[11:34] hi there :), anyone here who can tell me, how to check out a (mainline-) kernel? [11:36] It seems like kernel 3.9.* has only candidates up to "v3.9.7", but in the kernel-ppa is a "v3.9.11-saucy" version ("git checkout v3.9.11" says that git dosn't know this version) [11:39] SimonK, quite certain that 3.9.11 is a vaild version [11:40] hm, just to be on the safe side: if i am able to check out 3.10-rc1, that means 3.9.11 should be in my git-tree? [11:41] the highest number i can find is "2ea699d98cd6f9e9b813c24542d581dedacdc659 refs/tags/v3.11-rc6" [11:41] SimonK, you checked out http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=summary right? [11:42] Well, "master" only bases on Linus' tree [11:42] i.e. does not include any stable releases from upstream Linux [11:45] SimonK, mainline actually means https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/ (for stable kernels) [11:45] I cloned "git://kernel.ubuntu.com/ubuntu/ubuntu-saucy.git" [11:46] kernel.org didn't let me clone (it stopped at 13% every time, probably because of my dsl-speed) [11:47] well, then i do a bisect between 3.9.7 and 3.10-rc1. [11:48] SimonK, ubuntu-saucy != mainline............. [11:49] so I have to clone from kernel.org? Well, thats unfortunate. [12:40] SimonK, smartboyhw: the Ubuntu kernel versions are slightly different from upstream git (given that we usually start with the devel kernels) [12:40] an easy way to find that out is by looking at /proc/version_signature -- there you will see the Ubuntu version and the upstream version [12:41] I see... I'm trying to download from kernel.org at the moment *fingers crossed* [12:41] I'll never understand why you have to restart a cloning if you abrot it bevore :/ [12:47] we do have mainline kernels build for Ubuntu [12:47] SimonK: see https://wiki.ubuntu.com/Kernel/MainlineBuilds [12:49] Yes, I know which version is the last good (3.9.11) and the first bad (3.10-rc1) my problem now just comes down to bad internet-connection, git cloning from *.kernel.org is aborting every time i tried so far === psivaa is now known as psivaa-afk === psivaa-afk is now known as psivaa [18:02] anyone able to tell me what happens if an SRU goes to verification-failed ? [18:06] TheLordOfTime: the package should be removed from the -proposed pocket, and the bug should be put back into triaged [18:06] hggdh: should i let sponsors do those changes, or can I bump it back to Triaged myself? [18:06] because E: Bug 1206878 VERIFICATIONFAILED because E: New Bug [18:06] Launchpad bug 1206878 in nginx (Ubuntu Precise) "[SRU] Configuration should be purged only in nginx-common" [Critical,Fix committed] https://launchpad.net/bugs/1206878 [18:07] TheLordOfTime: you can bump it back to triaged yourself; removal from the -proposed will have to be done by someone with archive authority [18:08] hggdh: done, now, if I were to try and fix this, and actually fix the newly-introduced-bug from what's in 1.1.19-1ubuntu0.3, do I bump the version to -1ubuntu0.4, or...? [18:08] (this is the first patch I've actually had fail o.O) [18:10] good question... I think it would remain 0.3, but this might conflict with an already-uploaded source. [18:11] so... i should wait until I can go poke someone, like bdmurray who actually sponsored the upload, to explain what to do in this case? [18:11] yeah. Or -motu, or -devel, or -packaging [18:11] -MOTU's been nonresponsive [18:11] hence me asking here about the verification-failed thing [18:11] -devel will be my next target probably. [18:11] unless someone wakes up on the MOTUs [18:11] yeah, I saw your comment in -servers [18:12] * hggdh goes afk for a bit [18:21] -1ubuntu0.4 because the previous version of the package already existed in the archive [18:21] TheLordOfTime:, hggdh ^^ [18:22] bdmurray: so, I have the -1ubuntu0.3 package I pulled from proposed with dget, do i just add code modifications to that, add a new changelog entry for -1ubuntu0.4, and attach another debdiff, or do i start from -1ubuntu0.2, do changes, and then new changelog -1ubuntu0.4 ? [18:23] basically, which package do I base the updated debdiff from. [18:24] -1ubuntu0.3 [18:25] okay, so work off what's already in proposed, fix the newly-introduced-bug, test, and then attach the debdiff... [18:25] cool. [18:25] (this was the first patch that has been for nginx that verification-failed o.O) [18:25] (at least, that i've worked on) [18:27] bdmurray: yeah, I wondered about an already-uploaded debdiff. So the process is base on the failed fix, and add the fix's fix [18:31] bdmurray: for -1ubuntu0.4, though, do I also need to put in (closes LP: #bugnumber) that was referenced in -1ubuntu0.3 ? [18:32] TheLordOfTime: yes [18:32] ok [19:58] bdmurray: if you're not busy, can you perhaps guide me to a resolution of a bug issue? I'm stuck between Invalid/Wishlist and actually poking someone in a position to make a decisive answer on the bug, and that plus the emails I"m getting about it are driving me to the point of telling everyone to die in a fire... o.o [20:00] TheLordOfTime: I'm not terribly involved in anything at the moment [20:00] hggdh: you know the bug, i briefly discussed it here [20:00] bdmurray: okay, one moment [20:00] (btw, i'll need