| [11:00:43] | * gusaus has joined #aegir |
| [11:11:34] | * hestenet has joined #aegir |
| [11:12:09] | * hestenet has quit (Remote host closed the connection) |
| [12:04:39] | * freiheit has quit (Quit: Leaving.) |
| [12:53:52] | * utzulogigan has quit (Ping timeout: 272 seconds) |
| [12:55:37] | * utzulogigan has joined #aegir |
| [15:07:44] | * DecipheredAFK is now known as Deciphered |
| [15:07:44] | * Deciphered has quit (Changing host) |
| [15:07:44] | * Deciphered has joined #aegir |
| [15:32:56] | <gboudrias> | ergonlogic: That's actually pretty amazing |
| [15:34:38] | <gboudrias> | hefring: jonpugh++ awesome work on HSBM |
| [15:34:44] | <gboudrias> | I'll test it as soon as possible |
| [15:34:59] | <gboudrias> | note to self: figure out how the karma thing works |
| [16:37:35] | * Deciphered is now known as DecipheredAFK |
| [16:41:24] | * gusaus has quit (Quit: gusaus) |
| [16:53:28] | * jerryitt has quit (Quit: Connection closed for inactivity) |
| [17:39:44] | * DecipheredAFK is now known as Deciphered |
| [17:44:14] | * Deciphered is now known as DecipheredAFK |
| [18:57:09] | * boshtian has joined #aegir |
| [22:59:29] | * gandhiano has joined #aegir |
| [00:18:34] | * gandhiano has quit (Ping timeout: 240 seconds) |
| [00:44:25] | <jonpugh> | Ergonlogic: gboudrias: I'm sure out work on devshop 1.x on Aegir 3 the last three weeks helped boost that :). |
| [00:48:14] | <jonpugh> | Eh I guess not that much :) installations of 3.2 doubled to over 400 last week. We're still on 3.x |
| [00:48:25] | * boshtian has quit (Remote host closed the connection) |
| [00:51:29] | * gandhiano has joined #aegir |
| [00:56:46] | * gandhiano has quit (Ping timeout: 256 seconds) |
| [01:49:17] | * mstenta has joined #aegir |
| [03:00:14] | * gandhiano has joined #aegir |
| [03:25:04] | * gandhiano has quit (Ping timeout: 276 seconds) |
| [03:35:10] | * gandhiano has joined #aegir |
| [03:59:56] | * gandhiano has quit (Ping timeout: 240 seconds) |
| [04:49:46] | * colan has joined #aegir |
| [05:42:06] | * gandhiano has joined #aegir |
| [05:48:42] | * cweagans has joined #aegir |
| [05:49:48] | <cweagans> | ergonlogic: ping |
| [05:54:41] | <cweagans> | ergonlogic: I have to run, but had an idea I wanted to run by you. What would you think about building something like http://deploybot.com with Drupal and some of the stuff you built for Ran (as part of the Aegir project)? Just a generic, configurable deployment thingy that doesn't manage any server config or anything by itself. It would be a pretty easy thing, and would give us a less complicated place to flesh out some of the deployment |
| [05:54:41] | <cweagans> | triggers, processors, and destinations. Later, when we have Aegir NG working 100%, we can add that as a deployment destination. In fact, that might even be a good foundation for building on top of for a 100% Kubernetes/OpenShift environment. idk - just spitballing here. I need a tool like deploybot, but I really don't wanna pay for deploybot :P |
| [05:54:51] | * cweagans has quit (Quit: Leaving) |
| [06:00:31] | * christefano has joined #aegir |
| [06:04:16] | * hosttor has joined #aegir |
| [06:25:37] | * zombiebeard has joined #aegir |
| [06:27:53] | * colan has quit (Quit: colan) |
| [06:36:02] | * christefano has quit (Quit: christefano) |
| [06:59:59] | * FatherShawn has joined #aegir |
| [07:00:26] | * FatherShawn has left #aegir () |
| [07:02:20] | * FatherShawn has joined #aegir |
| [07:35:04] | * zombiebeard has quit (Quit: zombiebeard) |
| [07:47:50] | * christefano has joined #aegir |
| [07:47:56] | * realityloop has joined #aegir |
| [07:56:49] | * christefano has quit (Quit: christefano) |
| [08:06:19] | * Egyptian[Home] has joined #aegir |
| [08:12:45] | * cweagans has joined #aegir |
| [08:12:51] | <cweagans> | , |
| [08:32:17] | * hosttor has quit (Remote host closed the connection) |
