2020-07-03T02:54:09  *** okurz_ is now known as okurz
2020-07-03T07:07:34  <fcrozat> hmm, looks like mirrorbrain scan didn't notice free.fr aka http://opensuse.mirrors.proxad.net/ being up to date..
2020-07-03T07:07:47  <fcrozat> http://opensuse.mirrors.proxad.net/opensuse/distribution/leap/15.2/iso/?F=1 is right now, though
2020-07-03T07:29:29  <lkocman> fcrozat, I'll email mirror@ again
2020-07-03T07:29:38  <lkocman> should have all mirror admins on it
2020-07-03T07:30:31  <lkocman> Hello team, I need to ask for an advice, if following is a preferred way to handle the issue. Seems like we'd like to point Leap (15.1, 15.2 ...) ports sources to the sources of the main repository.
2020-07-03T07:30:48  <lkocman> https://github.com/yast/skelcd-control-openSUSE/pull/187 suggested to do redirect on apache
2020-07-03T07:31:23  <lkocman> The way how I'd usually handle it in "another distribution" is to disable publish for given repo, and make it a symlink. Would apache redirect be more of openSUSE way to do this?
2020-07-03T07:31:53  <lkocman> pjessen, bmwiedemann ^ what do you guys think?
2020-07-03T07:34:11  <lkocman> fcrozat, ah so it's a mirrorbrain issue not mirror issue. Let me just create ticket for admin so we have a tracker for it. I'll also add it to retro
2020-07-03T07:43:22  <lkocman> no response oky I take it as redirect is fine then
2020-07-03T11:12:15  *** okurz_ is now known as okurz
2020-07-03T11:50:25  <witek> hi, who's responsible for metrics.o.o?
2020-07-03T11:50:45  <witek> could I get sudoers permissions there, please?
2020-07-03T11:52:11  <cboltz> I'm not responsible for it, but I can give you permissions ;-)
2020-07-03T11:53:24  <witek> appreciate a lot :)
2020-07-03T11:53:44  <darix> witek: why do you need them?
2020-07-03T11:54:13  <witek> I want to extend the statistics available in the Grafana dashboard
2020-07-03T11:54:46  <witek> want to figure out how https://github.com/jberry-suse/openSUSE-release-tools/tree/master/metrics/access is set up
2020-07-03T11:56:20  <cboltz> witek: looks like we don't have a salt role for metrics yet, so as a quick (and hopefully temporary) solution,   ssh root@metrics.i.o.o
2020-07-03T11:56:52  <cboltz> I won't complain if you create a salt role so that we can handle sudo permissions via a "metrics-admins" group ;-)
2020-07-03T11:57:48  <darix> or just dont do any sudo and do everything via salt :P
2020-07-03T11:58:22  <witek> cboltz: could you give me reference, where similar is done?
2020-07-03T11:59:39  <cboltz> for example pillar/role/web_download.sls and salt/role/web_download.sls
2020-07-03T12:00:03  <cboltz> the new role also needs to be added to the pillar/id/ file
2020-07-03T14:07:41  <lkocman> One very stupid question
2020-07-03T14:07:59  <lkocman> are we discriminating people with first name "Dick" on the opensuse-factory and other mailing lists in general?
2020-07-03T14:08:27  <lkocman> I did sent a post to factory to respond to Dick Waite ... starting by "Hello Dick" and the message was rejected
2020-07-03T14:08:49  <lkocman> since there was no message from the server ... I suppose that's the only possible reason
2020-07-03T14:09:05  <lkocman> can we please not censor common US first names?
2020-07-03T14:26:25  <lkocman> Meanwhile I'll skip his name from the email
2020-07-03T14:29:16  <cboltz> this question isn't stupid ;-)
2020-07-03T14:29:55  <cboltz> and I wouldn't be surprised if we indeed have a rule blocking mails with "Dick" because that word is, well, not only used as a first name ;-)
2020-07-03T14:30:09  <darix> lkocman: can you imagine why it was blocked maybe?
