> | If I ever start a landscaping business, I'm calling it "Lawn & Order" |
| [15:30:04] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [15:43:44] | <azend> | ergonlogic: just loled |
| [15:47:51] | <realityloop> | ergonlogic: ping |
| [15:57:57] | <p_stampy> | lol ergonlogic |
| [16:02:48] | <realityloop> | whats the likley timeframe till D8 support in Aegir/AegirNG? |
| [16:13:01] | * VeggieMeat_ has quit (Quit: No Ping reply in 180 seconds.) |
| [16:13:37] | * VeggieMeat has joined #aegir |
| [16:14:40] | * ivanjaros has joined #aegir |
| [16:19:25] | * drastik-jw has quit (Remote host closed the connection) |
| [16:32:11] | * boshtian has joined #aegir |
| [17:08:42] | <p_stampy> | realityloop: x2 |
| [17:20:52] | <ergonlogic> | realityloop: Aegir3 will already install D8. the problem is largely that the other Drush maintainers insisted on blocking D8 support in Drush7. So full support isn't likely until Drupal8 AND Drush8 are released, which is likely to be around the same time. |
| [17:46:49] | <p_stampy> | ergonlogic: which is a long way away? |
| [17:48:47] | <p_stampy> | oh |
| [17:49:05] | <p_stampy> | :/ dw |
| [18:12:27] | * e-anima has joined #aegir |
| [18:27:30] | * sdrycroft has joined #aegir |
| [18:53:45] | * drupol has quit (Quit: Page closed) |
| [19:13:17] | * gandhiano has joined #aegir |
| [19:25:22] | * gandhiano has quit (Ping timeout: 272 seconds) |
| [19:35:43] | * stijnvbrande has joined #aegir |
| [19:39:19] | * gandhiano has joined #aegir |
| [20:08:11] | * Egyptian[Home] has joined #aegir |
| [21:00:30] | * drastik-jw has joined #aegir |
| [21:03:48] | * Egyptian[Home] has quit (Quit: Leaving.) |
| [21:04:22] | * mstenta has joined #aegir |
| [21:05:02] | * mstenta has quit (Client Quit) |
| [22:09:00] | * zombiebeard has joined #aegir |
| [22:12:29] | * Yaazkal has joined #aegir |
| [22:21:22] | * mstenta has joined #aegir |
| [22:21:50] | <mstenta> | Hey all, got a small problem with my Aegir install... wondering if anyone can provide some guidance... |
| [22:22:28] | <mstenta> | Yesterday I accidentally ran "drush @sites hosting-migrate @platform", when I meant to run "provision-migrate" - that screwed up my hostmaster site for a bit, but I got it working again. |
| [22:22:58] | <mstenta> | Now, however, in Aegir UI, I cannot "view" the hostmaster platform node. |
| [22:23:13] | <mstenta> | And all links to it are displayed as "http://aegir.barnowl.io/hosting/c/" (notice nothing after the /c/) |
| [22:23:45] | <mstenta> | And I cannot verify either the hostmaster platform or the hostmaster site |
| [22:24:28] | <mstenta> | (well, actually, I'm not sure HOW to verify the platform, because I can't get to it in the UI) |
| [22:24:43] | <mstenta> | When I try to verify the site, it halts with the following error: "Could not find provision alias named: @" |
| [22:25:07] | <mstenta> | This is Aegir 6.x-2.x, FYI |
| [22:31:13] | <mstenta> | I just ran "drush @platform_hostmaster provision-verify" and "drush @hostmaster provision-verify" and they both worked. |
| [22:31:29] | <mstenta> | So I'm guessing things are out of sync in the frontend... ? |
| [22:52:08] | * Yaazkal has quit () |
| [22:54:48] | * zz_drakythe is now known as drakythe |
| [22:55:52] | * Yaazkal has joined #aegir |
| [22:55:52] | * Yaazkal has quit (Client Quit) |
| [23:05:46] | <ergonlogic> | mstenta: hosting-import |
| [23:08:04] | <mstenta> | ergonlogic: that was my hunch... i remember you mentioning that yesterday |
| [23:08:08] | <mstenta> | what is the usage for that? |
| [23:08:18] | <mstenta> | drush help hosting-import just says @context is the argument |
| [23:08:25] | <ergonlogic> | yes |
| [23:08:30] | <mstenta> | what's @context? |
| [23:08:41] | <mstenta> | would that be the platform/site? |
| [23:08:57] | <ergonlogic> | it should be `drush @hostmaster hosting-import @platform_hostmaster` |
| [23:09:07] | <ergonlogic> | then `drush @hostmaster hosting-import @hostmaster` |
| [23:09:12] | <ergonlogic> | or something like that |
| [23:09:42] | <ergonlogic> | I don't recall ever using is on the hm site, but it should work as normal |
| [23:10:13] | <mstenta> | ok thanks, giving that a try now... |
| [23:11:05] | <mstenta> | hmm... |
