| *** masaki has joined #openstack-doc | 00:02 | |
| *** charcol has joined #openstack-doc | 00:06 | |
| *** gmann has quit IRC | 00:15 | |
| *** gmann has joined #openstack-doc | 00:15 | |
| *** edmondsw has joined #openstack-doc | 00:54 | |
| *** edmondsw has quit IRC | 00:58 | |
| *** phuongnh has joined #openstack-doc | 01:02 | |
| *** imacdonn_ has quit IRC | 01:08 | |
| *** yamamoto_ has quit IRC | 01:31 | |
| *** yamamoto has joined #openstack-doc | 01:35 | |
| *** thorst has joined #openstack-doc | 01:49 | |
| *** thorst has quit IRC | 01:49 | |
| *** dmacpher has joined #openstack-doc | 02:33 | |
| *** edmondsw has joined #openstack-doc | 02:42 | |
| *** edmondsw has quit IRC | 02:46 | |
| *** thorst has joined #openstack-doc | 03:22 | |
| *** thorst has quit IRC | 03:30 | |
| *** mriedem has quit IRC | 04:02 | |
| *** nicolasbock has joined #openstack-doc | 04:07 | |
| *** gouthamr has quit IRC | 04:38 | |
| *** suyog has quit IRC | 04:55 | |
| *** suyog has joined #openstack-doc | 05:09 | |
| *** masaki has quit IRC | 05:24 | |
| *** thorst has joined #openstack-doc | 05:26 | |
| *** thorst has quit IRC | 05:33 | |
| *** masaki has joined #openstack-doc | 05:39 | |
| *** masaki has quit IRC | 05:52 | |
| *** masaki has joined #openstack-doc | 05:52 | |
| *** thorst has joined #openstack-doc | 06:01 | |
| *** thorst has quit IRC | 06:05 | |
| *** edmondsw has joined #openstack-doc | 06:18 | |
| *** edmondsw has quit IRC | 06:23 | |
| *** suyog has quit IRC | 06:52 | |
| *** andreas_s has joined #openstack-doc | 06:53 | |
| AJaeger | dhellmann, asettle I proposed https://review.openstack.org/#/c/487714/ to update the minimal openstackdocstheme version in global-requirements. While the older versions work, I'd like to see this one used... | 07:07 |
|---|---|---|
| *** rcernin has joined #openstack-doc | 07:13 | |
| *** dmacpher has quit IRC | 07:23 | |
| *** alexchadin has joined #openstack-doc | 07:26 | |
| *** charcol has quit IRC | 07:31 | |
| *** vijaykc4 has joined #openstack-doc | 08:01 | |
| *** thorst has joined #openstack-doc | 08:02 | |
| *** edmondsw has joined #openstack-doc | 08:06 | |
| *** thorst has quit IRC | 08:06 | |
| *** edmondsw has quit IRC | 08:11 | |
| amotoki | hi, a horizon developer document refers to http://docs.openstack.org/admin-guide/cross_project_cors.html | 08:18 |
| amotoki | I see the corresponding file still exists in openstack-manuals repo. | 08:18 |
| amotoki | Will this file will be kept in openstack-manuals, or will it be migrated to some project? | 08:19 |
| amotoki | if latter, how can I know it? | 08:19 |
| amotoki | sorry for bothering you. I remember the doc-migration etherpad and it gave me the answer. | 08:21 |
| amotoki | but it might be not easy for folks not familiar with the doc-migration work. | 08:22 |
| *** efoley has joined #openstack-doc | 08:28 | |
| openstackgerrit | Yuval Brik proposed openstack/openstack-manuals master: karborclient: add docs https://review.openstack.org/487756 | 08:35 |
| asettle | Thanks AJaeger :) | 08:47 |
| *** masaki has quit IRC | 09:02 | |
| *** vijaykc4 has quit IRC | 09:08 | |
| *** sambetts|afk is now known as sambetts | 09:14 | |
| *** vijaykc4 has joined #openstack-doc | 09:17 | |
| openstackgerrit | Merged openstack/openstack-manuals master: Revert "the mistral team deleted their admin guide landing page" https://review.openstack.org/487352 | 09:23 |
| asettle | AJaeger dhellmann - did we ever finish the release notes conversation? https://review.openstack.org/#/c/486295/ | 09:38 |
| AJaeger | asettle: not really. if you or dhellmann want to work on the change, please do... | 09:38 |
| asettle | AJaeger: looking now :) | 09:40 |
| openstackgerrit | Merged openstack/openstack-manuals master: move docs for template generator into the contributor guide https://review.openstack.org/487596 | 09:48 |
| openstackgerrit | Merged openstack/openstack-manuals master: Karbor: has admin guide https://review.openstack.org/487439 | 09:48 |
| openstackgerrit | Merged openstack/openstack-manuals master: change config-guide links to the correct "config-reference" https://review.openstack.org/487562 | 09:48 |
| *** edmondsw has joined #openstack-doc | 09:54 | |
| *** alexchadin has quit IRC | 09:57 | |
| *** edmondsw has quit IRC | 09:58 | |
| *** yamamoto has quit IRC | 10:00 | |
| *** thorst has joined #openstack-doc | 10:03 | |
| *** thorst has quit IRC | 10:07 | |
| *** alexchadin has joined #openstack-doc | 10:11 | |
| *** phuongnh has quit IRC | 10:23 | |
| *** thorst has joined #openstack-doc | 10:24 | |
| *** vijaykc4 has quit IRC | 10:27 | |
| *** thorst has quit IRC | 10:28 | |
| *** yamamoto has joined #openstack-doc | 10:31 | |
| *** vijaykc4 has joined #openstack-doc | 10:33 | |
| *** thorst has joined #openstack-doc | 10:47 | |
| *** thorst has quit IRC | 10:51 | |
| *** vijaykc4 has quit IRC | 11:00 | |
| robcresswell | asettle: One thing I'd like to look into next cycle is how the docs theme handles header levels: h1, h2 etc | 11:03 |
| robcresswell | atm it is not semantically correct I think, which means anyone with accessibility issues is gonna have a bad time | 11:03 |
| asettle | robcresswell: put it on the list: https://etherpad.openstack.org/p/doc-future-problems :) | 11:04 |
| * setuid grins "Does the EOL process need to change?" Nope, all good, perfect! | 11:13 | |
| *** vijaykc4 has joined #openstack-doc | 11:15 | |
| *** kazsh has quit IRC | 11:18 | |
| *** kazsh has joined #openstack-doc | 11:19 | |
| *** alexchadin has quit IRC | 11:20 | |
| *** edmondsw has joined #openstack-doc | 11:42 | |
| *** edmondsw has quit IRC | 11:47 | |
| *** thorst has joined #openstack-doc | 11:54 | |
| *** vijaykc4 has quit IRC | 11:54 | |
| *** catintheroof has joined #openstack-doc | 12:11 | |
| *** yamamoto has quit IRC | 12:14 | |
| *** khushbu has joined #openstack-doc | 12:21 | |
| *** mriedem has joined #openstack-doc | 12:27 | |
| *** edmondsw has joined #openstack-doc | 12:33 | |
| *** khushbu has quit IRC | 12:37 | |
| *** khushbu has joined #openstack-doc | 12:39 | |
| *** khushbu has quit IRC | 12:44 | |
| openstackgerrit | Merged openstack/openstack-manuals master: karborclient: add docs https://review.openstack.org/487756 | 12:59 |
| *** lucasxu has joined #openstack-doc | 13:03 | |
| *** yamamoto has joined #openstack-doc | 13:14 | |
| *** yamamoto has quit IRC | 13:21 | |
| *** vijaykc4 has joined #openstack-doc | 13:23 | |
| *** alexchadin has joined #openstack-doc | 13:25 | |
| *** khushbu has joined #openstack-doc | 13:29 | |
| *** gouthamr has joined #openstack-doc | 13:31 | |
| *** dustins has joined #openstack-doc | 13:32 | |
| *** khushbu has quit IRC | 13:34 | |
| *** dustins has quit IRC | 13:34 | |
| *** dustins has joined #openstack-doc | 13:35 | |
| *** alexchadin has quit IRC | 13:35 | |
| mriedem | so we have a pretty serious issue with the docs migration in nova breaking cached urls all over google | 13:38 |
