2020-01-28T00:37:14  *** cboltz <cboltz!~cb@opensuse/member/Cboltz> has quit IRC ()
2020-01-28T00:39:39  *** ldevulder_ <ldevulder_!~ldevulder@176.167.97.11> has joined #opensuse-admin
2020-01-28T00:41:15  *** ldevulder <ldevulder!~ldevulder@176.167.124.200> has quit IRC (Ping timeout: 268 seconds)
2020-01-28T01:22:24  *** malcolmlewis <malcolmlewis!~malcolmle@opensuse/member/malcolmlewis> has joined #opensuse-admin
2020-01-28T01:50:18  *** boombatower <boombatower!~boombatow@drupal.org/user/214218/view> has quit IRC (Quit: Konversation terminated!)
2020-01-28T03:48:48  *** okurz_ <okurz_!~okurz@ppp-46-244-214-116.dynamic.mnet-online.de> has joined #opensuse-admin
2020-01-28T03:51:25  *** okurz <okurz!~okurz@ppp-46-244-216-37.dynamic.mnet-online.de> has quit IRC (Ping timeout: 272 seconds)
2020-01-28T03:51:25  *** okurz_ is now known as okurz
2020-01-28T07:31:44  *** jadamek <jadamek!~julienA@176.177.116.167> has joined #opensuse-admin
2020-01-28T07:42:49  *** dddh <dddh!~dddh@pdpc/supporter/active/dddh> has quit IRC (Excess Flood)
2020-01-28T07:43:49  *** dddh <dddh!~dddh@pdpc/supporter/active/dddh> has joined #opensuse-admin
2020-01-28T07:56:42  *** ldevulder_ is now known as ldevulder
2020-01-28T08:01:49  *** dddh <dddh!~dddh@pdpc/supporter/active/dddh> has quit IRC (Excess Flood)
2020-01-28T08:02:16  *** dddh <dddh!~dddh@pdpc/supporter/active/dddh> has joined #opensuse-admin
2020-01-28T08:23:08  *** ldevulder <ldevulder!~ldevulder@176.167.97.11> has quit IRC (Quit: Leaving)
2020-01-28T08:23:20  *** ldevulder <ldevulder!~ldevulder@176.167.97.11> has joined #opensuse-admin
2020-01-28T08:35:29  *** kl_eisbaer <kl_eisbaer!kl_eisbaer@pdpc/supporter/base/kl-eisbaer> has joined #opensuse-admin
2020-01-28T08:48:55  *** matthias_bgg <matthias_bgg!~mbgg@37.223.145.31> has joined #opensuse-admin
2020-01-28T09:01:46  *** marxin <marxin!marxin@nat/suse/x-xqmhffettllzpsxn> has joined #opensuse-admin
2020-01-28T10:23:19  *** ldevulder_ <ldevulder_!~ldevulder@176.167.123.76> has joined #opensuse-admin
2020-01-28T10:26:57  *** ldevulder <ldevulder!~ldevulder@176.167.97.11> has quit IRC (Ping timeout: 260 seconds)
2020-01-28T12:14:16  <bmwiedemann> note: new widehat hardware is about to get shipped to Nuremberg.