| [23:11:07] | <mstenta> | WD path: The alias 'hosting/c/platform_hostmaster' for path 'node/1616' (language '') was not created, because the path 'node/7' already [error] |
| [23:11:07] | <mstenta> | has the same alias. |
| [23:11:40] | <mstenta> | followed by: |
| [23:11:41] | <mstenta> | Task verify was added to the queue. Next queue run is 09:11:02-0400, server time is 09:10:24-0400. [status] |
| [23:12:17] | <mstenta> | OH! But I think it worked! |
| [23:12:32] | <ergonlogic> | p_stampy: https://drupalreleasedate.com/ says D8 is coming at the end of September. I believe they're trying to release at DC Barcelona |
| [23:12:47] | <mstenta> | it verified the platform and i can access it again in the frontend |
| [23:12:50] | <mstenta> | i guess the node changed |
| [23:13:02] | <ergonlogic> | looks like |
| [23:17:29] | <mstenta> | ah... so i ran hosting-import on @hostmaster, and then tried to verify the hostmaster site from the frontend... |
| [23:17:33] | <mstenta> | now the frontend is borked |
| [23:17:44] | <mstenta> | CSS seems to be gone, with warnings like: |
| [23:17:44] | <mstenta> | warning: Invalid argument supplied for foreach() in /var/aegir/hostmaster-6.x-2.x/profiles/hostmaster/modules/hosting/client/hosting_client.access.inc on line 174. |
| [23:17:58] | <mstenta> | :-/ |
| [23:19:31] | <mstenta> | it appears that there are two "hostmaster" platform nodes in the frontend now |
| [23:19:53] | <mstenta> | one i can get to at http://aegir.barnowl.io/node/1616 |
| [23:20:01] | <mstenta> | and one that is inaccessible at http://aegir.barnowl.io/hosting/c/ |
| [23:27:08] | <mstenta> | well the way i see this: i've been wanting to switch to Aegir 3 anyways - maybe this is reason to just start fresh |
| [23:33:25] | <bgm> | offtopic, but has anyone found a solution to Jenkins not being able to negotiate ssh ciphers with recent sshd versions? (besides weakening ciphers) |
| [23:34:02] | <bgm> | (all the threads I find suggest to basically go back to sha1, but recent jsch versions are supposed to support modern ciphers) |
| [00:03:08] | * fatguylaughing has joined #aegir |
| [00:20:03] | <ergonlogic> | mstenta: For my part, I'd take a backup of the hm site, then poke around in the db to see about pointing thing back to their original nodes. |
| [00:20:22] | <ergonlogic> | bgm: sorru, I have no idea |
| [00:20:23] | * anarchat has joined #aegir |
| [00:20:23] | * anarchat has quit (Client Quit) |
| [00:20:26] | <ergonlogic> | sorry* |
| [00:24:22] | * cweagans_afk is now known as cweagans |
| [00:40:50] | * hestenet has joined #aegir |
| [00:43:36] | <bgm> | np, tx :) |
| [00:51:55] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [00:58:58] | * ivanjaros has joined #aegir |
| [01:46:42] | * freiheit has joined #aegir |
| [01:55:18] | * Yaazkal has joined #aegir |
| [01:57:58] | * stijnvbrande has quit (Quit: Connection closed for inactivity) |
| [02:06:16] | * ratioweb has joined #aegir |
| [02:09:06] | * freiheit has quit (Quit: Leaving.) |
| [02:16:44] | * josh_k has joined #aegir |
| [02:19:24] | * sdrycroft has quit (Quit: Leaving.) |
| [02:28:56] | * liammcdermott has joined #aegir |
| [02:29:51] | <liammcdermott> | There's quite a bit of spam on: http://community.aegirproject.org/dashboard |
| [02:29:56] | <liammcdermott> | Just FYI. |
| [03:02:00] | * DecipheredAFK has quit (Ping timeout: 272 seconds) |
| [03:02:41] | * DecipheredAFK has joined #aegir |
| [03:08:25] | * freiheit has joined #aegir |
| [03:16:33] | <bgm> | so erm, anyone using something other than jenkins? |
| [03:17:04] | <bgm> | my understanding is that it relies on Java for AES256 crypto, and Java export restrictions locks it down, and it looks really annoying to play with java policies |
| [03:17:47] | <bgm> | I looked at buildbot, but it runs on python2, which seems like a bad idea |
| [03:19:51] | <bgm> | hmm, maybe gitlab CI? |
| [03:20:22] | * DecipheredAFK has quit (Ping timeout: 272 seconds) |
| [03:21:14] | * DecipheredAFK has joined #aegir |
| [03:21:39] | <anarcat> | why is python2 a bad idea? |
| [03:21:45] | <anarcat> | gitlab ci looks interesting |
| [03:22:37] | <anarcat> | but it's ruby :p |
| [03:26:04] | * DecipheredAFK has quit (Ping timeout: 272 seconds) |