you to also sponsor a new debdiff for 1206878, that fixes the issues) [20:00] bdmurray: this is the one driving me insane: https://bugs.launchpad.net/ubuntu/+source/nginx/+bug/1170586 [20:00] Launchpad bug 1170586 in nginx (Ubuntu) "Naxsi package lacking Stub Status" [Undecided,Incomplete] [20:01] bdmurray: it's stuck between Invalid/Wishlist (because although the code for that module exists, its rules file never included it)... [20:01] and trying to figure out whether there's any way to get an approved change to the package to build the module's code, as it exists in the 1.1.19 source [20:02] quantal and later all have the module's code compiled in the rules [20:02] as it stands i'm leaving it Incomplete/Undecided until someone more... senior... can make a decision on how to proceed [20:03] the whole initial issue was based on research the OP found that pointed at 1.2.1 (on Debian's wiki page) [20:03] comment 4, i narrowed the criterion for the issue to be specific to the package, and not what debian's saying, but nginx's debian maintainers can't give me a clear answer [20:03] so... this is now in the point where someone wants the thing, but activating it isn't SRU worthy [20:04] and I don't see a backport being relevant in this case either because they want something that wasn't activated in the rules, but is actually in the source code. [20:04] so... i'm stuck :/ [20:06] bdmurray: hggdh: anyone else: i welcome any guidance on how to tackle this to get everyone off my case about this... o.o [20:10] TheLordOfTime: everyone? I see only one person affected by it. [20:11] bdmurray: direct emails to me are causing the stress [20:11] not on the bug [20:11] either way, the bug is stuck in triager limbo until someone can figure it out [20:11] on the one hand, activating the module is as simple as one line of code, on the other, doing that wouldn't fit into SRU or backport [20:12] so... kinda stuck [20:12] (part of me wants to Invalid that bug but... i'm in too high a stress level to do so sanely) [20:17] bdmurray: on a more sane, not-as-stressful note, https://bugs.launchpad.net/ubuntu/+source/nginx/+bug/1206878 has a new debdiff ready for sponsoring [20:17] Launchpad bug 1206878 in nginx (Ubuntu Precise) "[SRU] Configuration should be purged only in nginx-common" [Critical,Triaged] [20:17] that apparently fixes the issue introduced in 1.1.19-1ubuntu0.3 (that was in -proposed) [20:18] at least, from my testing with the testcases specified, it works as expected, and doesn't trigger installation or purge errors [20:18] I'd prefer no to sponsor that since I'll may be the SRU team member reviewing it. [20:18] Do you know what the patch for the other bug would look like? [20:18] okay, i'll have to go find a sponsor then... [20:19] bdmurray: you mean the one that's driving me to wanting to tell people to burn? [20:19] bdmurray: i can probably slap up an example patch in a few minutes [20:19] but for all intents and purposes... Bug 1206878 is a much higher priority than 1170586 [20:19] Launchpad bug 1206878 in nginx (Ubuntu Precise) "[SRU] Configuration should be purged only in nginx-common" [Critical,Triaged] https://launchpad.net/bugs/1206878 [20:19] because config removal on a non-common-files purge is just bad. [20:25] bdmurray: a diff for LP Bug 1170586 would look something along the lines of this: http://paste.ubuntu.com/6189724/ [20:25] Launchpad bug 1170586 in nginx (Ubuntu) "Naxsi package lacking Stub Status" [Undecided,Incomplete] https://launchpad.net/bugs/1170586 [20:25] (note because of convenience, I built that off of the -1ubuntu0.4 package I uploaded a debdiff for which fixes 1206878) [20:26] note i tend to testbuild before I submit a debdiff... [20:26] so unless someone's actually going to approve adding in that line to build the module, i'm not even going to test-build [20:26] (although in theory it should build) [20:27] and I think that fits in this part of the "When" for SRUs [20:27] Bugs which do not fit under above categories, but (1) have an obviously safe patch and (2) affect an application rather than critical infrastructure packages (like X.org or the kernel). [20:28] bdmurray: then, if i were to upload a debdiff for that and propose it for SRU under that criterion... [20:28] but I'm pretty sure I'm still the new guy on that team [20:28] then perhaps someone would look at it? [20:28] ehh, where are the SRU team anyways [20:29] or, rather, would you mind poking them and seeing what they say :P [20:29] * TheLordOfTime yawns [20:29] i need to go beat my head against the wall for a bit [20:29] but the good news is that annoying bug that removes a user's configs when nginx-light or nginx-full or nginx-naxsi or nginx-extras is purged is (hopefully) in the queue to fix it [20:30] bdmurray: last question, do i leave the verification-failed tag on 1206878 since i uploaded a new debdiff, and wait for someone from sponsors and SRU team to take a look? [20:32] TheLordOfTime: yes, because we don't want it showing up as verified on the pending sru report [20:35] okay, done. if sponsors need to be resubscribed, can you do that for me, bdmurray? i'm going to go take a warm shower to try and calm down... for some reason that's relaxing... *shrugs*