2020-01-28T12:58:39  <pjessen> nice
2020-01-28T13:46:04  *** ldevulder_ is now known as ldevulder
2020-01-28T13:53:13  *** kl_eisbaer <kl_eisbaer!kl_eisbaer@pdpc/supporter/base/kl-eisbaer> has quit IRC (Ping timeout: 245 seconds)
2020-01-28T13:53:40  *** agraul <agraul!agraul@opensuse/member/agraul> has quit IRC (Ping timeout: 248 seconds)
2020-01-28T13:54:15  *** enavarro_suse <enavarro_suse!enavarro_s@nat/suse/x-xlaljdawrhnyhdhb> has quit IRC (Ping timeout: 258 seconds)
2020-01-28T13:58:54  *** jadamek <jadamek!~julienA@176.177.116.167> has quit IRC (Quit: Leaving)
2020-01-28T14:02:37  *** jadamek <jadamek!~julienA@176.177.116.167> has joined #opensuse-admin
2020-01-28T14:06:30  *** enavarro_suse <enavarro_suse!enavarro_s@nat/novell/x-yinkwwnzdagoabxp> has joined #opensuse-admin
2020-01-28T14:06:41  *** agraul <agraul!agraul@opensuse/member/agraul> has joined #opensuse-admin
2020-01-28T14:07:14  *** kl_eisbaer <kl_eisbaer!kl_eisbaer@pdpc/supporter/base/kl-eisbaer> has joined #opensuse-admin
2020-01-28T14:08:34  *** lkocman <lkocman!lkocman@nat/novell/x-epzcbecpaxjvzmfq> has joined #opensuse-admin
2020-01-28T14:48:04  *** foursixnine <foursixnine!foursixnin@nat/novell/x-ktuccbixhnndisgb> has quit IRC (Quit: foursixnine)
2020-01-28T15:20:44  *** dheidler <dheidler!dheidler@nat/novell/x-xvlioctonfvtgvkr> has quit IRC (Quit: WeeChat 2.6)
2020-01-28T15:27:21  *** dheidler <dheidler!dheidler@nat/novell/x-zemdrbmfvsfdqdgy> has joined #opensuse-admin
2020-01-28T15:31:49  *** dheidler <dheidler!dheidler@nat/novell/x-zemdrbmfvsfdqdgy> has quit IRC (Client Quit)
2020-01-28T15:50:33  *** dheidler <dheidler!dheidler@nat/novell/x-hrkegzamicdbtxqh> has joined #opensuse-admin
2020-01-28T17:08:45  *** matthias_bgg <matthias_bgg!~mbgg@37.223.145.31> has quit IRC (Read error: Connection reset by peer)
2020-01-28T17:09:32  *** matthias_bgg <matthias_bgg!~mbgg@37.223.145.31> has joined #opensuse-admin
2020-01-28T17:20:55  *** kl_eisbaer <kl_eisbaer!kl_eisbaer@pdpc/supporter/base/kl-eisbaer> has left #opensuse-admin ("Leaving.")
2020-01-28T17:24:10  *** cboltz <cboltz!~cb@opensuse/member/Cboltz> has joined #opensuse-admin
2020-01-28T17:33:13  <cboltz> hmm, that's... interesting
2020-01-28T17:33:49  <cboltz> I wanted to add a comment to SR 767253, but when submitting it, I got "failed to create a comment"
2020-01-28T17:34:08  <cboltz> The screenshot looks interesting[tm]: https://paste.opensuse.org/cf57b25e
2020-01-28T17:34:28  <cboltz> basically I get the OBS main page in the comment frame :-/
2020-01-28T17:36:59  <cboltz> DimStar: sadly that ^^^ means my comment got lost - the short version is that I also think that it isn't an AppArmor problem
2020-01-28T17:37:50  <cboltz> one surprising thing is that (according to apache-dmesg.log) only a few profiles get loaded (ping, ghostscript and lsb_release), but the apparmor-profiles package has more
2020-01-28T17:39:31  <DimStar> found the culprit
2020-01-28T17:39:39  <DimStar> cboltz: thanks for looking into ittoo
2020-01-28T17:46:35  <DimStar> cboltz: just added my last findings to https://build.opensuse.org/request/show/757447
2020-01-28T17:47:13  *** ldevulder <ldevulder!~ldevulder@176.167.123.76> has quit IRC (Remote host closed the connection)
2020-01-28T17:47:20  <cboltz> ah, fun with nscd ;-)
2020-01-28T17:47:43  *** ldevulder <ldevulder!~ldevulder@176.167.123.76> has joined #opensuse-admin
2020-01-28T17:47:44  <cboltz> I've long replaced it with a real nameserver (unbound) on my laptop
2020-01-28T17:49:20  *** matthias_bgg <matthias_bgg!~mbgg@37.223.145.31> has quit IRC (Ping timeout: 265 seconds)
2020-01-28T17:51:54  *** ldevulder_ <ldevulder_!~ldevulder@176.167.19.235> has joined #opensuse-admin
2020-01-28T17:53:50  <DimStar> cboltz: yep, fun stuff.. and quiet lucky to have a test that caught the issue
2020-01-28T17:54:10  *** ldevulder <ldevulder!~ldevulder@176.167.123.76> has quit IRC (Ping timeout: 268 seconds)
2020-01-28T17:54:31  *** ldevulder__ <ldevulder__!~ldevulder@176.167.182.81> has joined #opensuse-admin
2020-01-28T17:55:18  <cboltz> indeed :-)
2020-01-28T17:55:45  <cboltz> any idea what could have caused the funny comment failure in OBS (see paste.o.o link above), or should I better ask the experts in the OBS channel?