| [03:27:15] | * DecipheredAFK has joined #aegir |
| [03:36:06] | * dpayne has joined #aegir |
| [03:36:14] | * dpayne is now known as DanielP |
| [03:46:53] | * gusaus has joined #aegir |
| [04:11:18] | * boshtian has quit (Quit: boshtian) |
| [04:31:43] | * formatC_vt has quit (Ping timeout: 244 seconds) |
| [04:32:07] | * formatC_vt has joined #aegir |
| [04:38:40] | <liammcdermott> | Been using gitlab ci myself, it's okay. |
| [04:41:38] | <liammcdermott> | Does anyone care about the spam on community.aegirproject.org or is that site officially unmaintained/dead now? |
| [04:44:24] | <anarcat> | liammcdermott: i think people are migrating to readthedocs.org |
| [04:44:27] | <anarcat> | but i haven't followed |
| [04:44:29] | <anarcat> | i think people care |
| [04:44:32] | <anarcat> | but it's a hard problem |
| [04:44:39] | <gboudrias> | liammcdermott: I'll look at the spam now |
| [04:44:48] | <formatC_vt> | yes, migrating to http://aegir.readthedocs.org |
| [04:44:49] | <gboudrias> | But there's always more |
| [04:48:24] | <gboudrias> | Done |
| [04:48:44] | <gboudrias> | Maybe we should disable new user creation? It's all spammers at this point as far as I can tell |
| [04:49:18] | <cweagans> | gboudrias: yeah, let's kill it. |
| [04:49:33] | <cweagans> | gboudrias: possibly disable new content creation for all roles too |
| [04:49:51] | <gboudrias> | Ok will do. Anyway people can always email us if they really really want a new user. |
| [04:52:08] | <cweagans> | gboudrias: imo, we shouldn't do anything new there at all. We have a static mirror at community-archive.aegirproject.org, and we could totally just set up a redirect from community -> community-archive |
| [04:52:16] | <gboudrias> | Okay, disabled anonymous user creation and removed the permission to create or edit content (except edit own content because eh) |
| [04:52:48] | <liammcdermott> | gboudrias, thanks! |
| [04:52:54] | <gboudrias> | No problem :) |
| [04:53:07] | <gboudrias> | cweagans: It's probably time. Might be worth an email to the list but I for one approve. |
| [04:56:47] | <liammcdermott> | Does anyone remember a bug back in Aegir 2.1 where running a migrate action for a site would delete that site (database and site files)? |
| [04:57:16] | <ergonlogic> | we're just waiting for someone (volunteers?) to add a bit of js to the archive (on github, btw) that'll add a banner pointing folks at the new resources. |
| [04:57:50] | <liammcdermott> | That is Aegir version >=2.1 |
| [04:59:00] | <ergonlogic> | liammcdermott: I don't |
| [04:59:24] | <ergonlogic> | but there was an issue, since fixed, where backups could be corrupted |
| [04:59:31] | <ergonlogic> | which could have a similar effect |
| [04:59:35] | <anarcat> | ergonlogic: hey |
| [04:59:49] | <ergonlogic> | though the migrate task should just fail and rollback in such a case |
| [04:59:53] | <ergonlogic> | anarcat: hi |
| [04:59:57] | <liammcdermott> | ergonlogic, Heh backups are the one thing that are working. |
| [05:00:10] | <liammcdermott> | And the migrate *appears* to work completely. |
| [05:00:23] | <ergonlogic> | well, the problem was that the backup would *appear* to work |
| [05:00:49] | <DanielP> | i can do the js redirects at around 10pm PST tonight |
| [05:00:56] | <DanielP> | email me daniel.payne.unlimited@gmail.com |
| [05:01:24] | <ergonlogic> | that is, we'd invoke a command, and them pipe it's result into `tar`, iirc |
| [05:01:30] | <ergonlogic> | within the same system call |
| [05:01:38] | <liammcdermott> | ergonlogic, presumably if that were the problem the backups wouldn't work when I manually restore them, but they do. |
| [05:01:40] | <ergonlogic> | I think it was mysqldump |
| [05:01:54] | <ergonlogic> | ok, well, make sure all your tables are there |
| [05:02:08] | <ergonlogic> | the problem manifested as partial db dumps |
| [05:02:20] | <ergonlogic> | often on admin_menu cache tables |
| [05:02:27] | <liammcdermott> | ergonlogic, I see, so it could still be the same issue. |
| [05:02:34] | <ergonlogic> | so early enough that a site couldn't run on them |
| [05:02:44] | <ergonlogic> | well, it's a thought |
| [05:02:58] | <ergonlogic> | I don't recall anything else that'd have that effect |