2020-07-03T14:30:15  <cboltz> pjessen: can you please check if we indeed have such a rule?
2020-07-03T14:36:38  <lkocman> well yeah, but then it's a valid name
2020-07-03T14:36:54  <lkocman> and I feel like he wouldn't be happy if his emails would be rejected because of e.g. signature
2020-07-03T14:38:15  <lkocman> We could also Wilhelm or Willie? I hope you see my point
2020-07-03T14:38:46  <lkocman> *We could also block
2020-07-03T14:39:28  <cboltz> I fully agree - if we indeed have such a rule, let's drop it (and take the risk of mails using the word not as a first name ;-)
2020-07-03T14:39:46  <lkocman> yup
2020-07-03T14:40:00  <lkocman> let me check email header meanwhile perhaps there is X-something-reason
2020-07-03T15:14:42  *** lurchi_ is now known as lurchi__
2020-07-03T15:15:38  *** lurchi__ is now known as lurchi_
2020-07-03T15:33:21  <pjessen> cboltz: yes, in general people called Dick are not welcome :-)
2020-07-03T15:39:34  <cboltz> Can you please drop that rule? ;-)
2020-07-03T15:41:53  <cboltz> (or maybe only deny lowercase "dick"?)
2020-07-03T16:03:38  <lcp28> cboltz: mind taking a look https://gitlab.infra.opensuse.org/infra/salt/-/merge_requests/416
2020-07-03T16:04:08  <lcp28> the new design is in the default branch, but it doesn't currently work, since gh-pages can't handle some stuff added there
2020-07-03T16:04:36  <lcp28> (I could fix it, but it seems easier just to add it to our jekyll pipeline)
2020-07-03T16:09:47  <lcp28> thanks!
2020-07-03T16:10:16  <lcp28> I guess that still leaves proxy and dns
2020-07-03T16:10:54  <cboltz> right, just tell me when you have deployed the salt changes on jekyll.i.o.o
2020-07-03T16:11:08  <cboltz> I'll do the DNS change then
2020-07-03T16:11:44  <cboltz> the haproxy config will be ready in a minute, so if you adjust your /etc/hosts, you can test before we switch over
2020-07-03T16:11:45  <lcp28> gotta wait for deployment in gitlab, don't I
2020-07-03T16:12:00  <cboltz> right, you'll have to wait for the pipeline
2020-07-03T16:14:33  <cboltz> (and also for the cronjob at the full hour)
2020-07-03T16:14:57  <lcp28> hm, yeah
2020-07-03T16:15:02  <lcp28> I can run it manually
2020-07-03T16:15:13  <lcp28> I did that for doug on 15.2 release
2020-07-03T16:17:32  <cboltz> haproxy config done
2020-07-03T16:18:25  *** lurchi_ is now known as lurchi__
2020-07-03T16:21:05  <cboltz> as a sidenote - did you see my reply about the sql query some days ago? (not sure if you were really online - "lcp" was, but "lcp$number" wasn't ;-)
2020-07-03T16:22:08  <lcp28> yeah, normal lcp is dead
2020-07-03T16:22:19  <lcp28> lcp number is when I am online recently
2020-07-03T16:22:41  <lcp28> I will figure out normal lcp soonish >:D
2020-07-03T16:22:48  <cboltz> ok, let me dig out what I wrote ;-)
2020-07-03T16:25:15  <cboltz> lcp28: https://paste.opensuse.org/0bd03308
2020-07-03T16:26:19  <lcp28> alright, yeah, that makes sense
2020-07-03T16:27:06  <cboltz> just wondering - what's the number of expected results from that query?
2020-07-03T16:27:17  <cboltz> (IIRC I got an empty result when testing one of them ;-)
2020-07-03T16:27:38  <lcp28> they are empty when they finished once before
2020-07-03T16:28:33  <lcp28> and there is no easy answer to how many, because that can vary, since the answer is events that happened on the server since the last cache
2020-07-03T16:30:38  <cboltz> do you at least know a (somewhat typical) range for the number? Like 10, 100, 1k, 10k, 100k, 1M?
