File size: 7,291 Bytes
4aa5fce |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 |
[08:08] <harmw> hm, isn't there something like a codestandard on readthedocs..? I know of some openstack projects that have just that [12:40] <smoser> philips, it doesn't actually matter there. [12:40] <smoser> the '-' get converted to '_'. [12:41] <smoser> harmw, suck. [12:42] <smoser> i don't know what to do about that really. i like that there is a version string in the image, but I really wish that [12:42] <smoser> a.) i didn't screw that up [12:42] <smoser> b.) i could properly "promote" a release [12:42] <smoser> (ie, rename) [12:43] <smoser> build-release doens't do anything that shouldn't be doable in a ci build [12:47] <smoser> harmw, the '~pre1' is by design (versus '-pre1'). '~' in dpkg means "less than". [13:04] <harmw> uhm, whats wrong with committing a change in bin/build-release since that's where you keep track of version numbering :) [13:05] <harmw> and using the ~ annoyed me, I wasn't aware of a dpkg design thingy [13:06] <harmw> but whats wrong with using - instead of ~? 0.3.2-pre1 would still be the same as 0.3.2~pre1, right? [13:30] <smoser> harmw, 0.3.2-pre1 is > 0.3.2 [13:31] <smoser> $ dpkg --compare-versions 0.3.2-pre1 gt 0.3.2 && echo yes, greater || echo no [13:31] <smoser> yes, greater [13:31] <harmw> ah like that [13:32] <smoser> the issue with committing a change in bin/build-release is that then you cant feed it a tag as input. [13:33] <smoser> the other issue is that then 0.3.2~pre4 can't just be "promoted" to 0.3.2 [13:33] <smoser> as it has that yucky string in it. [13:33] <smoser> i don tknow. its over thinking things i know. [13:37] <harmw> can't we just go from 0.3.2rc1 to 0.3.2? [13:37] <smoser> what do you mean? [13:38] <harmw> nah, just thinking out loud [13:44] <smoser> harmw, i'm ok to punt on that, and just change that one thing and call it 0.3.2. [13:44] <smoser> assuming my kernel change didn't cause anything else [13:45] <harmw> it builds just fine, I haven't tested the new image though [13:45] <harmw> planning to do just that tonight [13:53] <smoser> crap. [13:53] <smoser> /dev/null has bad perms in the tarball. [13:55] <smoser> and disk image [14:04] <harmw> and you only updated the kernel :p stupid kernel [14:04] <smoser> ? [14:04] <smoser> oh. no. it was a bug before. [14:04] <smoser> fix in hand. [14:05] <smoser> and it explains why running 'cirros-status' as root would not get you network debug info [14:05] <harmw> hm, so my images are broken as well? [14:05] <smoser> right [14:05] <harmw> (since I'm not using bin/bla) [14:06] <smoser> fixed and pushed. [14:06] <harmw> nice, now I only need to wait till midnight :p [14:06] <smoser> you weren't using bin/bla ? [14:06] <smoser> at all ? [14:06] <smoser> you probably were. [14:07] <harmw> haha, then I'm not going to get the fun out of this commit [14:07] <harmw> I was using the steps you describe in README [14:07] <smoser> you were. xmakedeves gets called from [14:07] <smoser> * build disk images using bin/bundle [14:07] <smoser> $ sudo ./bin/bundle -v output/$ARCH/rootfs.tar download/kernel-$ARCH.deb output/$ARCH/images [14:08] <harmw> bin/bundle, yes [14:08] <harmw> not bin/build-release [14:08] <smoser> right. build-release just does what is in the README [14:08] <smoser> its proably almost identical to yours [14:08] <harmw> indeed it is :) [14:08] <harmw> (though mine doesn't upload to LP ) [14:09] <smoser> it doesn't upload [14:09] <smoser> where did you think it did that ? [14:09] <smoser> (actually, theres a bug that *i*haven't uploaded to launchpad :) [14:15] <harmw> hm well I thought to have skimmed over some bits that pushed