#archlinux-ports | Logs for 2017-10-30

[01:55:38] eschwartz is now known as guys
[02:30:35] -!- p71 has quit [Remote host closed the connection]
[02:47:30] -!- p71 has joined #archlinux-ports
[02:50:57] -!- tyzoid has quit [Ping timeout: 248 seconds]
[02:53:11] -!- tyzoid has joined #archlinux-ports
[03:41:40] -!- fhdrin has joined #archlinux-ports
[04:00:46] -!- fhdrin has quit [Ping timeout: 252 seconds]
[04:26:46] -!- tyzoid has quit [Ping timeout: 264 seconds]
[04:28:15] -!- tyzoid has joined #archlinux-ports
[05:12:22] guys is now known as es
[06:30:02] -!- deep42thought has joined #archlinux-ports
[06:36:54] <deep42thought> isacdaavid: thanks for the duplication notice, actually the build master checks the master mirror for that - so it should not happen on the master mirror
[06:37:35] <deep42thought> the problem with built-for-i686-instead-of-any was an error on my end
[06:37:57] <deep42thought> first, I added 'i686' to all arch arrays, so all packages were built for i686
[06:38:17] <es> oh wow :D
[06:38:21] <deep42thought> this is corrected for some time, now, but older packages may still be i686 instead of any
[06:38:27] <deep42thought> but their next update will fix that
[06:39:45] <deep42thought> as to why you don't see a difference on https://github.com - only "real" differences are stored there - the additional architecture is added automatically and ist not tracked via git
[06:41:45] <deep42thought> ca-certificates-* is not duplicate on the master mirror
[06:44:36] <deep42thought> isacdaavid: on which mirror did you see that? Maybe the operator forgot to add some --delete option?
[06:45:23] <isacdaavid> the main one (afaict) -> https://mirror.archlinux32.org
[06:46:24] <deep42thought> well, ca-certificates is any, but ca-certificates-cacert is (still) i686
[06:46:32] <deep42thought> should become any, when it receives an update
[06:46:38] <isacdaavid> ok
[06:48:19] <isacdaavid> i can't imagine why you would need to change the arch array in the first place, but whatever
[06:48:40] <deep42thought> well, at some point upstream will/does remove i686 from the arch array
[06:48:44] <deep42thought> so we need to add it
[06:48:54] <deep42thought> I simply forgot to check, that the arch is not "any"
[06:49:10] <isacdaavid> i see
[06:49:35] <deep42thought> and apparently makepkg creates i686 packages if it is "arch=(i686 any)"
[06:49:36] <deep42thought> :-D
[06:51:07] <isacdaavid> yeah, doesn't make much sense
[06:51:18] <isacdaavid> makepkg has the -A flag btw
[06:51:24] <isacdaavid> dunno if the build server could make use of it to avoid changing packages that only say arch=(x86_64)
[06:52:18] <deep42thought> yeah, I noticed that flag, too - but too late
[06:53:02] <deep42thought> and now the other logic is in place
[06:53:19] <deep42thought> and we need the possibility to change PKGBUILDs anyway
[06:56:28] <isacdaavid> i'll use the occasion to thank everyone involved with keeping i686 alive
[06:57:35] <deep42thought> yeah, it's been fun :-)
[07:12:54] -!- isacdaavid has quit [Quit: Leaving.]
[07:26:39] -!- deep42thought has quit [Quit: Leaving.]
[07:57:51] -!- nmxjkrbjhx has joined #archlinux-ports
[07:57:52] -!- nmxjkrbjhx has quit [Remote host closed the connection]
[08:27:34] -!- deep42thought has joined #archlinux-ports
[11:04:11] -!- shentey has joined #archlinux-ports
[11:14:59] -!- shentey has quit [Remote host closed the connection]
[11:15:25] -!- shentey has joined #archlinux-ports
[11:20:05] -!- es has quit [Quit: A random quit message]
[11:22:26] -!- shentey has quit [Read error: Connection reset by peer]
[11:22:38] -!- shentey has joined #archlinux-ports
[11:34:01] -!- shentey has quit [Remote host closed the connection]
[11:34:27] -!- shentey has joined #archlinux-ports
[11:56:38] -!- shentey has quit [Quit: Konversation terminated!]
[13:21:01] -!- guys has joined #archlinux-ports
[14:31:32] -!- yar has quit [Excess Flood]
[14:34:44] -!- yar has joined #archlinux-ports
[15:34:29] <deep42thought> tyzoid: I'm unable to access the forum via sshfs
[15:34:42] <deep42thought> have (not) you migrated my login already?
[15:43:25] -!- shentey has joined #archlinux-ports
[16:09:41] -!- deep42thought has quit [Quit: Leaving.]
