|
=== Kiall_ is now known as Kiall |
|
[02:08] <cjohnston> /3/9 |
|
[02:34] <RAOF> I'm getting OOPS-e24b88e5a18fffb5e46b9db4c41f84b4 when trying to accept the nominations for https://bugs.launchpad.net/ubuntu/+source/meta-kde/+bug/1002336 |
|
[02:34] <ubot5> Ubuntu bug 1002336 in smokeqt (Ubuntu) "SRU tracking bug for KDE SC 4.8.3" [Undecided,Fix released] |
|
[02:34] <ubot5> https://lp-oops.canonical.com/oops.py/?oopsid=e24b88e5a18fffb5e46b9db4c41f84b4 |
|
[02:38] <micahg> it was suggested to me to use a blueprint for something like this |
|
[02:39] <wgrant> oooooooh deary me |
|
[02:39] <StevenK> RAOF: I'd look, but lp-oops loves me not. |
|
[02:39] <wgrant> That is *not* going to work well |
|
[02:40] <RAOF> wgrant: Am I going to be timing out one of your precious LP daemons with a hundred bug nominations? |
|
[02:40] <RAOF> :) |
|
[02:40] <wgrant> That's, like, more than a kernel team level of task abuse :) |
|
[02:41] <RAOF> Anything I can do about it? |
|
[02:41] <wgrant> The solution is to not do things like that. |
|
[02:41] <micahg> RAOF: fix LP :) |
|
[02:42] <RAOF> Let me clarify: anything I can do in the next 10 minutes to make my SRU process work? |
|
[02:42] <wgrant> No. |
|
[02:42] <wgrant> Other than not using one bug for 55 pacakges |
|
[02:42] <wgrant> It might work in couple of weeks when one of my DB optimisations is deployed. |
|
[02:43] <wgrant> But the chance of it working before then is roughly zero |
|
[02:43] <wgrant> And the chance of it not making me angry after then is less than zero. |
|
[02:43] <RAOF> wgrant: You might want to alert the KDE team to that, then. |
|
[02:44] <micahg> LP isn't happy with more than 10 tasks in general (disabling javascript and what not) |
|
[02:46] <lifeless> StevenK: lp-oops issue? |
|
[02:47] <wgrant> lifeless: Just taking ages, because the OOPS is horrible. |
|
[02:47] <lifeless> wgrant: I'm asking in case its other than that |
|
[02:48] <RAOF> wgrant: You've got the ability to disable timeouts or use a longer timeout or something, right? Could you accept those nominations, or will I resign myself to not tracking that bug as easily? |
|
[02:48] <lifeless> RAOF: in extreme situations we do. You might be able to reject the nominations and add separate individual tasks. |
|
[02:49] * micahg would like to fix this one day |
|
[02:49] <lifeless> using one bug, if the group of packages have to move in lockstep is, in principle, ok. It is however tricky with LP's guts today to do so many. |
|
[02:49] <wgrant> RAOF: The timeout on that page is already more than twice the normal timeout. |
|
[02:49] * micahg hopes one day we get bug linking |
|
[02:52] * RAOF vents at LP timeouts again. |
|
[02:53] <lifeless> RAOF: the timeouts aren't the root cause |
|
[02:53] <RAOF> Apparently another thing to not do is accept multiple packages from the queue at once. |
|
[02:53] <lifeless> RAOF: email sending is a major factor for that one |
|
[02:53] <lifeless> RAOF: in particular determining *who* to email takes a while. |
|
[02:54] <wgrant> (because every package may have different subscribers. so for every package it creates a new task which looks up the subscribers for every package and so on) |
|
[02:55] <lifeless> s/may have/probably has/ |
|
[02:55] <lifeless> what it should do is dispatch the event notification and move on |
|
[02:55] <lifeless> that would let us to much more work. |
|
[02:55] <wgrant> Huh? |
|
[02:55] <lifeless> wgrant: notification service |
|
[02:55] <wgrant> We won't have that until next decade. |
|
[03:00] <RAOF> Grr! Ok, launchpad. Will you work if I only accept 4 packages from the queue at once? |
|
[03:20] <lifeless> RAOF: 1 |
|
[03:26] <wgrant> RAOF: Queue batch acceptance is a bonus at the best of times. It's even less likely to work when all the uploads reference an enormous bug. |
|
=== czajkowski changed the topic of #launchpad to: https://launchpad.net/ | Help contact: czajkowski | Launchpad is an open source project: https://dev.launchpad.net/ | This channel is logged: http://irclogs.ubuntu.com/ | User Guide: https://help.launchpad.net/ | Support: https://answers.launchpad.net/launchpad | For packaging help: join #ubuntu-packaging |