something to LP [14:15] * harmw should probably just stop thinking every know and then === harlowja_away is now known as harlowja [17:50] <harmw> smoser: I see 0.3.2 is tagged as such? [17:50] <harmw> I'm auto-building in about 10 minutes [17:51] <smoser> no tag [17:51] <harmw> just ned to think of something to only build a new buildroot when it is really -really- neded [17:52] <smoser> i'mo building trunk right now. [17:52] <smoser> i'll put it on download and just not call it [17:52] <smoser> but can point you at it [17:53] <harmw> fair enough :) [17:53] <smoser> its built i386, amd64 and is 10 minutes into arm [17:54] <smoser> i suspect 20 minutes it'll finish. [17:54] <harmw> ah yes, the other archs - I'm still only building x86_64 [17:54] <harmw> btw why is make br-source done witj i386 as arch? [17:54] <smoser> just because [17:55] <harmw> hmk, and its wrong to have it build with $ARCH ? [17:55] <harmw> i386/arm/x86_64 ? [17:56] <smoser> probably, yeah. they do have different packages though i think. [17:57] <smoser> i forget. [17:57] <harmw> well having it build against x86_64 went great, haven't tried it on arm/i386 though [18:00] <smoser> oh. right. it iwll work. [18:00] <smoser> and it i think does it anyway. [18:00] <smoser> the explicit step is just to have the download done [18:00] <smoser> so that it hits cache [18:01] <harmw> ah lol [18:01] <harmw> well its caching those downloads anyway [18:01] <harmw> any thoughts on this btw https://bugs.launchpad.net/cirros/+bug/1292973 ? [18:07] <harmw> harlowja: news from the bsd front? [18:07] <harlowja> harmw ah, let me see if i can get sean in, didn't chat last week [18:07] <harlowja> pinging now [18:08] <harmw> hm, sucks my server hasn't got enough firepower left to host a jenkins instance for c-i nightlies [18:09] <smoser> harmw, is that busted in current buildroot ? [18:09] <smoser> we should forward there if it is [18:09] <harmw> yea, it regards the 2012.05 buildroot [18:09] <harmw> or just upgrde buildroot [18:10] <harmw> which i've already done in some branch [18:15] <harlowja> harmw no response ping from sean, i'll let u know [18:15] <harlowja> will get him in here or bug him about whats happening/status [18:16] <smoser> http://paste.ubuntu.com/7109590/ [18:17] <smoser> harmw, ^ . its odd. i built on an azure instance to a tmpfs, thkning it would go crazy faster. [18:17] <smoser> but it did not. [18:19] <smoser> harmw, http://download.cirros-cloud.net/pre-release/0.3.2/ [18:29] <harmw> tmpfs, nice, I was thinking of going that way aswell [18:30] <harmw> my buildroot is 2G, thats gonna be a close call... [18:31] <smoser> well, 7G was within a cuple hundred meg. [18:31] <smoser> but that included everything. [18:32] <harmw> 2.1GB, without the downloads folder [18:32] <harmw> oh and its still compiling [18:33] <harmw> ah no, just finishing up [18:33] <harmw> took 33 minutes [18:34] <harmw> http://buildservice.weites.com/cirros/trunk/ [18:34] <harmw> nice :> [18:35] <harmw> smoser: do you have reallife results from ARM-based cirros boots? [18:35] <smoser> no. other than lxc. [18:36] <harmw> cool, I'm expecting a cubietruck which has hardware virt stuff [18:36] <smoser> oh nice. [18:37] <smoser> canonical has some such hardware, and its possible people here have used it. [18:37] <smoser> but i'm not sure actually. i really did it for lxc. but i'd like to have it known functional on som eactual hardware [18:37] <smoser> and at some point would want arm64 stuff too [18:38] <harmw> it should arrive next week or so [18:38] <harmw> and then Ill load it up as nova-compute node === harlowja is now known as harlowja_away === harlowja_away is now known as harlowja |