| [05:03:07] | <liammcdermott> | ergonlogic, in this case there isn't even a database created, it's almost like the site just disappears. |
| [05:03:13] | <ergonlogic> | even though, technically, that bug shouldn't really either |
| [05:03:23] | <liammcdermott> | I suspect it's a problem with the CI scripts I've written |
| [05:03:55] | <liammcdermott> | I was just hoping that upgrading Aegir would magically fix my problem. :) |
| [05:03:57] | <ergonlogic> | there was something, a long time ago (2+ years, iirc) where Aegir could get confused about the db it was trying to delete |
| [05:04:16] | <ergonlogic> | I think the issue was titled 'db gets deleted on edge cases' or something |
| [05:04:31] | <ergonlogic> | well, upgrading to Aegir3 might :) |
| [05:04:33] | <liammcdermott> | Hmmm, that sounds interesting. |
| [05:04:56] | <ergonlogic> | but we've been backporting most stability fixes to 2.x anyway |
| [05:05:13] | <ergonlogic> | so that db one is relatively recent, and would be fixed with an upgrade to 2.4 |
| [05:05:38] | <ergonlogic> | DanielP: thanks for offering to help |
| [05:05:49] | * Yaazkal has quit (Read error: Connection reset by peer) |
| [05:05:51] | <ergonlogic> | cweagans: can you point DanielP at the proper repo? |
| [05:06:39] | <ergonlogic> | DanielP: the archives are hosted on github pages, since cweagans flattened everything to html |
| [05:06:39] | <liammcdermott> | Okay, might as well make the jump to 3, what could go wrong? |
| [05:06:53] | <cweagans> | DanielP: https://github.com/aegir-project/community-archive.github.io, I think? |
| [05:06:59] | <liammcdermott> | ergonlogic, thanks for the help! |
| [05:07:04] | <cweagans> | DanielP: it's the one that starts with community-archive in github.com/aegir-project |
| [05:07:05] | <cweagans> | :) |
| [05:07:08] | <formatC_vt> | ergonlogic> hi, i found a problem with https://www.drupal.org/node/2555129 |
| [05:07:10] | <hefring> | https://www.drupal.org/node/2555129 => Split site and platform Git features [#2555129] => 20 comments, 5 IRC mentions |
| [05:07:24] | <ergonlogic> | DanielP: but we don't want straight redirects. We want a simple banner pointing users at the new resources. |
| [05:07:38] | <formatC_vt> | because hook form alter call order is: |
| [05:07:39] | <formatC_vt> | hosting_git_pull_form_alter |
| [05:07:39] | <formatC_vt> | hosting_platform_git_form_alter |
| [05:07:39] | <formatC_vt> | hosting_site_git_form_alter |
| [05:08:15] | <formatC_vt> | but hosting_git_pull_form_alter should be called last, not first |
| [05:08:55] | <ergonlogic> | formatC_vt: we could make it hosting_git_pull_form_FORM_ID_alter() |
| [05:08:59] | <ergonlogic> | that'd come later |
| [05:09:16] | * mstenta has quit (Quit: Leaving.) |
| [05:09:20] | <formatC_vt> | ok, i will try |
| [05:15:13] | <ergonlogic> | anyone here with in-depth knowledge of Drupal's '#state' element in forms? |
| [05:16:06] | <ergonlogic> | sorry, '#states' |
| [05:18:14] | <gboudrias> | gusaus: ping |
| [05:20:07] | * boshtian has joined #aegir |
| [05:22:00] | <formatC_vt> | ergonlogic> i work sometime with this |
| [05:22:35] | <ergonlogic> | I'm trying to get something like this OR logic working: https://www.drupal.org/node/1464758 |
| [05:22:35] | <hefring> | https://www.drupal.org/node/1464758 => Fixed conditionals to allow OR and XOR constructions => 0 comments, 1 IRC mention |
| [05:22:52] | <ergonlogic> | the problem is that all the keys down the array are identical |
| [05:23:32] | <ergonlogic> | https://api.drupal.org/comment/24708#comment-24708 seems to suggest that you can inject random stuff into one of the keys |
| [05:23:34] | <gboudrias> | ergonlogic: So I'm looking to make an Aegir3 cluster of sorts, is it the new cluster mode that has "smart nodes"? |
| [05:23:38] | <ergonlogic> | experimenting with that |
| [05:23:56] | <ergonlogic> | gboudrias: no smart nodes that I'm aware of |
| [05:24:26] | <gboudrias> | So that's just not a thing? I thought we were discussing it at some point |
| [05:24:34] | <gboudrias> | I'm talking about having Aegirs talk to Aegirs |
| [05:24:47] | <ergonlogic> | by 'smart node' I understand one that has Drush and Provision installed, and can do things like build a platform on a remote server directly, rather than the rsyncs that we do now |