| mriedem | google for "docs nova-manage list_cells" and the first hit is https://docs.openstack.org/nova/latest/cells.html which is a broken link now | 13:38 |
| mriedem | it's now at https://docs.openstack.org/nova/latest/user/cells.html | 13:38 |
| mriedem | do we need to be putting redirects in the nova docs? | 13:38 |
| mriedem | also why i had to make this change in the ocata manual https://review.openstack.org/#/c/487141/ | 13:39 |
| AJaeger | mriedem: our sitemap currently contains https://docs.openstack.org/nova/latest/cells.html - I regenerated a few days before you moved content around. | 13:40 |
| AJaeger | I can regenerate the sitemap again, this should help with google... | 13:41 |
| AJaeger | mriedem: but in general we can put in redirects... | 13:45 |
| *** dims has joined #openstack-doc | 13:46 | |
| mriedem | ok right now i'm just going to keep track of the broken links and we can deal with them after FF | 13:46 |
| mriedem | thanks | 13:46 |
| *** dustins has quit IRC | 13:53 | |
| *** dustins has joined #openstack-doc | 13:56 | |
| setuid | What is wrong with this line: "When the ``router_distributed = True`` flag is configured, routers created" | 14:02 |
| setuid | in the mitaka-eol tag, tox barfs on it: "unknown option: --distributed True" | 14:02 |
| AJaeger | setuid: check how the line looks in newer branches of that file. | 14:03 |
| AJaeger | That looks like a problem where a newer sphinx broke it. | 14:04 |
| setuid | AJaeger: what's odd, is that it builds on Linux, does not build on Mac, using the same versions of tooling | 14:04 |
| AJaeger | I think you have :option:``--distributed`` there - and it should be just ``distributed``. | 14:04 |
| setuid | Doesn't exist in master | 14:05 |
| setuid | config-dvr-ha-snat.rst that is | 14:05 |
| openstackgerrit | Andreas Jaeger proposed openstack/openstack-manuals master: Update sitemap https://review.openstack.org/487863 | 14:06 |
| AJaeger | setuid: check stable/ocata branch | 14:06 |
| setuid | nor older tags. Hrmph, this is beginning to get really messy | 14:06 |
| AJaeger | check for :option: like I said aboe | 14:06 |
| setuid | Yep, looking now | 14:06 |
| AJaeger | mriedem, asettle, dhellmann , updated sitemap to fix cells.html page and include all the new pages and links ^ | 14:07 |
| setuid | I guess tox doesn't really report the exact line of the error, only the start of the struct | 14:08 |
| *** kjw3 has joined #openstack-doc | 14:11 | |
| setuid | AJaeger: Oh the irony. I fixed it so it looks like the one in stable/ocata, and now it complains: "WARNING: Malformed :option: u'distributed True', does not contain option marker - or -- or / " | 14:13 |
| AJaeger | remove :option: | 14:13 |
| setuid | This is precisely the kind of errata that I was referring to on the ML, about doing post-retirement patching so these eol versions can still be built by operators in their own offline or staged copies | 14:17 |
| setuid | I don't mind supplying a bulk tactical patch, but I don't want to be the maintainer of an eol doc fork somewhere on the outside ;) | 14:18 |
| AJaeger | setuid: totally different problem. | 14:19 |
| AJaeger | setuid: the problem here is that our tools evolved over time, nothing we could have known at that time. | 14:19 |
| asettle | Hi hi hi hi what? | 14:19 |
| asettle | AJaeger: got it | 14:20 |
| AJaeger | sorry to confuse you, asettle | 14:27 |
| asettle | Hhaha I'm just not following today | 14:28 |
| asettle | >.< | 14:28 |
| *** lucasxu has quit IRC | 14:30 | |
| *** lucasxu has joined #openstack-doc | 14:32 | |
| *** mihalis68 has joined #openstack-doc | 14:37 | |
| setuid | AJaeger: Yep, I understand that, things change. | 14:43 |
| setuid | Managed to get past the option issue, now stuck back at the same one as yesterday: "ConnectionError: HTTPSConnectionPool(host='git.openstack.org', port=443): Max retries exceeded with url: /cgit/openstack/cinder/plain/etc/cinder/api-paste.ini?h=stable/mitaka (Caused by NewConnectionError('<urllib3.connection.VerifiedHTTPSConnection object at 0x105b2cbd0>: Failed to establish a new connection: [Errno 65] No route to host',))" | 14:43 |
| setuid | the 'No route to host' is interesting, since everything else I've thrown at it, can get to that URI | 14:43 |
| setuid | It should be a 404 (my patch corrects stable/mitaka to mitaka-eol, so a 200), but shouldn't be an SSL timeout | 14:45 |
| dhellmann | AJaeger , asettle : o/ -- I'm getting a late start today and have some TC & release duties to take care of before I catch up here. | 14:47 |
| AJaeger | dhellmann: no worries, no fires today ;) | 14:48 |
| setuid | very weird, always dies on block-storage/block-storage-sample-configuration-files, times out and tosses the above error | 14:58 |
| setuid | This, among many other reasons, is why I avoid python wherever and whenever I can | 14:59 |
| *** lucasxu has quit IRC | 14:59 | |
| *** vijaykc4 has quit IRC | 15:01 | |
| *** lucasxu has joined #openstack-doc | 15:02 | |
| *** rcernin has quit IRC | 15:03 | |
| setuid | This may be down to the py2.7 + SSL/tls issues that plague everyone | 15:04 |
| *** annegentle has joined #openstack-doc | 15:07 | |
| openstackgerrit | Merged openstack/openstack-manuals master: Update sitemap https://review.openstack.org/487863 | 15:13 |
| *** annegentle has quit IRC | 15:16 | |
| *** annegentle has joined #openstack-doc | 15:17 | |
| *** annegentle has quit IRC | 15:21 | |
| *** sdague has joined #openstack-doc | 15:22 | |
| asettle | Ah shite, AJaeger dhellmann - do you guys have the docs team meeting in your cal? | 15:28 |
| AJaeger | asettle: nope. | 15:28 |
| asettle | I just realised it's been 2 weeks. But for some reason it's been removed from my calendar | 15:28 |
| AJaeger | When is it? In 30 mins? | 15:29 |
| asettle | Ugh *rolls eyes* | 15:29 |
| asettle | Well, I'm not sure anymore :/ cause we cancelled the last one. LEt me check that date we cancelled. | 15:29 |
| *** andreas_s has quit IRC | 15:29 | |
| *** skelso has joined #openstack-doc | 15:29 | |
| asettle | Ah lord, it is in 30 minutes | 15:29 |
| asettle | Normally I'm a bit more aware of it D: | 15:30 |
| *** annegentle has joined #openstack-doc | 15:30 | |
| asettle | annegentle: have you got the docs meeting in your cal for 30 mins time? | 15:30 |
| asettle | Mine seems to be removed, and I think it's today D: | 15:30 |
| annegentle | asettle yep, sure do! | 15:30 |
| asettle | Ohhh ffs | 15:30 |
| asettle | I have a conflict | 15:30 |
| asettle | Naturally | 15:30 |
| annegentle | asettle though of course I'm double-booked | 15:30 |
| annegentle | asettle ha yeah... sigh | 15:31 |
| asettle | So am I :( | 15:31 |
| * AJaeger is double booked as well... | 15:31 | |
| asettle | This time isn't working particularly well for anyone, is it? | 15:31 |
| dhellmann | AJaeger, asettle : how easy/hard is it to get info from the web server about 404 pages so we could add some redirects for popular old pages that have moved? | 15:31 |