2020-01-28T17:57:34  *** lkocman <lkocman!lkocman@nat/novell/x-epzcbecpaxjvzmfq> has quit IRC (Quit: lkocman)
2020-01-28T17:57:40  <DimStar> cboltz: I'd seen that already every now and then; usually I have to logout/login to make it work again
2020-01-28T17:57:42  *** lkocman <lkocman!lkocman@nat/novell/x-chmdmbyafukcynpx> has joined #opensuse-admin
2020-01-28T17:58:18  *** ldevulder_ <ldevulder_!~ldevulder@176.167.19.235> has quit IRC (Ping timeout: 268 seconds)
2020-01-28T17:59:14  <cboltz> did you already open a ticket for it?
2020-01-28T18:00:14  <DimStar> no...
2020-01-28T18:00:24  <DimStar> it's been a while that I saw it
2020-01-28T18:00:33  <DimStar> I thought this vanished
2020-01-28T18:01:02  <cboltz> as much as I like self-fixing bugs - I'm afraid they are very rare...
2020-01-28T18:03:14  <cboltz> I'll open a ticket ;-)
2020-01-28T18:03:39  <DimStar> not that it would have fixed itself.. but with so much code changes on obs lately, it's not impossible it might have gone
2020-01-28T18:04:27  <DimStar> might also 'just' happen when the services are restarted and your user session is logged out
2020-01-28T18:09:58  <cboltz> maybe, but that would be very unlucky timing - I logged in 5 minutes before hitting the error
2020-01-28T18:10:07  <cboltz> anyway - reported as https://github.com/openSUSE/open-build-service/issues/9009
2020-01-28T18:17:11  *** lkocman <lkocman!lkocman@nat/novell/x-chmdmbyafukcynpx> has quit IRC (Quit: lkocman)
2020-01-28T19:01:36  *** boombatower <boombatower!~boombatow@drupal.org/user/214218/view> has joined #opensuse-admin
2020-01-28T19:03:26  *** matthias_bgg <matthias_bgg!~mbgg@95.169.235.131> has joined #opensuse-admin
2020-01-28T19:28:18  *** lkocman <lkocman!~lkocman@x5f732b8e.dyn.telefonica.de> has joined #opensuse-admin
2020-01-28T19:35:40  <lcp> cboltz: out of curiosity, how are lists & mlmmj hosted and who owns the database for them?
2020-01-28T19:37:11  *** lkocman <lkocman!~lkocman@x5f732b8e.dyn.telefonica.de> has quit IRC (Quit: lkocman)
2020-01-28T19:47:36  <cboltz> on baloo.i.o.o
2020-01-28T19:47:52  <cboltz> AFAIK there isn't a database, just plain files
2020-01-28T19:50:04  <lcp> yeah, I just read how mlmmj stores stuff, pretty cool for parsing manually I would imagine
2020-01-28T19:50:17  <lcp> not so great for speed for thousands of people
2020-01-28T19:50:41  <cboltz> well, those thousands of people don't subscribe and unsubscribe every 5 minutes ;-)
2020-01-28T19:51:34  <cboltz> and the subscribers list is split by (IIRC) first letter of the mail address
2020-01-28T19:51:51  <cboltz> so in the end it works pretty well
2020-01-28T19:52:13  <cboltz> besides that - not too long ago, I had to debug a mysql database where a query like
2020-01-28T19:52:26  <cboltz>     select foo from bar where baz = 42 limit 1;
2020-01-28T19:52:35  <cboltz> took more than 30 seconds
2020-01-28T19:53:06  <cboltz> any guesses why and what the fix was? ;-)
2020-01-28T19:53:39  <lcp> database was too slow
2020-01-28T19:53:57  <cboltz> yes, obviously, but why?