|
=== dpm__ is now known as dpm |
|
=== Mission-Critical is now known as MissionCritical |
|
[09:02] <bencer> hi all, when trying to access pad.ubuntu.com despite i'm loggged in says: "Either you have not been granted access to this resource or your entitlement has timed out. Please try again.", any ideas of what could be wrong? |
|
[09:06] <czajkowski> bencer: ehterpad is a group you need to get access granted to you |
|
[09:07] <bencer> czajkowski: uhm the thing is that i was at uds and i had to request other people to take notes of the sessions |
|
[09:07] <bencer> because i couldnt access, and now i want to see the notes... how could i do it? can i request access to the etherpad group? |
|
[09:07] <czajkowski> bencer: right so mst people asked to be added to the team so they could take notes :) |
|
[09:08] <czajkowski> bencer: https://launchpad.net/~ubuntu-etherpad |
|
[09:08] <czajkowski> bencer: what is your lp nick ? |
|
[09:08] <bencer> czajkowski: bencer |
|
[09:09] <bencer> czajkowski: just requested join |
|
[09:09] <czajkowski> bencer: done |
|
[09:09] <bencer> thanks!! |
|
[09:09] <czajkowski> bencer: in future if you're struggling at uds to use the tools there ask and someone will help you |
|
[09:09] <czajkowski> rahter than waiting till after uds :) |
|
[09:10] <bencer> well, was a kind of excuse to have somebody else take notes :D |
|
[09:10] <czajkowski> bencer: yes :/ |
|
[09:10] <bencer> but maybe the error message could be a bit more explainful |
|
[09:10] <czajkowski> people tend to frown upon that, and like to rotate it |
|
[09:11] <czajkowski> grant access does imply you need to get access |
|
[09:11] <czajkowski> anyways you have it now |
|
=== Ng_ is now known as Ng |
|
=== danilos_ is now known as danilos |
|
=== yofel_ is now known as yofel |
|
=== epsy is now known as \u03b5 |
|
[13:16] <vibhav> Does anybody have any idea why https://launchpad.net/~pitti/+related-software fails to load? |
|
[13:16] <mgz> because ~pitti is too darn prolific? |
|
[13:17] <vibhav> Error ID: OOPS-5423cdd9bbf86c43ee83358e9fe88c8e |
|
[13:17] <ubot5> https://lp-oops.canonical.com/oops.py/?oopsid=5423cdd9bbf86c43ee83358e9fe88c8e |
|
[13:17] <vibhav> mgz: profilic? |
|
[13:21] <mgz> so, the direct answer is that three sql queries related to finding he software related to him on take over a second |
|
[13:22] <mgz> the longest query on my page is 21ms by contrast |
|
[13:22] <mgz> so, scales badly with large numbers of hits. |
|
[13:23] <mgz> see bug 735972 |
|
[13:23] <ubot5> Launchpad bug 735972 in Launchpad itself "Person:+related-software timeouts" [Critical,Triaged] https://launchpad.net/bugs/735972 |
|
=== andreas__ is now known as ahasenack |
|
=== deryck is now known as deryck[afk] |
|
=== debfx_ is now known as debfx |
|
[15:30] <fugue88> staging.launchpad.net has been "Code Update in Progress" since yesterday. Any chance it will be working today? |
|
[15:32] <czajkowski> fugue88: let me go and see |
|
=== deryck[afk] is now known as deryck |
|
[15:46] <c4ff31n> hi |
|
[17:28] <fugue88> czajkowski: Any word on staging.launchpad.net? |
|
[17:28] <fugue88> My irc client went offline, not sure if I missed anything. |
|
[17:28] <czajkowski> fugue88: thedac is working on it |
|
[17:28] <fugue88> czajkowski, thedac: Cool, thanks! |
|
[17:28] <thedac> fugue88: I am doing a manual restore now |
|
[18:51] <thedac> fugue88: czajkowski staging is back up. |
|
[18:51] <czajkowski> thedac: cheers |
|
[18:52] <fugue88> thedac: Awesome, thanks! |
|
=== czajkowski changed the topic of #launchpad to: https://launchpad.net/ | Help contact: | Launchpad is an open source project: https://dev.launchpad.net/ | This channel is logged: http://irclogs.ubuntu.com/ | User Guide: https://help.launchpad.net/ | Support: https://answers.launchpad.net/launchpad | For packaging help: join #ubuntu-packaging |
|
=== matsubara is now known as matsubara-afk |
|
=== Logan_ is now known as AntiTrollMeta |
|
=== AntiTrollMeta is now known as frog |
|
=== frog is now known as Logan_ |
|
|