[16:37:32] -!- p71 has quit [Read error: Connection reset by peer]
[17:15:25] -!- p71 has joined #archlinux-ports
[17:26:39] -!- guys has quit [Ping timeout: 248 seconds]
[17:42:39] -!- guys has joined #archlinux-ports
[17:52:48] -!- deep42thought has joined #archlinux-ports
[20:13:30] -!- tyzoid_work has joined #archlinux-ports
[20:15:05] <tyzoid_work> deep42thought: That'd be correct. I've been more concerned with keeping it *online*.
[20:15:18] <deep42thought> yeah, no problem :-)
[20:15:27] <tyzoid_work> After experiencing RAID failure on the new dedi, I decided to do a stress test before finishing the roll out
[20:15:30] <tyzoid_work> they replaced the disks
[20:15:32] <tyzoid_work> so...
[20:15:41] <deep42thought> oh
[20:15:54] <tyzoid_work> *yay for backups*
[20:15:59] <deep42thought> :-)
[20:16:00] <rewbycraft> Oof
[20:16:31] <tyzoid_work> dd if=/dev/urandom of=testfile for 800GB should be pretty close.
[20:16:34] <rewbycraft> One disk fail or both fail?
[20:16:49] <tyzoid_work> One disk, but something corrupted and I did a complete wipe/reinstall anyway
[20:17:01] <tyzoid_work> It's supposed to be software raid1
[20:17:06] <rewbycraft> I'd run some smart tests
[20:17:16] <rewbycraft> Let the drives detect issues themselves
[20:17:28] <tyzoid_work> Can that be done from the OS over ssh? or would I need to have them set up a IP-KVM?
[20:17:41] * tyzoid_work knows about SMART, never ran one
[20:19:02] <tyzoid_work> rewbycraft: Also, can it be run from the disk mounted as / ?
[20:20:12] -!- shentey has quit [Quit: Konversation terminated!]
[20:48:32] <guys> deep42thought: https://bugs.archlinux.org
[20:48:47] <deep42thought> I saw it
[20:48:48] <guys> Seems the maintainer cannot duplicate, but has updated to a new release anyway
[20:48:53] <rewbycraft> tyzoid_work: Depends on the disk model. https://www.thomas-krenn.com
[20:49:14] <guys> does it work for you now?
[20:49:49] <deep42thought> it's currently being compiled
[20:50:38] <deep42thought> nope, still opens the debugger (not the bug which was reported)
[20:55:29] <deep42thought> https://buildmaster.archlinux32.org
[20:55:37] <deep42thought> it's the same error as before for i686
[20:55:38] <tyzoid_work> rewbycraft: Thanks, I'll take a look at that later
[21:10:35] -!- shentey has joined #archlinux-ports
[21:30:47] <tyzoid_work> hey deep42thought: I've got an old 32bit laptop kicking around somewhere. Assuming I can find it, any use of me setting it up over ssh and letting you test remotely with it?
[21:31:17] <deep42thought> hmm, not really
[21:31:24] <deep42thought> I wouldn't know what to run on it
[21:31:50] <deep42thought> It would fit for some automated testing, though - once we got that running :-)
[21:32:16] <tyzoid_work> deep42thought: IIRC it doesn't have sse2, so it might be useful there. I'll definately keep it in mind for automated testing, though.
[21:32:58] <tyzoid_work> deep42thought: Also, prepping for the ISO build in two days, any other changes to archiso? I think you mentioned something about archiso being updated in mainline.
[21:33:09] <deep42thought> hmm, if it really does not have sse2, it might be nice for testing some stuff occasionally
[21:33:40] <tyzoid_work> If that's the case, i'll see if I can't dig it up and throw arch32 on it, then.
[21:34:12] <deep42thought> archiso changed some hashing algorithms
[21:34:28] <deep42thought> but I think, it's not really important
[21:35:28] <deep42thought> there is a "new" commit which is not yet in a released version upstream - and therefor also not in a released version in archlinux32
[21:36:16] <deep42thought> maybe you can use the current git branch tips for creating the isos?
[21:38:14] <tyzoid_work> deep42thought: The head of mainline? or head of archlinux32?
[21:38:23] <deep42thought> our heads :-)
[21:38:37] <tyzoid_work> either way - my build script makes using the packaged version easier
[21:38:52] <tyzoid_work> so I'll do a test with the head and report issues, but I'd rather build the monthly from a released package
[21:39:07] <deep42thought> yeah, np
[21:39:16] <deep42thought> I think the change is a minor one
[21:40:04] <tyzoid_work> deep42thought: Yeah, doesn't look all that significant.
[21:40:12] <tyzoid_work> Plus, the git message is funny: "Nobody wants to use md5 these days..."