| setuid | doc/install-guide/source/index.rst has: .. title:: OpenStack Installation Guide | 15:32 |
| AJaeger | dhellmann: infra team runs the apache server and has logs, best ask them | 15:32 |
| dhellmann | AJaeger : ack, thanks | 15:32 |
| setuid | tools/build-install-guides.sh has: sed -i -e "s/\.\. title::.*/.. title:: ${title}/" ${INDEX} | 15:32 |
| asettle | yep ^^ what AJaeger said | 15:32 |
| AJaeger | asettle: it could be worse ;) | 15:32 |
| asettle | AJaeger: I'm going to email out about changing the meeting time again. I just can't do it, and it appears you guys can't either. | 15:32 |
| setuid | Why does this work on Linux (whether default or ${title}), but fails on Mac, claimming an error in "title" directive | 15:32 |
| setuid | I fee like I'm falling all over subtle platform/pip/python issues, not actual doc issues | 15:33 |
| sdague | AJaeger: yeh, we've got a lot of 404s on nova docs that we only started realizing today | 15:37 |
| asettle | What do you mean it fails? As in, fails to build/render? | 15:37 |
| sdague | it would be extremely useful to get that log set exposed to help in building the redirects | 15:38 |
| *** annegentle has quit IRC | 15:38 | |
| asettle | sdague: also a lot of doc patches from Stephen to fix too: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:doc-migration but I know he's out at the moment | 15:39 |
| asettle | Might be helpful to get someone into those | 15:40 |
| sdague | asettle: sure, but that doesn't affect existing links being 404, right? | 15:40 |
| sdague | asettle: my use case / concern - https://docs.openstack.org/nova/latest/cells.html | 15:41 |
| sdague | that's linked all over the place, as are similar things | 15:41 |
| asettle | sdague: https://docs.openstack.org/nova/latest/user/cells.html | 15:41 |
| asettle | I believe the restructure will help | 15:41 |
| sdague | asettle: how? | 15:42 |
| asettle | I mean, don't get me wrong, it's redirects too, and ensuring everything goes into the right places. It will just be helpful to have the migration completed, and ensured that all 404s are caught, and redirected appropriately | 15:42 |
| asettle | it is redirects* (not too) | 15:42 |
| *** edmondsw has quit IRC | 15:42 | |
| asettle | Just saying it will help :) | 15:43 |
| sdague | asettle: ok, I guess I don't understand how the restructure helps on the 404s, it seems to just make more 404s | 15:43 |
| sdague | I want a path to be able to address those with redirects baked in before pushing more in | 15:43 |
| asettle | We proposed a new layout for the documentation, as you're aware of. The build job that was implemented for all docs relies on that layout. This determines the publishing locations/URLs | 15:45 |
| *** sambetts is now known as sambetts|afk | 15:45 | |
| sdague | asettle: right, sure, I get all those things. I'm trying to understand how we address the redirect blackhole issue as we do those things | 15:46 |
| sdague | I'd be all for a fast push on these if I had a way to fix latent 404s at the same time | 15:48 |
| *** edmondsw has joined #openstack-doc | 15:50 | |
| asettle | sdague: adding redirects currently is basically band-aiding the problem. I genuinely recommend you consider stephenfin 's patches, to fix the actual problem. Passing a patch for redirects would potentially take the same amount of time to pass patches to fix the problem as a whole. Basically, I don't encourage adding redirects when the link is adjusted, not broken. Our redirects live here: https://github.com/openstack/openstack-manuals/blob | 15:51 |
| asettle | /master/www/.htaccess | 15:51 |
| setuid | Looks like rolling sphinx-build back to 1.3.6 on Mac solved the build issues there | 15:52 |
| asettle | setuid: that's random | 15:53 |
| *** vijaykc4 has joined #openstack-doc | 15:53 | |
| setuid | Yep, 1.3.6 on Linux, 1.6.3 on Mac, fails on Mac, rolled back to 1.3.6 there by forcing version with pip, now the manuals built (running the manual sphinx-build target that failed previously). Re-running with tox now. | 15:54 |
| asettle | Well I'm so glad that was easy for you :P | 15:54 |
| setuid | I can't get Linux to pull 1.6.3, so it'll stay at 1.3.6 for now. | 15:54 |
| setuid | I'm here to solve doc problems, not patch core python problems | 15:54 |
| sdague | asettle: I guess I don't understand not solving this problem. For all those wikis, docs, and other bookmarks that people have linking into the nova documentation, sending them to a 404 page seems really suboptimal. | 15:55 |
| setuid | asettle: An idea about sdague's issue: Would it make sense to have a simple query that they can link to, which would always 301 them to the current/latest version of pages they're linking to? | 15:56 |
| *** efoley has quit IRC | 15:56 | |
| setuid | vs. getting a 404, a simple search query that lands on a 301 | 15:56 |
| setuid | That way, you solve it once, and don't have to do it for every page/dir you move | 15:56 |
| dhellmann | I think maybe sdague and asettle are not talking about the same thing? | 15:56 |
| asettle | I'm starting to think we might not be haha! | 15:56 |
| asettle | Sorry sdague for misunderstanding you. Let's start this again :) | 15:56 |
| dhellmann | after we finish the migration, there will be lots of links pointing to the old locations | 15:56 |
| sdague | asettle: ok, cool | 15:57 |
| dhellmann | it would be useful to be able to redirect from those locations to the new ones, without having to have the full list of those redirects live in the openstack-manuals repo | 15:57 |
| dhellmann | here's a strawman: | 15:57 |
| dhellmann | add a .htaccess file to each project for their redirects | 15:57 |
| dhellmann | express them in terms of /$project/latest/$old-path redirecting to /$project/latest/$new-path | 15:57 |
| setuid | But then you need to maintain that, and .htaccess is read for every single fopen() on the server side, which would auger performance down | 15:58 |
| dhellmann | then set a flag in openstack-manuals/www/project-data/latest.yaml to indicate that the team has such an .htaccess file | 15:58 |
| asettle | Hmmm | 15:58 |
| dhellmann | and instead of redirecting to /latest/index.html redirect to /latest/the-rest-of-the-path | 15:58 |
| setuid | How about /$project/latest/find?doc=nova-foo, which lands in a 301 in $new-path? | 15:58 |
| dhellmann | to allow those in-tree redirects to work? | 15:58 |
| sdague | setuid: it is, however the performance concerns on that do go back to the 90s, and these files are going to end up in page cache pretty quick | 15:59 |
| sdague | I don't want to not solve a UX problem because of an extra fstat | 15:59 |
| dhellmann | if that design will work, I can write it up in more details | 15:59 |
| setuid | sdague: That depends, but it does dramatically increase backend IO | 15:59 |
| *** phuongnh has joined #openstack-doc | 16:00 | |