2020-01-28T19:55:11  <lcp> did it run out of space?
2020-01-28T19:55:19  <cboltz> no
2020-01-28T19:57:30  <cboltz> maybe I should give you a hint: it was a quite simple, but still "typical" problem everybody who works with databases should know
2020-01-28T19:59:16  <lcp> was it corrupted?
2020-01-28T19:59:34  <cboltz> no ;-)
2020-01-28T19:59:44  <cboltz> technically everything was ok
2020-01-28T20:01:26  <lcp> I clearly won't get it, my best guess was it run out of resources 🤷‍♂️
2020-01-28T20:01:41  <cboltz> ok, another hint: for debugging, I did
2020-01-28T20:01:48  <cboltz>     explain select foo from bar where baz = 42 limit 1;
2020-01-28T20:02:52  <cboltz> and that made the problem _very_ obvious
2020-01-28T20:10:14  <cboltz> any other ideas, or should I tell you what was wrong?
2020-01-28T20:11:17  <lcp> idk, a loop or something >:D
2020-01-28T20:11:32  <cboltz> no, you are thinking too complicated ;-)
2020-01-28T20:11:44  <cboltz> it's a simple query, which means not too much can go wrong
2020-01-28T20:12:08  <cboltz> the "explain select" told me "using index: none"
2020-01-28T20:12:15  <cboltz> for a 2 GB table ;-)
2020-01-28T20:12:23  <lcp> oh god
2020-01-28T20:12:35  <klein> oh... I just arrived, I was bet on lack of indexes
2020-01-28T20:12:36  <klein> :-(
2020-01-28T20:12:36  <lcp> yeah, I was thinking way too big
2020-01-28T20:12:51  <klein> indexes <== always none, or too much
2020-01-28T20:13:14  <cboltz> I don't think there's something like "too many indexes" ;-)
2020-01-28T20:13:34  <cboltz> sure, in theory you could loose a bit of disk space and maybe 1% performance
2020-01-28T20:13:39  <lcp> you can never have too many, associate everything with everything
2020-01-28T20:13:49  <klein> I saw developers saying "Maybe we should put all the attributes on indexes, than any query will run fast"
2020-01-28T20:13:52  <cboltz> but a missing index makes the difference between 30 seconds and 0.001 seconds ;-)
2020-01-28T20:14:15  <klein> I stoped listening and ran to the hills after that
2020-01-28T20:14:21  <klein> :-)
2020-01-28T20:14:24  <lcp> if you associate everything with everything, even if wrong, your query will run very fast and return wrong stuff
2020-01-28T20:14:35  <lcp> but speed matters more than correct response anyway
2020-01-28T20:14:59  <cboltz> indexes should (and will) never give you wrong results ;-)
2020-01-28T20:16:06  <lcp> sounds more like a challenge to me
2020-01-28T20:17:27  <cboltz> well, all my "create index" queries I did over the last years were always helpful, and never caused wrong results
2020-01-28T20:17:32  <klein> lol... I can see the ticket already: "Please create a new vm on the Heroes network so we can test some database concepts"
2020-01-28T20:17:48  <cboltz> and if you know my talent to find bugs (especially obscure ones) that means something ;-)
2020-01-28T20:17:52  <lcp> we need mongodb somewhere just for breaking
2020-01-28T20:18:02  <klein> mongodb?