2020-07-03T16:32:36  <lcp28> no more than 100k for sure
2020-07-03T16:33:08  <lcp28> I would say for the current server it would probably even be under 10k
2020-07-03T16:33:59  <cboltz> that makes the >1M rows sequence scan (I guess that's "full table scan" in mysql terms?) interesting[tm]
2020-07-03T16:39:49  <lcp28> yeah, honestly I have no clue what happened and why
2020-07-03T16:40:43  <cboltz> the query is interesting[tm] and I'm afraid it's beyond my (postgre)sql knownledge
2020-07-03T16:41:05  <cboltz> so either ask upstream or a postgres expert to have a look at it ;-)
2020-07-03T16:41:16  <lcp28> yeah, I asked upstream to see if they can help me
2020-07-03T16:42:22  <lcp28> cboltz: mentoring built in jekyll, and should be in the right place
2020-07-03T16:44:21  <lcp28> replacing hosts entry with proxy ip shows the working site
2020-07-03T16:44:32  <cboltz> :-)
2020-07-03T16:44:40  <cboltz> I'll switch over the DNS entry
2020-07-03T16:45:57  <cboltz> done
2020-07-03T16:46:53  <cboltz> I see several "Please wait for loading the projects..." lines - is this a known issue?
2020-07-03T16:47:28  <lcp28> sadly yes
2020-07-03T16:49:45  <cboltz> sounds like your TODO list doesn't become empty ;-)
2020-07-03T16:49:58  <lcp28> apparently
2020-07-03T16:50:00  * cboltz also has an ever-full TODO list
2020-07-03T16:54:30  <pjessen> cboltz: no such rule - in fact, the only bounce from factory to dick.waite was due to html content and was 28jun
2020-07-03T16:55:45  <cboltz> also no bounce with "Dick" in the mail body?
2020-07-03T16:56:02  <pjessen> lkocman: we rejected one message from you due to html content. basiscally, our lists welcome any tom, dick and harry.
2020-07-03T16:56:23  <pjessen> cboltz: nope, mlmmj can't check the body
2020-07-03T16:56:40  <cboltz> right, now that you say it, I remember that ticket ;-)
2020-07-03T17:01:54  <lkocman> pjessen, ufff
2020-07-03T17:02:18  <lkocman> ufufuf
2020-07-03T17:02:35  <lkocman> I think I must have used the office365 client, as My evolution is set up to send emails in plain text
2020-07-03T17:02:36  <lkocman> oky
2020-07-03T17:02:43  <lkocman> thanks for double checking!
2020-07-03T17:03:11  <cboltz> lkocman: don't worry too much - I'm sure some people (hopefully including you) enjoyed reading this discussion ;-)
2020-07-03T17:20:34  <pjessen> lkocman: no prob, happens all the time.
2020-07-03T17:32:08  <lkocman> hehe :-)
2020-07-03T17:32:50  *** lurchi__ is now known as lurchi_
2020-07-03T19:54:21  <lcp28> cboltz: I might try to rewrite the countdown.o.o renderer to python3, and make it use gettext instead of the current system
2020-07-03T19:55:08  <lcp28> I don't particularly like dealing with python, but oh well
2020-07-03T19:55:20  <cboltz> I won't stop you ;-)
2020-07-03T19:56:01  <cboltz> actually I like python (so if you have any problems, feel free to ask) - but I'm afraid there are too many things on my TODO list that are more important
2020-07-03T19:56:47  <cboltz> oh, and please make sure to tell me (or update the salt code) if you change the requirements of the script ;-)
2020-07-03T19:58:53  *** lurchi_ is now known as lurchi__
2020-07-03T20:18:35  *** lurchi__ is now known as lurchi_
2020-07-03T21:25:20  *** lurchi_ is now known as lurchi__
2020-07-03T21:33:09  *** lurchi__ is now known as lurchi_
2020-07-03T23:09:18  *** lurchi_ is now known as lurchi__
2020-07-03T23:13:38  *** lurchi__ is now known as lurchi_