| setuid | sdague: agreed, the issues are separate, but I don't think .htaccess is the best approach, because now you have to edit/maintain the contents of .htaccess for every page you change/move, which means you'll miss pages and have to fix it up each time someone finds a broken 404. | 16:00 |
| sdague | setuid: but at least it's fixable | 16:00 |
| dhellmann | another option of course is to put real pages at the old locations with links that say "this moved" but that's a bit janky | 16:01 |
| sdague | the auto search redirect thingy I've done in the past on drupal sites, there is a module for it | 16:01 |
| sdague | dhellmann: agree it is janky | 16:01 |
| sdague | and it won't correct search engines | 16:01 |
| dhellmann | right | 16:01 |
| asettle | Sorry guys, I've had to tap out - next meeting :( | 16:01 |
| asettle | sdague: hopefully you get your answer, thanks for helping setuid and dhellmann | 16:01 |
| asettle | I'll read scroll back I get a chance | 16:01 |
| sdague | returning a 301 is the right answer | 16:01 |
| setuid | Short-term, .htaccecss, longer-term, auto-redirect when 404's are found | 16:02 |
| sdague | for all the crawlers to update results | 16:02 |
| dhellmann | how does the auto search work? what are the search terms? | 16:02 |
| ianychoi | asettle, fine although I have been one of active meeting attendees haha | 16:02 |
| setuid | dhellmann: You can make it as simple or as rich as you need | 16:02 |
| sdague | dhellmann: typically it takes the path chunked up | 16:02 |
| sdague | but, it's kind of a bit of tech all on it's own | 16:02 |
| setuid | ?query=doc vs. ?query=doc&version=ocata&redir=permanent, or whatever | 16:03 |
| dhellmann | setuid : assume there are 0 people to populate the data and it has to work based only on the URL. How useful could you make that? | 16:03 |
| dhellmann | ok | 16:03 |
| sdague | dhellmann: you end up building a search engine | 16:03 |
| dhellmann | path chunks might be minimally useful | 16:03 |
| setuid | dhellmann: Pretty simple, you take everything after /project/ and split on '/', and then match | 16:03 |
| sdague | that's fine, it's just ends up being it's own deal | 16:03 |
| dhellmann | but there's noone to build a search engine for us, so *shrug* | 16:03 |
| sdague | project level htaccess also lets projects rename pages that are named poorly | 16:03 |
| sdague | I would love for that to be a thing | 16:04 |
| setuid | It doesn't need to be a heavy-weight engine, just something that does a simple GET to see if it returns a 200. If 200, passthrough, if 404, search/split/301, if 500, send to an error page + email to operators | 16:04 |
| setuid | sdague: fuzzy logic search works well on drupal/wordpress style plugins | 16:04 |
| dhellmann | I think it's already possible to have an in-tree .htaccess, it's just that it would never see the old-old url before this migration | 16:04 |
| sdague | setuid: sure, but you have a database with all the path names | 16:04 |
| *** egallen has quit IRC | 16:04 | |
| ianychoi | asettle, just one issue with I18n I would like to discuss is how to deal with translation support for migrated documents - for example, networking-guide -> doc/source/admin in neutron repository: http://git.openstack.org/cgit/openstack/neutron/tree/doc/source - have a good Thursday evening & night! | 16:04 |
| dhellmann | setuid : remember, there are 0 developers on this project | 16:04 |
| setuid | sdague: you already do, it's your filesystem ;) | 16:04 |
| sdague | setuid: so you are going to replace an fstat with a find :) | 16:05 |
| sdague | and not expect that to hit your io | 16:05 |
| setuid | nooooo | 16:05 |
| sdague | dhellmann: ok, do you have an example of .htaccess redirect that I could do? | 16:05 |
| setuid | Well, wait a second, is sitemap auto-generated? | 16:05 |
| setuid | What if the same process that gen'd sitemap, dropped .htaccess in the right places? | 16:06 |
| sdague | dhellmann: I've got a bootstrapping problem in figuring out how to do the first one of these | 16:06 |
| *** lucasxu has quit IRC | 16:06 | |
| asettle | ianychoi: might be best to talk to dhellmann about that :) | 16:07 |
| setuid | sdague: Question (this sounds very similar to what I'm trying to solve here too), you're basically talking about off-domain wiki/community pages/resources that link back to docs.o.o, and find 404's there, right? | 16:07 |
| sdague | setuid: correct | 16:07 |
| setuid | sdague: Is that because they're linking to docs that have gone end-of-life, or because of site reorg/compoonent renames? | 16:07 |
| sdague | right now, the reorg is the big hitter | 16:07 |
| ianychoi | asettle, yep - thanks! | 16:07 |
| sdague | but it's a general problem for the eol case as well | 16:08 |
| dhellmann | stand-by | 16:08 |
| setuid | sdague: Right, I'm 110% on the eol case, that's why I asked | 16:08 |
| dhellmann | sdague : let's experiment this way | 16:09 |
| dhellmann | what's that old cells URL? | 16:09 |
| dhellmann | we'll put a redirect in the openstack-manuals .htaccess to point from that to /nova/latest/$path | 16:09 |
| sdague | dhellmann: https://docs.openstack.org/nova/latest/cells.html | 16:09 |
| dhellmann | right | 16:09 |
| dhellmann | oh, we don't even need to do that in the global .htaccess if it's already pointing to latest | 16:10 |
| dhellmann | let me try a patch, hang on | 16:10 |
| sdague | yeh, I'm totally fine if I can do this in nova specific htaccess | 16:10 |
| sdague | I just need examples | 16:10 |
| dhellmann | that'll work for this url, but it's not going to work for all | 16:10 |
| dhellmann | it's going to take me a try or two to figure out the incantation, but I think we can do it | 16:11 |
| sdague | sure, I guess with all the docs getting decentralized back into the projects, conceptually this means stuff like this should get project level managed as well, otherwise we're going to have these log jams in fixing things | 16:11 |
| sdague | dhellmann: cool | 16:11 |
| dhellmann | sdague : let's see what https://review.openstack.org/487932 does | 16:14 |
| dhellmann | we should be able to tell by poking at the draft docs output once the job finishes | 16:14 |
| sdague | dhellmann: ok, cool | 16:14 |
| sdague | thank you | 16:14 |
| dhellmann | sdague : is there a ML thread on this (I got a late start today and haven't caught up yet) | 16:14 |
| sdague | there is not | 16:15 |
| sdague | it started with a debug task where wrong links turned up, and mriedem incorrectly thought the sitemap fix fixed this | 16:15 |
| dhellmann | if you could start a thread stating the problem, I can try to describe my proposed fix in more detail | 16:15 |
| ianychoi | dhellmann, hello! Can I discuss briefly with the translation support for migrated documents - for example, networking-guide -> doc/source/admin in neutron repository: http://git.openstack.org/cgit/openstack/neutron/tree/doc/source ? | 16:16 |
| sdague | sure | 16:16 |