[21:40:19] <deep42thought> lol
[21:40:24] <deep42thought> didn't read that :-)
[21:40:54] <tyzoid_work> yeah, the multiline messages don't show with --oneline :(
[21:40:55] <tyzoid_work> go figure
[21:41:01] <tyzoid_work> but yeah, still pretty funny
[21:41:36] <tyzoid_work> deep42thought: You should check out signing your release tags
[21:41:50] <deep42thought> they are signed
[21:41:54] <tyzoid_work> git allows tags and commits to be gpg-signed, using the -S arg.
[21:42:08] <tyzoid_work> deep42thought: https://github.com ?
[21:42:24] <tyzoid_work> I don't see it, unless github is just showing that it's signed, but you haven't uploaded your pubkey yet
[21:42:31] <deep42thought> I never figured out, how to let github import the signatures from git
[21:42:46] <tyzoid_work> ah, you add that to your profile in the same spot as ssh keys
[21:42:52] <tyzoid_work> below the box for ssh keys, there's one for gpg keys
[21:43:20] <deep42thought> > git tag -v dual-v32.3
[21:43:20] <deep42thought> shows the correct signature and everything
[21:43:38] <tyzoid_work> yup, though github doesn't know about it yet.
[21:44:36] <deep42thought> the key was in the profile already
[21:44:43] <deep42thought> (I must have inserted it already)
[21:45:19] <tyzoid_work> deep42thought: Hmm, is your git commit author the same as the uid on your gpg key?
[21:45:25] <tyzoid_work> Github says this: "The email in this signature doesn't match the committer email."
[21:45:36] <deep42thought> I think, the problem is, that I use new email addresses for everything - and the git-commit address is not on the key
[21:45:44] <tyzoid_work> That'd be it
[21:45:58] <tyzoid_work> If you add that uid to your key, you can upload that to github
[21:46:03] <tyzoid_work> I'll sign the new version when I get home
[21:46:37] <tyzoid_work> or you can switch your author email. Either way works
[21:48:17] <deep42thought> better?
[21:49:32] <tyzoid_work> deep42thought: Yup, now showing verified
[21:51:06] <tyzoid_work> deep42thought: did you upload it to a keyserver?
[21:51:14] <deep42thought> nope
[21:51:18] <deep42thought> let me try again
[21:51:21] <tyzoid_work> and if so, can you upload to the mit keyserver? That way, I can get it pretty much instantaneously
[21:51:36] <tyzoid_work> deep42thought: Works fine on github, this is just so I can sign the new uid
[21:52:20] <deep42thought> ok, should be done
[21:54:35] <tyzoid_work> deep42thought: alright, pull back from MIT and you should see my sig
[21:55:23] <deep42thought> yup, worked
[21:55:26] <deep42thought> thanks for signing
[21:55:57] <tyzoid_work> np. Now git should verify for people who trust my key too :)
[21:56:39] <tyzoid_work> btw, just discovered that you can add .patch at the end of a github commit URL
[21:56:41] <tyzoid_work> https://github.com
[21:56:48] <tyzoid_work> and it'll show the text of the commit
[21:57:01] <deep42thought> yes, i know
[21:57:13] <guys> tyzoid_work: you never learned all this stuff?
[21:57:44] <tyzoid_work> Github isn't my primary code hosting environment. I usually use the CLI, and my previous company used stash/bitbucket from Atlassian.
[21:58:18] <tyzoid_work> plus, I ran my own private instance of gitlab for a few years
[21:58:20] <guys> `curl https://github.com`
[21:58:23] <guys> -_-
[21:59:45] <deep42thought> I didn't know _that_ interface O.o
[21:59:54] <tyzoid_work> hmmm
[21:59:59] <tyzoid_work> ~$ curl https://www.github.com
[22:00:00] <tyzoid_work> curl: (6) Couldn't resolve host 'www.github.com'
[22:00:19] <guys> Note: The keys with the following IDs couldn't be exported and need to be reuploaded 3B08EFC6BA974CFC
[22:01:28] <tyzoid_work> guys: hmm, wonder why that is
[22:02:46] <tyzoid_work> guys: Github still verifies my key, though: https://github.com
[22:03:43] <tyzoid_work> anyway, I'm heading out, I'll check irc when I get back home in ~1hr
[22:04:21] <deep42thought> good by
[22:04:32] <deep42thought> "bye"?
[22:05:10] <tyzoid_work> bye!
[22:05:17] -!- tyzoid_work has quit []
[22:10:49] -!- deep42thought has quit [Quit: Leaving.]
[22:17:00] -!- guys has quit [Quit: A random quit message]
[22:31:53] <rewbycraft> deep42thought: I may be able to supply a 32 bit machine that has IPMI
[22:31:58] <rewbycraft> May be useful for automated testing
[22:32:29] <rewbycraft> Because the script can access the serial console, powercycle the machine, load isos directly into the virtual cd drive, etc
[23:21:04] -!- shentey has quit [Quit: Konversation terminated!]