| [05:25:11] | <gboudrias> | I see the pack module is well documented though so that's pretty cool |
| [05:25:11] | <ergonlogic> | the closest we have is remote_import |
| [05:25:15] | <gboudrias> | hm |
| [05:25:57] | <gboudrias> | ergonlogic: Yeah building platforms remotely is the only essential part, I take it that's the Pack module? |
| [05:26:03] | <ergonlogic> | remote_import has a 'remote' class that includes some stuff that'd be helpful with 'smart nodes' imo |
| [05:26:10] | <ergonlogic> | no |
| [05:26:31] | <ergonlogic> | webpack mounts /var/aegir one remotes via NFS |
| [05:26:32] | <gboudrias> | "When configuring a Platform to be deployed on a Pack, choose the "Pack" server node from the Web server radio group during the Platform node creation" ? |
| [05:26:43] | <gboudrias> | okay I see |
| [05:27:07] | <ergonlogic> | from a UI standpoint, you're always going to choose a cluster or pack, as if it was a server |
| [05:27:25] | <ergonlogic> | under the covers they do things differently |
| [05:27:28] | <gboudrias> | Okay thanks |
| [05:27:59] | <gboudrias> | I'm guessing multiserver has more users/support? |
| [05:28:18] | <anarcat> | ergonlogic: i believe you needed my help for some paypal thing |
| [05:28:24] | <anarcat> | ergonlogic: tomorrow is my last day! |
| [05:28:37] | <ergonlogic> | oh right, thanks for the reminder! |
| [05:28:43] | <ergonlogic> | anarcat: you're at the office? |
| [05:28:47] | <anarcat> | i am |
| [05:28:53] | <ergonlogic> | or at least able to answer #208, iirc? |
| [05:28:55] | <formatC_vt> | ergonlogic> I don't use XOR/OR logic before, but if you can share code with pastebin i can try |
| [05:28:55] | <ergonlogic> | ok |
| [05:29:15] | <ergonlogic> | formatC_vt: it's ok, I've got some avenues to explore |
| [05:29:46] | <ergonlogic> | anarcat: hang on, I'll get into paypal now, and try the contact reset stuff |
| [05:29:56] | <anarcat> | ergonlogic: now yes |
| [05:30:10] | <ergonlogic> | ys now :p |
| [05:31:19] | <anarcat> | ? |
| [05:31:23] | <helmo> | anarcat: ergonlogic the paypal login now works .. or was there another issue? |
| [05:31:30] | <anarcat> | helmo: i don't know |
| [05:39:08] | <ergonlogic> | anarcat: it let me change the contact number without confirming w/ the old one this time |
| [05:39:15] | <anarcat> | ack |
| [05:39:31] | <ergonlogic> | but it doesn't look like it wants to let me send funds without further verification |
| [05:39:39] | <ergonlogic> | working thorugh it now |
| [05:39:58] | <ergonlogic> | but it looks like I have to confirm business operations at my address, or the like |
| [05:40:10] | <ergonlogic> | anyway, I'm doing so now |
| [05:43:37] | <ergonlogic> | helmo: ok, it looks like I was able to send you the funds |
| [05:43:43] | <ergonlogic> | helmo: please confirm |
| [05:44:57] | <ergonlogic> | anarcat: so it looks like the last thing is to change the contact name from yours to mine. They're asking me to fax a bill with my name and address. I don't think I'll need your help here. Sorry for the hassle. But last time it was requiring a call to the confirmation number. |
| [05:45:14] | <anarcat> | ergonlogic: no problem |
| [05:49:13] | <formatC_vt> | ergonlogic> hosting_git_pull_form_FORM_ID_alter() -- not help, hosting_git_pull_form_site_node_form_alter called before hosting_site_git_form_alter |
| [05:49:17] | <formatC_vt> | The module order is determined by system weight, then by module name |
| [05:49:41] | <ergonlogic> | oh right |
| [05:49:56] | <helmo> | ergonlogic: Thanks, paypal confirmed. |
| [05:50:23] | <ergonlogic> | formatC_vt: then maybe try making the other ones hook_form_FORM_ID_alter()'s |
| [05:50:47] | <ergonlogic> | formatC_vt: or setting the module weight |
| [05:50:55] | <ergonlogic> | helmo: good :) |
| [05:50:59] | <ergonlogic> | sorry for the delay |
| [05:52:52] | * ratioweb has quit (Ping timeout: 246 seconds) |
| [05:55:44] | <gusaus> | gboudrias: thanks for the tips - that worked :D |
| [05:55:53] | <gboudrias> | No problem :) |
| [05:58:05] | <formatC_vt> | ergonlogic> set module weight - this is a duct tape solution, other ones form_id alter not help because of modules names |