| dhellmann | that way we have a better record and more folks can contribute to make sure we're not going down a bad path | 16:16 |
| dhellmann | cool, thanks | 16:16 |
| setuid | I think this is another topic for the PTG... specifically about the current reorg, vs. the proposed eol archived docs site/solution/strategy | 16:16 |
| AJaeger | dhellmann: a .htaccess can break the complete page, I would rather have them in one place... | 16:17 |
| AJaeger | complete site I mean | 16:17 |
| dhellmann | AJaeger : it would only break stuff under the project team's URL, right? | 16:17 |
| dhellmann | I'm trying to figure out a way to do this that doesn't require the docs team to maintain a list of redirects for every old docs page | 16:17 |
| AJaeger | dhellmann: I fear it would break the whole site - like we did already last wek ;) | 16:18 |
| sdague | AJaeger: well, you have to decide if this is going to be a decentralized effort or a centralized one | 16:18 |
| dhellmann | well, that was because we broke the root .htaccess file, right? | 16:18 |
| dhellmann | I mean, I share your concern if it's possible to break the whole site, I just thought it wouldn't be | 16:18 |
| AJaeger | dhellmann: yes we did - let's evaluate that | 16:19 |
| dhellmann | we could probably build some sort of separate list of changed paths, but I'd rather not do anything more complex than necessary | 16:19 |
| *** khushbu has joined #openstack-doc | 16:19 | |
| dhellmann | ianychoi : hi, yes! | 16:20 |
| dhellmann | ianychoi: do you have thoughts on how best to make that work? | 16:21 |
| ianychoi | dhellmann, thanks - I think you are dealing with doc migration a lot :) Can translation support be also dealt as a Jenkins job within openstack-unified-publish-jobs or different jenkin job? | 16:22 |
| dhellmann | ianychoi : we should be able to tie it into the new publishing jobs. I don't have a good way to separate *part* of the output of those jobs for translation though. Is that going to be a problem? | 16:23 |
| * dhellmann has to step away but will be back in a few minutes | 16:23 | |
| *** khushbu has quit IRC | 16:23 | |
| ianychoi | dhellmann, actually there may include two things: One is to sync translation resources (e.g., .po, .pot) with project repositories and the other is how to publish translated documents if translation ratio is good | 16:25 |
| AJaeger | dhellmann: release notes are partially translated today | 16:26 |
| ianychoi | And unfortunately I am a little bit far from previous https://docs.openstack.org/draft/draft-index.html . Would someone know the status for draft publication? | 16:26 |
| asettle | Thanks for the email, sdague :) | 16:26 |
| sdague | asettle: no prob | 16:28 |
| ianychoi | Aha, https://docs.openstack.org/draft/draft-index.html was moved to https://docs.openstack.org/pike/index.html . Then for I18n team, where would be the best location to review draft translated documents? New index.html in openstack/i18n repository? | 16:30 |
| *** phuongnh has quit IRC | 16:32 | |
| dhellmann | AJaeger : it sounds like the release notes job will be a good place to look for an example | 16:34 |
| dhellmann | ianychoi : we don't do "draft" publishing any more. everything from openstack-manuals goes live as soon as it merges now | 16:34 |
| dhellmann | ianychoi : that may disrupt your current processes for translation, but we can work on that now that the majority of the migration is done | 16:35 |
| * dhellmann steps away to pick up lunch | 16:35 | |
| ianychoi | dhellmann, I see. Have a good lunch :) | 16:36 |
| openstackgerrit | Ian Y. Choi proposed openstack/openstack-manuals master: Removes 404 links in ja landing page https://review.openstack.org/487939 | 16:50 |
| ianychoi | dhellmann, plz review ^ . I investigated in seeing all language landing pages and just two links in Japanese language landing page affected. | 16:52 |
| dhellmann | ianychoi : looking now | 17:02 |
| *** hyakuhei has quit IRC | 17:02 | |
| *** hyakuhei has joined #openstack-doc | 17:02 | |
| *** hyakuhei has quit IRC | 17:02 | |
| *** hyakuhei has joined #openstack-doc | 17:02 | |
| *** vijaykc4 has quit IRC | 17:03 | |
| *** skelso has quit IRC | 17:04 | |
| dhellmann | ianychoi : we should talk about how to handle language landing pages with the new templating structure | 17:04 |
| dhellmann | for example https://docs.openstack.org/ja/ | 17:04 |
| dhellmann | should that be https://docs.openstack.org/$series/ja/ now? | 17:04 |
| dhellmann | or is it ok to keep the language page at the top level since it has sub-sections for different series | 17:05 |
| ianychoi | dhellmann, IMO the # of translated documents is pretty lower than documents in release. | 17:05 |
| dhellmann | that's true | 17:05 |
| dhellmann | I suppose users look by language first, too, and not by release | 17:05 |
| dhellmann | ianychoi : /ja/ops-guide/ still exists on the web server. I think we have a bad redirect preventing access. Do you want to remove the content or fix the redirect? | 17:06 |
| dhellmann | same for user-guide | 17:07 |
| ianychoi | dhellmann, oh really? Then it would be much better to fix redirect | 17:07 |
| dhellmann | ok, let me see if I can do that | 17:07 |
| amotoki | +1 for having a language first. by doing so, we can have a same structure as for the source language (english). | 17:07 |
| *** khushbu has joined #openstack-doc | 17:08 | |
| dhellmann | yes, I like that idea | 17:08 |
| ianychoi | Previously, zh-CN team argued like https://blueprints.launchpad.net/openstack-i18n/+spec/redirct-local-landing-page . Would it be a good idea to make a redirection from https://docs.openstack.org/$series/ja/ to https://docs.openstack.org/ja/ ? | 17:08 |
| ianychoi | amotoki, +1 for master publishing documents but stable/{$series} documents, /$series/ja/ would be fine? | 17:09 |
| amotoki | on the other hand, as my hat of Japanese coordinator, we will lose all translated documents as a result of doc-migration and I am not sure we can recover translations sooner or later. | 17:10 |
| openstackgerrit | Doug Hellmann proposed openstack/openstack-manuals master: anchor redirects for old guides https://review.openstack.org/487952 | 17:11 |
| dhellmann | ianychoi : I think this is what we want ^^ | 17:11 |
| *** vijaykc4 has joined #openstack-doc | 17:12 | |
| *** khushbu has quit IRC | 17:12 | |
| dhellmann | amotoki : I thought that the translation tool recognized text that was "the same" but in different places? perhaps I underestimated the impact this would have on translation :-/ | 17:13 |
| amotoki | dhellmann: we can keep translations for past releases, but Japanese team lost most active translators on documentations and at now there is no active translators for documents. | 17:14 |
| amotoki | so I don't think we can have translated documents for pike or later soon. | 17:15 |
| amotoki | this is my point. | 17:15 |