2020-01-28T20:18:05  * klein run to the hills
2020-01-28T20:18:09  <lcp> the older the better, without that new license bs
2020-01-28T20:18:11  <cboltz> I thought we are talking about databases ;-)
2020-01-28T20:19:03  <klein> *but* talking about Databases, Darix and me will probably upgrade our PostgreSQL to v12
2020-01-28T20:19:23  <klein> we discussed that last week, but had no time to do it yet
2020-01-28T20:20:49  <lcp> I can see matrix machine was added to salt repo, I might as well start working on that already :D
2020-01-28T20:21:29  <klein> kernel upgrade... brb
2020-01-28T20:21:47  <cboltz> just checked - looks like the initial highstate wasn't done yet
2020-01-28T20:25:05  <cboltz> even better - needs a small fix in our salt repo
2020-01-28T20:25:09  <cboltz> (any guesses?)
2020-01-28T20:25:27  * cboltz already starts to fix it
2020-01-28T20:25:35  <klein> pillars?
2020-01-28T20:25:38  *** matthias_bgg <matthias_bgg!~mbgg@95.169.235.131> has quit IRC (Ping timeout: 240 seconds)
2020-01-28T20:25:42  <cboltz> yes, right direction
2020-01-28T20:25:58  <klein> lavk of that new structure, or roles?
2020-01-28T20:26:04  <klein> *lack
2020-01-28T20:26:17  <cboltz> no, more basic ;-)
2020-01-28T20:26:42  <klein> country, virt_cluster and that things?
2020-01-28T20:26:50  <cboltz> no, even more basic ;-)
2020-01-28T20:27:16  <klein> they are not there at all?
2020-01-28T20:27:24  <klein> I mean, the pillar file
2020-01-28T20:27:43  <klein> we didnt accepted the minion key?
2020-01-28T20:28:04  <cboltz> I just accepted the key
2020-01-28T20:28:14  <cboltz> it's really a bug in what we have in our salt repo
2020-01-28T20:28:38  * klein gives up
2020-01-28T20:29:51  <cboltz> ok, then please review https://gitlab.infra.opensuse.org/infra/salt/merge_requests/323 ;-)
2020-01-28T20:30:48  <klein> ok, i am in my cellphone, waiting btrfs maintenance script after reboot
2020-01-28T20:31:07  <klein> will check as soon as my machine is back, 3mpre minutes :-(
2020-01-28T20:31:10  <lcp> that was a simple mistake to make
2020-01-28T20:31:23  <lcp> you do expect to write a dot after openSUSE after a while
2020-01-28T20:36:25  <klein> ohhh.. I see... did that many times already haha
2020-01-28T20:38:22  <cboltz> sounds like we have a volunteer to write a little CI check ;-)
2020-01-28T20:38:37  <klein> cboltz: and this pipeline taking forever to run, and, testing almost nothing huh? I will change that!
2020-01-28T20:39:58  <klein> the simple test for this specific case should be something like "if you have a sls file in pillars/id and it has "." == fail, and, if there is any file there that doesn't have a correspondent accepted key, maybe fail too?
2020-01-28T20:40:43  <cboltz> cd pillar/id && find -not -name '*_infra_opensuse_org.sls' -not -name README.md
2020-01-28T20:40:48  <cboltz> and fail if it finds anything ;-)
2020-01-28T20:41:14  <klein> nice idea
2020-01-28T21:15:14  <lcp> oh right, I might want to plug matrix.o.o into ldap if that's possible
2020-01-28T21:15:41  <lcp> I just need to actually submit the packages to openSUSE:Infrastructure because it seems we don't build with ldap support
2020-01-28T21:16:18  <lcp> question is, that would be accessible out of vpn, is that even a good idea
2020-01-28T21:17:28  <cboltz> if you ask that way, the answer is typically "no" ;-)  - but I have to admit that I don't really know matrix, and therefore can't really judge
2020-01-28T21:18:39  <klein> What is the Matrix ?