| [05:59:39] | <gusaus> | gboudrias: with regards to creating a site from a git repo, is that different from deploying a platform from a makefile? |
| [06:00:16] | <ergonlogic> | formatC_vt: module weight is global and can have other effects, but go ahead with that solution if you prefer |
| [06:00:26] | <gboudrias> | gusaus: Yes, a platform can contain many sites, and it would just be one site's folder in Git |
| [06:00:57] | <gboudrias> | Also it doesn't really "build" anything, it just downloads it, unlike the makefile which is more of a blueprint |
| [06:04:04] | <gusaus> | gboudrias: ahhh.. ok - i don’t think i’ve had a need to do that…. at least i think :P |
| [06:05:09] | <gboudrias> | gusaus: Well, you're talking about deploying sites with scp, I prefer Git since it allows versioning and all that, but it's more a matter of preference at this point :) |
| [06:07:00] | <gusaus> | gboudrias: ahh… i was talking about copying modules/themes/features from my local computer into a site |
| [06:08:38] | <gboudrias> | Well, you could still use git for existing sites |
| [06:09:21] | * freiheit has left #aegir () |
| [06:12:25] | * DanielP has quit (Quit: DanielP) |
| [06:16:31] | <gusaus> | gboudrias: first i need to create a repository for each on github, yes? |
| [06:16:57] | <gboudrias> | gusaus: Yeah or any git provider. Gitlab allows private git repos for free, for example. |
| [06:18:36] | <gusaus> | ok - so first i’ll need to create repos for all the features - keeping them all on github for consistancy |
| [06:19:33] | <gboudrias> | Well, it depends what you're trying to do, but that's always a good idae |
| [06:19:35] | <gboudrias> | idea* |
| [06:27:58] | <gusaus> | right now i’ve been working on features that i’d include in the next release of the distribution - which also has some security updates |
| [06:28:40] | <gboudrias> | Okay, in that case think you'll want everything in git one way or anyother |
| [06:28:42] | <gboudrias> | another* |
| [06:29:02] | <gboudrias> | At least for everything distro-related |
| [06:29:59] | <formatC_vt> | ergonlogic> oh, this is hard. i don't like this split from 2 modules to 4, for example: now you need enable git, git pull task, but nothing changes because you should also enable platform or site for git |
| [06:31:33] | <cweagans> | formatC_vt: ergonlogic: If we split it into a bunch of modules, we shouldn't have any of them disabled if they already had hosting_git enabled before. We don't want to change available functionality on upgrade. |
| [06:37:22] | <formatC_vt> | I doesn't see a easy solution for this, i need time to think. |
| [06:39:26] | <ergonlogic> | cweagans: formatC_vt helmo added an update hook to enable the site and platform submodules is hosting_git was enabled |
| [06:40:11] | <ergonlogic> | formatC_vt: tbh, I'm not sure why we'd split out the different git commands into separate modules |
| [06:40:29] | <ergonlogic> | it seems like all that should be in hosting git. |
| [06:40:45] | <ergonlogic> | after all, the provision commands are all in the common drush/ dir |
| [06:42:03] | <ergonlogic> | a simple list of checkboxes on a config page could turn the functionality on or off, as far as tasks go |
| [06:42:17] | <ergonlogic> | or better yet, permissions |
| [06:44:15] | <formatC_vt> | the problem is - form alter, pull and checkout should be change form after site_git and platform_git alter's |
| [06:46:19] | <formatC_vt> | but call order depends on module name |
| [06:47:00] | <ergonlogic> | and moving to hosting_site_git_form_site_node_form_alter() and hosting_platform_git_form_platform_node_form_alter()? |
| [06:47:29] | <ergonlogic> | hosting_git_form_alter() should come after both of those |
| [06:47:50] | <ergonlogic> | or hosting_git_pull_form_alter() |
| [06:53:31] | <formatC_vt> | a b c d e f g h i j k l m n o p q r s |
| [06:53:32] | <formatC_vt> | hosting_git_pull_form_alter() |
| [06:53:32] | <formatC_vt> | hosting_platform_git_form_platform_node_form_alter() |
| [06:53:32] | <formatC_vt> | hosting_site_git_form_site_node_form_alter() |
| [06:55:33] | <formatC_vt> | ergonlogic> may be i'm wrong, but this is my module order by module name |
| [06:59:24] | <ergonlogic> | the latter two should run first |