| dhellmann | oh, we lost the translators, not the translations | 17:15 |
| dhellmann | I misread that | 17:15 |
| dhellmann | that is indeed unfortunate | 17:15 |
| ianychoi | I can do merge activities if all texts previously and later would be the same | 17:15 |
| amotoki | current active translators work on horizon and their plugins. | 17:15 |
| dhellmann | they will be *mostly* the same, but I don't know if they will be identical | 17:15 |
| ianychoi | But I have not tested whether the merge works with different Zanata projects in translate.o.o | 17:16 |
| *** mihalis68 has quit IRC | 17:16 | |
| *** rcernin has joined #openstack-doc | 17:16 | |
| *** skelso has joined #openstack-doc | 17:17 | |
| ianychoi | And I am glad that now translated htmls for Japanese ops-guide and user-guide are still alive. However, since translated publish job executions were already deleted, I am a little bit worried about lifetime for translated htmls which are not cared by publish job executions. | 17:17 |
| amotoki | I am one of document translators the ja team lost (from the other reason) and focus on UI translation as a translator :p | 17:17 |
| setuid | dhellmann: What magical incantations must I chant over tox to get it to respect all of the various http{,s}_proxy settings I've shoved at it? | 17:19 |
| amotoki | ianychoi: thanks anyway | 17:19 |
| ianychoi | amotoki, it's my pleasure and me too - thanks a lot! | 17:20 |
| dhellmann | setuid : I think you use the -e option on the command line to pass the names of the variables to have it pass through | 17:20 |
| dhellmann | wait, sorry, wrong option | 17:20 |
| dhellmann | wrong type of env :-) | 17:20 |
| setuid | dhellmann: I've put the passenv bits in tox.ini, but they're... not making it through all the sub-plugins tox calls | 17:20 |
| dhellmann | ugh | 17:21 |
| setuid | this feels a lot like how I had to do it with Maven, using MAVEN_OPTS, java -Dhttp.proxy and updating /etc/maven/settings.xml to make it all work | 17:21 |
| setuid | some plugins respect it, others don't, it's... painful | 17:21 |
| dhellmann | what parts aren't respecting it? | 17:22 |
| setuid | That's what I'm trying to bisect, which of the various dozens of pip modules is dying. It looks like 'requests', which sounds like core python, but it successfully builds hundreds of prior docs via sphinx-build | 17:22 |
| setuid | import heisenbug | 17:23 |
| dhellmann | so it's failing to build the environment because pip can't install something? | 17:23 |
| setuid | pip appears to have all the deps it needs, and I can build the docs when I'm using a non-proxied connection | 17:23 |
| dhellmann | so sphinx is dying? | 17:24 |
| setuid | Once I connect behind the proxy, it fails to get https://git.openstack.org/ URIs | 17:24 |
| setuid | One sec, I'll get a pastebin | 17:24 |
| dhellmann | I'm trying to understand which step is actually the issue | 17:24 |
| setuid | https://paste.debian.net/hidden/0689da28/ | 17:25 |
| dhellmann | that site wants me to install some sort of cert | 17:25 |
| dhellmann | is that normal? | 17:25 |
| setuid | No idea, it's debian's pastebin | 17:26 |
| dhellmann | safair wants to install a cert that seems to come from apple | 17:26 |
| dhellmann | let me try chrome | 17:26 |
| dhellmann | ok, so that's sphinx dying | 17:27 |
| dhellmann | it's hard to tell what input file it's reading there, though | 17:27 |
| setuid | It's always the same one... block-storage/block-storage-sample-configuration-files | 17:28 |
| dhellmann | where do you get that from? the console output? | 17:31 |
| setuid | Yes | 17:31 |
| dhellmann | which guide is that in? | 17:31 |
| dhellmann | I'm checking out the source to look at it | 17:31 |
| dhellmann | config ref? | 17:31 |
| setuid | Looks like it... but syntactically, it's fine. It works when I'm not behind a proxy. | 17:32 |
| dhellmann | ok, the remote-code-block directive is implemented in that tree in doc/source/config-reference/source/ext/remote.py | 17:33 |
| dhellmann | and that just calls requests directly | 17:33 |
| dhellmann | I wonder if it needs to do something to enable the proxy? | 17:33 |
| setuid | It's odd, because the error implies the host is invalid (it isn't), but the python dump is '.VerifiedHTTPSConnection object...', and then it times out | 17:33 |
| setuid | but requests is straight python, which should read the environment where i've exported the vars | 17:34 |
| setuid | requests and urrlib that is | 17:34 |
| *** skelso has quit IRC | 17:34 | |
| setuid | This is where my python-fu starts to fall off ;) | 17:35 |
| dhellmann | you could modify the module to print out the environment variables right before it calls requests to ensure that they're making it through | 17:35 |
| setuid | requests.py? | 17:36 |
| dhellmann | I've never had to deal with proxies before | 17:36 |
| setuid | Or something in the doc tree? | 17:36 |
| dhellmann | ext/remote.py | 17:36 |
| dhellmann | in the doc tree | 17:36 |
| setuid | I could just force it with: os.environ['http_proxy'] = ... | 17:36 |
| setuid | Ok, one sec, trying that now | 17:36 |
| dhellmann | http://docs.python-requests.org/en/master/user/advanced/#proxies does seem to imply that if you set HTTP_PROXY it will pick it up | 17:36 |
| dhellmann | oh, though it does also imply there may be a different proxy for https | 17:36 |
| *** skelso has joined #openstack-doc | 17:37 | |
| dhellmann | are you setting both variables? | 17:37 |
| setuid | I'm setting both, though they're lowercase | 17:37 |
| setuid | I'll set both cases | 17:37 |
| dhellmann | yeah | 17:37 |
| dhellmann | I would expect it to be case sensitive | 17:37 |
| setuid | I wonder why they'd go off and invent their own version, standard is lowercase | 17:38 |
| * dhellmann shrugs | 17:38 | |
| setuid | "Ours is not to reason why..." :D | 17:38 |
| dhellmann | I would have to look at the code to see if it does deal with both | 17:39 |
| dhellmann | if we're right, you could file a bug with the requests project | 17:39 |
| setuid | It's re-running now, we'll know shortly | 17:39 |
| setuid | Nope, dies in the same spot. I'll try asserting os.environ[] | 17:40 |
| *** notmyname has quit IRC | 17:43 | |
| setuid | Success | 17:43 |
| setuid | Forcing the proxy in remote.py worked | 17:44 |
| setuid | Let me adjust that to inherit any exported environment vars | 17:44 |
| *** notmyname has joined #openstack-doc | 17:44 | |
| dhellmann | cool | 17:50 |
| setuid | The jump between tools is _painful_. Now I've pushed past the issue with sphinx+proxies, it's dying because it claims the title directive in index.rst is missing an argument, but that same version of sphinx builds it on Linux, fails on Mac. | 17:55 |
| dhellmann | we usually focus on linux for builds; I can't imagine what the difference there might be | 17:55 |
| dhellmann | maybe a different version of docutils? | 17:55 |
| setuid | Linux dies here now: https://paste.debian.net/hidden/7ac399e6/ | 17:56 |