| [08:32:42] | * Yaazkal has joined #aegir |
| [09:03:23] | <ergonlogic> | cweagans: sounds interesting. I'll take a look. |
| [09:13:40] | <ergonlogic> | cweagans: at first pass, deploybot looks something like the source-to-image stuff in openshift |
| [09:14:11] | <ergonlogic> | but with a separate front-end |
| [09:14:50] | <ergonlogic> | that said, yes, it should be pretty easy |
| [09:16:37] | <ergonlogic> | and it could probably act as a front-end for git-based workflows w/ Aegir3 as a deployment destination. |
| [09:21:58] | <ergonlogic> | cweagans: can you flesh out your use case a bit further? |
| [09:24:37] | <cweagans> | ergonlogic: It's sort of like source-to-image, but deploybot is a bit more abstract than that. For instance, I might host my git repo at Github. When I push, the webhook trigger on Deploybot is activated. There can be one or more processing steps that 1) pull down my code, 2) run whatever build step needs to happen, 3) optimize assets, 4) etc, then different deployment targets. So the way I see something like this fitting in would be that |
| [09:24:38] | <cweagans> | we might have a webhook trigger module, some intermediate processing, and then an openshift destination (Where the result of the processing steps are committed and force-pushed to openshift, where a Docker image would be built and deployed). I hadn't considered the usefulness for Aegir 3 users, but yeah, we could have a destination that just triggers tasks in Aegir. |
| [09:25:21] | <ergonlogic> | right |
| [09:25:23] | <cweagans> | ergonlogic: I see cases where STI in openshift might not do everything that a user wants to do with their source when they deploy, so having a way to plug into that process (which, realistically, is just putting a facade in front of it and plugging into *that*) might be a nice thing |
| [09:25:36] | <cweagans> | it would also mean that we have a tool for somebody that just wants to SFTP their code up to shared hosting or whatever |
| [09:25:53] | <cweagans> | being able to have atomic deploys that can be rolled back on shared hosting would be pretty amazing. |
| [09:26:00] | <ergonlogic> | it could also handle some of the state in such apps |
| [09:26:33] | <cweagans> | Right. I think it would be best done as a separate tool from the one that does all the openshift stuff, but they could compliment each other in nice ways. |
| [09:26:34] | <ergonlogic> | backups or snapshots of dbs and filestores, for example |
| [09:27:06] | <cweagans> | So that we can serve a wider market than just people that can run > 3 servers for etcd reliability |
| [09:28:02] | <ergonlogic> | ok, so what's your use-case, specifically? |
| [09:28:48] | <ergonlogic> | we could perhaps start there, spec it out, and start using BDD from the ground up |
| [09:29:36] | <cweagans> | ergonlogic: I have a Jekyll site that I host on Bitbucket. I want to push to the Bitbucket repo and have it automatically deployed (including `jekyll build` + image, css, js optimization and creating gzipped versions of everything) to a shared hosting account. |
| [09:29:53] | <cweagans> | or possibly S3. I haven't decided. |
| [09:30:20] | <ergonlogic> | ok |
| [09:30:23] | <cweagans> | So there's no application lifecycle or anything. It's just literally "generate this thing and copy the files somewhere" |
| [09:30:23] | <ergonlogic> | brb |
| [09:30:33] | <cweagans> | ok. I may have to step away here at some point |
| [09:38:11] | * realityloop has quit (Remote host closed the connection) |
| [09:45:25] | * cweagans has quit (Quit: Computer has gone to sleep) |
| [10:05:21] | * mstenta has quit (Ping timeout: 240 seconds) |
| [10:17:03] | <ergonlogic> | hefring: tell cweagans sorry, had family stuff going on. Let's work on something then. |
| [10:17:03] | <hefring> | ergonlogic: I'll pass that on when cweagans is around. |
| [10:44:58] | * gandhiano has quit (Ping timeout: 256 seconds) |