| [06:59:35] | <ergonlogic> | they are implementing a different hook |
| [06:59:42] | <formatC_vt> | yes, but not |
| [06:59:47] | <ergonlogic> | hook_form_FORM_ID_alter() |
| [06:59:56] | <ergonlogic> | as opposed to hook_form_alter() |
| [07:00:08] | <ergonlogic> | the FORM_ID ones run first |
| [07:01:03] | <formatC_vt> | Within each module, form alter hooks are called in the following order: first, hook_form_alter(); second, hook_form_BASE_FORM_ID_alter(); third, hook_form_FORM_ID_alter(). |
| [07:01:32] | <formatC_vt> | each module, not modules |
| [07:08:23] | <formatC_vt> | i need time to think and to sleep, may be tomorrow make some changes in call order to me =) |
| [07:14:30] | <ergonlogic> | hmm, just read drupal_alter() again |
| [07:14:52] | <ergonlogic> | that hadn't been my understanding, but it looks like they're keyed by module, as you say |
| [07:15:01] | <ergonlogic> | so use module weights, I guess |
| [07:29:07] | <ergonlogic> | formatC_vt: thanks for testing. I'll try to take a look at the issue later tonight. |
| [07:47:33] | * Egyptian[Home] has joined #aegir |
| [07:53:47] | * ivanjaros has quit (Quit: https://drupal.org/user/135190) |
| [08:04:27] | * fatguylaughing has quit (Quit: fatguylaughing) |
| [08:08:11] | * liammcdermott has quit (Ping timeout: 244 seconds) |
| [08:08:55] | * zombiebeard has quit (Quit: zombiebeard) |
| [08:30:40] | * gusaus has quit (Quit: gusaus) |
| [08:45:19] | * gusaus has joined #aegir |
| [08:51:55] | * josh_k has quit (Disconnected by services) |
| [08:51:55] | * josh_k_ has joined #aegir |
| [09:07:17] | * hestenet has quit (Remote host closed the connection) |
| [09:10:06] | <realityloop> | cweagans: ping |
| [09:10:14] | <cweagans> | realityloop: pong |
| [09:10:30] | <realityloop> | cweagans: who is best to talk to about Drupal 8 and aegir? |
| [09:10:58] | <cweagans> | realityloop: probably anyone in this room. ergonlogic would probably know best though |
| [09:11:08] | <realityloop> | ergonlogic: are you around? |
| [09:12:15] | <realityloop> | I'm basically trying to figure out whats required to get Drupal 8 support and where it's targeted.. |
| [09:13:31] | <ergonlogic> | realityloop: Drush8 |
| [09:13:40] | <ergonlogic> | realityloop: I'm only here for another minute |
| [09:13:59] | <realityloop> | ergonlogic: is this going to happen in aegir 3.x.. or slated for NG? |
| [09:14:08] | <ergonlogic> | but really, drush7 has an init_hook that die()s if it detects D8 |
| [09:14:17] | <realityloop> | ergonlogic: will you be in Barcelona? |
| [09:14:20] | * drakythe is now known as zz_drakythe |
| [09:14:26] | <ergonlogic> | once Drush8 is out, we should support it shortly thereafter |
| [09:14:36] | <ergonlogic> | probably not |
| [09:14:58] | <ergonlogic> | much as I'd like to, since a branch of my family is from there, and I still have cousins there |
| [09:15:29] | <realityloop> | do we know who will be there? |
| [09:15:48] | <ergonlogic> | Acquia is pushing very hard to release Drupal8 (and Drush8) at DC Barcelona |
| [09:16:03] | <ergonlogic> | I'm guessing some of our European colleagues |
| [09:16:08] | <realityloop> | ergonlogic: I figured that would be the case for Drupal at least.. |
| [09:16:29] | <ergonlogic> | Drush8 should be out around the same time |
| [09:16:38] | <ergonlogic> | ok, gtg |
| [09:16:42] | <realityloop> | we're having internal arguments about whether to stay with Aegir or not atm.. partially stemming from D8 not working under it.. |
| [09:16:44] | <ergonlogic> | back later |
| [09:16:57] | <ergonlogic> | D8 doesn't work under Drush either, atm |
| [09:16:58] | <realityloop> | ergonlogic: can you please ping me when your back.. |
| [09:17:10] | <realityloop> | ergonlogic: yes, I'm aware that Drush is the holdup.. |
| [09:17:24] | <ergonlogic> | what would you consider an alternative then? |
| [09:17:45] | <ergonlogic> | I mean, as soon as Drush starts working for D8, we'll resume efforts to work with it |
| [09:17:53] | <realityloop> | only options we can think of at the moment are acquia, pantheon, or platform.sh |
| [09:18:05] | <ergonlogic> | for that matter, as soon as Drupal 8 starts working :p |