| setuid | I didn't even touch remote.py on the Linux box | 17:56 |
| setuid | this is like whack-a-mole now | 17:56 |
| dhellmann | oh, that's the point where the extension was dealing with errors fetching the content by reporting an error with no details | 17:58 |
| dhellmann | that means it didn't get a 200 back | 17:58 |
| dhellmann | you'd have to change the way it raises an exception to have it show more info | 17:58 |
| setuid | Wait, ok... so that's it eating the 404 to stable/mitaka | 17:58 |
| dhellmann | probably, yeah | 17:58 |
| setuid | I forgot to git apply my patch that fixes those refs, now it works | 17:58 |
| setuid | Chasing the wrong reason for the error | 17:59 |
| setuid | That's part of what makes me grit my teeth debugging python, it's not always clear what the real subsystem is that is throwing the execption | 17:59 |
| dhellmann | oh, this is just someone's bad practice of raising an exception without any text | 17:59 |
| dhellmann | do you need any mitaka guides not referenced from https://docs.openstack.org/mitaka/ ? | 18:00 |
| setuid | I'll run the missing check in a sec | 18:00 |
| dhellmann | because we did have a lot of the old docs online, they just weren't linked from any landing pages, but we fixed that this week | 18:00 |
| dhellmann | maybe that landed today? | 18:01 |
| dhellmann | anyway, we now have a page for every series we've ever released, and links to whatever built docs still exist | 18:01 |
| setuid | hrm, README.rst in mitaka-eol references tox targets that don't exist | 18:01 |
| setuid | 'checklinks' for example | 18:01 |
| *** khushbu has joined #openstack-doc | 18:11 | |
| *** khushbu has quit IRC | 18:15 | |
| *** annegentle has joined #openstack-doc | 18:29 | |
| dhellmann | asettle : I just saw your comments on the rst doc changes. I'm going to have to figure out how to change my emacs config to use a different precedent order for the headers it automatically gives me, I guess :-) | 18:34 |
| *** vijaykc4 has quit IRC | 18:36 | |
| *** skelso has quit IRC | 18:44 | |
| *** skelso has joined #openstack-doc | 18:47 | |
| setuid | dhellmann: Made some progress, but I think there's a subtle bug between versions of components on Mac vs. Linux, where it parses that 'title' directive wrong on Mac, but works on Linux. I'll try to figure out the stack of components used, and find the discrepancy | 19:01 |
| dhellmann | if it's python I wouldn't expect it to be platform dependent but I could see a version difference | 19:01 |
| setuid | Checking with yolk now | 19:08 |
| *** khushbu has joined #openstack-doc | 19:16 | |
| *** khushbu has quit IRC | 19:20 | |
| *** donghao has joined #openstack-doc | 19:30 | |
| *** skelso has quit IRC | 19:36 | |
| *** khushbu has joined #openstack-doc | 19:38 | |
| *** khushbu has quit IRC | 19:42 | |
| *** nicolasbock has quit IRC | 19:48 | |
| setuid | dhellmann: What file did you get this from? http://paste.openstack.org/show/616467/ | 19:50 |
| dhellmann | setuid : I have an afs client connected to the openstack domain | 19:51 |
| dhellmann | let me find the directions for setting that up | 19:51 |
| dhellmann | https://docs.openstack.org/infra/system-config/afs.html | 19:52 |
| setuid | ah, afs... you at Morgan? | 19:52 |
| dhellmann | morgan? no...? | 19:52 |
| setuid | Morgan Stanley, big users of afs.. I've helped on a project they started, but was taken over by ML -> BAC, and then we open sourced it, called "EFS" (Enterprise File System), globally distributed, change management and development/deployment environment | 19:53 |
| setuid | EFS was originallydesigned and built on top of AFS | 19:53 |
| dhellmann | no, no associate with them | 19:54 |
| setuid | gotcha | 19:54 |
| dhellmann | I almost went to CMU, where AFS was created, but didn't so can't even really claim that :-) | 19:54 |
| setuid | dhellmann: so the output in that pastebin, is the default build targets, no edits, yes? | 20:01 |
| setuid | I'm trying to see where guides are commented out (specifically liberty and later), so I have a full copy in local builds | 20:01 |
| dhellmann | that pastebin shows a list of what was in the series-specific directories on the web server yesterday when I ran the command | 20:02 |
| dhellmann | I can do it again in case it changed | 20:02 |
| dhellmann | it will have, since I added series landing pages for the other series | 20:03 |
| setuid | ok | 20:03 |
| setuid | Look at this: https://paste.debian.net/hidden/598df9c6/ | 20:03 |
| dhellmann | http://paste.openstack.org/show/616787/ | 20:03 |
| setuid | It claims the title:: directive is empty, but it contains: .. title:: OpenStack Installation Guide | 20:04 |
| *** KeithMnemonic1 has joined #openstack-doc | 20:05 | |
| dhellmann | which version of sphinx and docutils are you using? | 20:06 |
| *** khushbu has joined #openstack-doc | 20:07 | |
| *** KeithMnemonic2 has quit IRC | 20:08 | |
| setuid | docutils 0.13.1, sphinx 1.3.6 | 20:08 |
| setuid | On the Linux side, docutils is 0.12, sphinx is 1.3.6 | 20:09 |
| dhellmann | the mitaka cosntraints file says docutils 0.12 and sphinx 1.2.3, give those a try | 20:09 |
| dhellmann | in fact, try installing everything constrained | 20:09 |
| setuid | Hrm, ok, I just took the default versions when pip satisfied the deps, I'll downgrade | 20:10 |
| dhellmann | if you don't pass the constraints file, you won't have upper bounds | 20:10 |
| setuid | I meet the minimum, but apparently being "too" new, breaks too | 20:10 |
| dhellmann | it doesn't look like the manuals repo is set up to use constraints at mitaka | 20:10 |
| dhellmann | yeah | 20:10 |
| setuid | I'll take a stab at trying to assert specific versions in the deps, not >= | 20:11 |
| *** khushbu has quit IRC | 20:11 | |
| *** skelso has joined #openstack-doc | 20:13 | |
| dhellmann | setuid : use -c http://git.openstack.org/cgit/openstack/requirements/plain/upper-constraints.txt?h=stable/mitaka | 20:13 |
| dhellmann | you need to combine that with -r test-requirements.txt | 20:13 |
| setuid | s,stable/mitaka,mitaka-eol,' but yes, I'll use that | 20:14 |
| dhellmann | come to think of it, you might have to download the constraints list from that url and pass a filename | 20:14 |
| dhellmann | no, the requirements repo still has a stable/mitaka branch | 20:14 |
| dhellmann | we haven't gotten around to eoling that | 20:14 |
| setuid | Fails with 0.12/1.3.6 on Mac | 20:14 |
| dhellmann | this is probably the point at which I say "we don't build on Mac" :-/ | 20:15 |
| dhellmann | I don't know what's going on there. | 20:15 |
| dhellmann | it shouldn't be any different | 20:15 |
| setuid | HAhAhha, yes, I know. I'll have to make it portable though, Bloomberg builds everything on Mac | 20:15 |
| setuid | I'll figure it out, if it takes ripping out all the deps and starting from scratch again | 20:15 |
| dhellmann | ah, customers | 20:16 |