2020-01-28T21:18:54  <lcp> the movie or the service >:D
2020-01-28T21:18:55  <klein> (please, don't send me a link for any Morpheus image)
2020-01-28T21:18:58  <klein> hahahahhaha
2020-01-28T21:19:00  <klein> the service
2020-01-28T21:19:32  <lcp> it's basically a service that currently competes with ircv3 and rocketchat for developer choice dominance
2020-01-28T21:19:44  <lcp> well, it's a protocol, as those 2 are
2020-01-28T21:21:00  <lcp> the gist is that we have an instance so we have an ability to bridge our communication channels together, so we can use them as if they were one
2020-01-28T21:22:31  <lcp> and my wondering about the current ldap setup for freeipa, Conan Kudo is hoping to use that for external stuff too in terms of authentication, so idk if that's a good starting point (I would guess it's better than nothing though)
2020-01-28T21:22:40  <klein> so, this will act like a gateway for rocketchat and irc at the same time?
2020-01-28T21:23:28  <lcp> it will be both by itself a service that can be used and a set of bridges to different services, yes
2020-01-28T21:24:29  <lcp> if we decide to bridge anything with rocketchat, I can't imagine SUSE internal would be interested in bridging their instance and we don't have much community in rocket chat
2020-01-28T21:25:15  <lcp> we do however have community in discord, matrix, telegram and irc (duh, we are here rn), so that would make the most sense to bridge
2020-01-28T21:25:40  <lcp> also uyuni is on gitter, so that's under the consideration too
2020-01-28T21:28:11  <klein> ok, does matrix store my credentials for irc/rocketchat/discord/whatever ?
2020-01-28T21:28:21  <klein> or does this happens in the "client side" ?
2020-01-28T21:32:12  <lcp> the ldap ones? it stores only on matrix
2020-01-28T21:32:51  <lcp> if you login into freenode from matrix, matrix stores your freenode login/password
2020-01-28T21:33:01  <lcp> if you login into telegram, also stores that
2020-01-28T21:33:52  <lcp> well, it doesn't exactly store the ldap stuff, considering it asks ldap for that, but you know what I mean, it stores what it's given by you locally
2020-01-28T21:35:24  <lcp> you don't have to login into those bridges btw, it's optional
2020-01-28T21:41:47  <cboltz> klein: even with the pillar/id/ filename filed, I still get   Specified SLS 'id.matrix_infra_opensuse_org' in environment 'production' is not available on the salt master
2020-01-28T21:42:33  <cboltz> I already checked some things, restarted the saltmaster etc. - no change
2020-01-28T21:42:45  <cboltz> an additional pair of eyes would be helpful ;-)
2020-01-28T21:52:34  <klein> caches on that *nice* pygit storage?
2020-01-28T21:57:42  <cboltz> well, in theory the deploy job should have triggered a refresh, and I also tried to manually trigger it
2020-01-28T21:58:03  <cboltz> (and yes, I also "love" the pygit storage - life would be boring without it ;-)
2020-01-28T22:27:48  *** jadamek2 <jadamek2!~julienA@176.167.89.122> has joined #opensuse-admin
2020-01-28T22:31:08  *** jadamek <jadamek!~julienA@176.177.116.167> has quit IRC (Ping timeout: 245 seconds)
2020-01-28T22:35:21  <Pharaoh_Atem> cboltz: I found my wallet, but erm, trying to remember my passphrase so I can extend my expired key :<
2020-01-28T23:04:43  <cboltz> sounds like you are making progress ;-)
2020-01-28T23:08:29  *** jadamek2 <jadamek2!~julienA@176.167.89.122> has quit IRC (Quit: Leaving)
2020-01-28T23:13:10  <cboltz> klein: for whatever reason, /etc/salt/grains on minnie was missing
2020-01-28T23:13:32  <cboltz> fixed, and highstate done (which only changed some base system stuff)
2020-01-28T23:14:15  <cboltz> unfortunately I still get    Specified SLS 'id.matrix_infra_opensuse_org' in environment 'production' is not available on the salt master
2020-01-28T23:39:45  *** cboltz <cboltz!~cb@opensuse/member/Cboltz> has quit IRC (Ping timeout: 265 seconds)