| [09:18:22] | <realityloop> | I'd prefer to stay with something that is open though.. |
| [09:18:28] | <ergonlogic> | do any of them actually duport D8, atm? |
| [09:18:33] | <ergonlogic> | support* |
| [09:18:55] | <ergonlogic> | they all depend on Drush, unless I'm mistaken |
| [09:19:09] | <ergonlogic> | and D8 is a *fast-moving* target right now |
| [09:19:35] | <ergonlogic> | as I mentioned earlier, Aegir should be able to install D8 |
| [09:19:58] | <ergonlogic> | but so long as drush can't do a sql-dump/sqlc, we're kind of stuck |
| [09:20:28] | <ergonlogic> | I used to track Drush head in various places, but it just got too unstable |
| [09:20:38] | <ergonlogic> | like breaking in multiple ways every day |
| [09:21:11] | <realityloop> | is moshe the best person to trach drush 8 status with? |
| [09:21:20] | <realityloop> | track? |
| [09:21:29] | <ergonlogic> | that'd be a good place to start |
| [09:21:35] | <realityloop> | cheers |
| [09:21:37] | <ergonlogic> | greg-1-anderson too, I suppose |
| [09:21:45] | <ergonlogic> | realityloop: let me know what you find out |
| [09:21:56] | <realityloop> | I'd also lilke to have a chat about AegirNG at some stage too if possible.. |
| [09:22:01] | <ergonlogic> | or, I'll just read the #drush backlog later |
| [09:22:11] | <ergonlogic> | realityloop: ok, cweagans is the best choice there |
| [09:22:15] | <realityloop> | thanks |
| [09:22:15] | <ergonlogic> | I really gtg now |
| [09:22:18] | <ergonlogic> | bye |
| [09:23:27] | <cweagans> | Hello again! realityloop - any specific questions about ng ? |
| [09:25:33] | <realityloop> | cweagans: I've been trying to build a vagrant for BOA, but we've also been thinking of looking at Docker.. and I think you were pannong for NG to be docker based..? |
| [09:26:26] | <cweagans> | realityloop: yeah, NG is all going to be docker |
| [09:26:30] | <realityloop> | cweagans: one of the issues with local development using aegir is if DB dies you loose db for all sites.. happened to Deciphered the other day |
| [09:27:02] | <cweagans> | realityloop: Aegir 4 is going to be a D8 UI on top of Kubernetes and likely OpenShift. |
| [09:27:23] | <realityloop> | cweagans: any idea on timeframe? |
| [09:27:24] | <cweagans> | realityloop: however, you won't need to bring that up locally. The idea is that it should be able to run whatever docker image you throw at it |
| [09:27:37] | <cweagans> | realityloop: "when it's ready" :P |
| [09:27:41] | <cweagans> | honestly, no |
| [09:27:58] | <realityloop> | I'm fully aware of that.. :) just looking for some indication.. |
| [09:28:07] | <realityloop> | 6 months.. a year.. less.. |
| [09:29:16] | <realityloop> | cweagans: are you able to do a skype call by any chance? |
| [09:29:26] | <cweagans> | The problem is that the people that are working on it (myself included) are really strapped for time and funding for that matter. To unblock things, we need more people or project sponsors. If I had to estimate it, I'd say 6 months is an okay timeline for at least a prototype assuming we can get some help. Maybe sooner than that |
| [09:29:47] | <cweagans> | it's a pretty complicated stack, but our part of it is relatively simple (it's literally just integrating with some REST APIs) |
| [09:30:12] | <cweagans> | realityloop: sure |
| [09:30:16] | <cweagans> | cweagans on skype |
| [09:30:17] | <realityloop> | what does it mean from infrastructure point of view.. we often host on linde for instance.. |
| [09:31:04] | <cweagans> | realityloop: let's talk on Skype - this will be a much easier conversation :) |
| [09:31:18] | <cweagans> | short answer is that you should be able to host anyway |
| [09:31:21] | <cweagans> | anywhere* |
| [09:35:32] | * e-anima has quit (Read error: Connection reset by peer) |
| [09:36:46] | * gandhiano has quit (Ping timeout: 240 seconds) |
| [09:40:21] | <cweagans> | realityloop: http://kubernetes.io/v1.0/examples/openshift-origin/README.html |
| [09:41:05] | * DecipheredAFK is now known as Deciphered |
| [09:41:06] | * Deciphered has quit (Changing host) |
| [09:41:06] | * Deciphered has joined #aegir |
| [09:41:32] | <p_stampy> | ah |
| [09:41:35] | <p_stampy> | thnks ergonlogic |
| [09:46:30] | <realityloop> | https://github.com/phinze/landrush |