| setuid | They're the one driving the open case to address the missing eol docs gap, so it's one of my highest priorities | 20:16 |
| dhellmann | that pastebin I gave you is the docs that do still exist upstream. Can you be specific about which manual you need that is not there? | 20:17 |
| setuid | I'm just trying to complete a build of liberty-eol and mitaka-eol. Right now, liberty works, with my smallish patch to address the missing branch names in the URLs, mitaka fails because of these weird dep problems | 20:18 |
| dhellmann | ok | 20:19 |
| setuid | curl http://git.openstack.org/cgit/openstack/requirements/plain/upper-constraints.txt?h=stable/mitaka | while read line; do sudo pip install $line; done; | 20:21 |
| setuid | :) | 20:22 |
| openstackgerrit | Doug Hellmann proposed openstack/openstack-manuals master: add has_in_tree_htaccess flag to template generator https://review.openstack.org/488190 | 20:23 |
| *** khushbu has joined #openstack-doc | 20:28 | |
| *** khushbu has quit IRC | 20:32 | |
| *** donghao has quit IRC | 20:36 | |
| *** KeithMnemonic2 has joined #openstack-doc | 20:38 | |
| *** KeithMnemonic1 has quit IRC | 20:42 | |
| *** mihalis68 has joined #openstack-doc | 20:59 | |
| *** khushbu has joined #openstack-doc | 21:05 | |
| *** khushbu has quit IRC | 21:10 | |
| *** yamamoto_ has joined #openstack-doc | 21:13 | |
| *** rbowen has quit IRC | 21:14 | |
| *** yamamoto_ has quit IRC | 21:20 | |
| *** rbowen has joined #openstack-doc | 21:22 | |
| *** thorst has quit IRC | 21:25 | |
| *** thorst has joined #openstack-doc | 21:26 | |
| *** thorst_ has joined #openstack-doc | 21:27 | |
| *** thorst has quit IRC | 21:30 | |
| *** thorst_ has quit IRC | 21:32 | |
| *** thorst has joined #openstack-doc | 21:45 | |
| *** thorst has quit IRC | 21:50 | |
| amotoki | hi, after we started to use the new openstackdocstheme 1.16.0, the left toc tree looks weird like https://docs.openstack.org/neutron/latest/ | 21:58 |
| amotoki | only the first title has less indent, even though all titles should have a same indent level.... | 21:58 |
| amotoki | is there any ongoing fix for this? | 21:59 |
| *** khushbu has joined #openstack-doc | 22:03 | |
| *** yamamoto has joined #openstack-doc | 22:04 | |
| annegentle | amotoki hey, in our testing we only had a few pages to look at, what do you suggest? | 22:05 |
| amotoki | annegentle: have you looked at my example? https://docs.openstack.org/neutron/latest/ | 22:05 |
| annegentle | amotoki yes, what indent are you looking for? | 22:05 |
| amotoki | is it not suggested to have this kind of title structure? | 22:06 |
| *** edmondsw has quit IRC | 22:06 | |
| amotoki | in the left-top side toc-tree, you see Installation Guide, OpenStack Networking Guide, .... | 22:06 |
| amotoki | but only Installation Guide has a different indent level. | 22:07 |
| amotoki | is it only for me? | 22:07 |
| *** edmondsw has joined #openstack-doc | 22:07 | |
| *** khushbu has quit IRC | 22:07 | |
| amotoki | annegentle: ^ I would like to see all of Installation Guide, OpenStack Networking Guide, ... in the same left margin. | 22:08 |
| *** gouthamr has quit IRC | 22:10 | |
| *** gouthamr_ has joined #openstack-doc | 22:10 | |
| *** edmondsw has quit IRC | 22:12 | |
| *** thorst has joined #openstack-doc | 22:13 | |
| *** thorst has quit IRC | 22:14 | |
| dhellmann | amotoki : indeed, that is a bug | 22:20 |
| *** rcernin has quit IRC | 22:26 | |
| jungleboyj | dhellmann: Are the docs that were at https://docs.openstack.org/newton/admin/ going to come back at some point? | 22:26 |
| jungleboyj | And for Ocata, etc. | 22:26 |
| *** catintheroof has quit IRC | 22:27 | |
| dhellmann | jungleboyj : I don't think for newton | 22:28 |
| *** skelso has quit IRC | 22:28 | |
| dhellmann | I'd have to go look at the spec to see what was said about the admin guide and why it was ok to remove it | 22:28 |
| openstackgerrit | Doug Hellmann proposed openstack/openstackdocstheme master: fix alignment of toc tree items in sidebar https://review.openstack.org/488222 | 22:31 |
| dhellmann | amotoki, annegentle : ^^ | 22:32 |
| jungleboyj | dhellmann: Ok, I will look at the spec. When we cut Pike though, there will be a new https://docs.openstack.org/pike/admin/ (Or something like that) based on content pulled from https://docs.openstack.org/cinder/latest ? | 22:32 |
| dhellmann | jungleboyj : that's right | 22:32 |
| dhellmann | when you start landing patches on the branch it will build docs with the series name in the URL | 22:32 |
| dhellmann | so that first patch to update .gitreview will do it | 22:33 |
| jungleboyj | Ok. So we can then backport fixes to that branch? | 22:33 |
| dhellmann | to pike? | 22:34 |
| jungleboyj | dhellmann: Right. | 22:34 |
| dhellmann | yeah, you can definitely backport doc fixes | 22:35 |
| jungleboyj | dhellmann: Cool, is there a process for generating the things that are automatically generated at the time we cut the release? | 22:35 |
| dhellmann | I think the admin guide was one that wasn't version-specific, and that's why it disappeared completely when we removed it from master | 22:36 |
| dhellmann | the release scripts that create the branch also propose some patches to set it up (.gitreview, tox.ini, release notes, etc.) | 22:36 |
| dhellmann | they do a best effort to make a patch that works, but you should plan to take over the patches and fix them if they're not right | 22:36 |
| *** gouthamr_ has quit IRC | 22:37 | |
| dhellmann | that mostly happens with the reno update | 22:37 |
| *** suyog has joined #openstack-doc | 22:37 | |
| jungleboyj | I had that question yesterday for the parts of the CLI documentation that are automatically generated. | 22:38 |
| jungleboyj | dhellmann: Ok, I will plan time around the end of Pike to make sure everything comes out right. | 22:38 |
| dhellmann | that'll be good | 22:38 |
| *** KeithMnemonic1 has joined #openstack-doc | 22:38 | |
| *** openstack has joined #openstack-doc | 22:45 | |
| *** med_ is now known as Guest13936 | 22:45 | |
| *** dustins has quit IRC | 22:45 | |
| *** dustins has joined #openstack-doc | 22:45 | |
| *** dustins has quit IRC | 22:45 | |
| *** mriedem is now known as mriedem_afk | 22:46 | |
| *** KeithMnemonic2 has joined #openstack-doc | 22:46 | |
| *** KeithMnemonic1 has quit IRC | 22:49 | |
| *** khushbu has joined #openstack-doc | 22:50 | |
| *** annegentle has quit IRC | 22:53 | |
| *** gouthamr has joined #openstack-doc | 22:53 | |
| *** khushbu has quit IRC | 22:54 | |
| *** thorst has joined #openstack-doc | 23:15 | |
| *** thorst has quit IRC | 23:21 | |
| *** sdague has quit IRC | 23:21 | |
| *** edmondsw has joined #openstack-doc | 23:37 | |
| *** edmondsw has quit IRC | 23:41 | |
| *** thorst has joined #openstack-doc | 23:46 | |
| *** catintheroof has joined #openstack-doc | 23:52 | |
| *** thorst has quit IRC | 23:55 | |
| *** thorst has joined #openstack-doc | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!