File size: 7,105 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 |
[08:31] <eoli3n> Hi
[08:38] <eoli3n> As said few days ago : 14:40 +R0Ck3R
[08:38] <eoli3n> 15:44 +mbond
[08:38] <eoli3n> 16:33 -mbond
[08:38] <eoli3n> 19:00 abbe_>abbe
[08:38] <eoli3n> 19:35 +mbond
[08:39] <eoli3n> sorry for this, wrong paste :/
[08:40] <eoli3n> so as said few days ago about this : https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1882232
[08:40] <ubottu> Launchpad bug 1882232 in chromium-browser (Ubuntu Focal) "Fresh install of chromium-browser during installer-like process, in a chroot fails (when preparing machines prior to first boot)" [Medium,Fix committed]
[08:40] <eoli3n> snap packages can't work with mounted homes
[08:40] <eoli3n> we use nfs homes
[08:41] <eoli3n> cannot open path of the current working directory: Permission denied
[08:42] <eoli3n> i think zyga said me that he could fix this
[08:46] <zyga> eoli3n: o/
[08:46] <zyga> I'm a bit indisposed lately
[08:47] <zyga> eoli3n: can you please tell me where you mount nfs homes?
[08:48] <eoli3n> under /home
[08:48] <zyga> directly in /home/foo or in a deeper structure?
[08:49] <zyga> eoli3n: can you show me how the mount is declared in either /etc/fstab or in a systemd mount unit please
[08:55] <eoli3n> zyga its mounted with autofs
[08:55] <eoli3n> prodpeda-samba.domain.fr:/home/p00000012366 on /home/p00000012366 type nfs4 (rw,relatime,vers=4.0,rsize=8192,wsize=8192,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=162.38.80.24,local_lock=none,addr=162.38.151.4)
[08:56] <zyga> eoli3n: can you paste the fstab line as well please
[08:56] <zyga> this is what we use for detection
[08:56] <zyga> *detection
[08:56] <eoli3n> there is no fstab line
[08:56] <eoli3n> i use autofs
[08:56] <zyga> ah, I see
[08:57] <zyga> I'm not familiar with autofs, how does it work?
[08:57] <eoli3n> https://help.ubuntu.com/community/Autofs#Wildcard_characters
[08:58] <eoli3n> but i don't get what information you need ?
[08:58] <eoli3n> autofs mount a directory when it is accessed
[08:58] <zyga> snapd has a global flag that indicates if NFS/CIFS workaround should be enabled
[08:58] <zyga> we use two data sources for that
[08:58] <zyga> we scan /etc/fstab
[08:58] <zyga> and we scan actually mounted filesystems
[08:58] <eoli3n> oh ok i get what you mean
[08:59] <zyga> I'm looking for something that indicates that autofs is in use
[08:59] <zyga> so the data is defined in /etc/auto.master?
[08:59] <zyga> could you please pastebin /proc/self/mountinfo
[08:59] <zyga> perhaps there's an automount mount point in place
[08:59] <zyga> that would be useful to detect as well
[09:01] <eoli3n> zyga: http://ix.io/2qtb
[09:01] <zyga> aha
[09:01] <zyga> 137 29 0:50 / /home rw,relatime shared:87 - autofs /etc/auto.master.d/home rw,fd=7,pgrp=22588,timeout=300,minproto=5,maxproto=5,indirect,pipe_ino=173399
[09:01] <zyga> this is interesting
[09:01] <zyga> so we can detect that
[09:01] <zyga> I think that's enough for me to fix this now
[09:01] <eoli3n> not really
[09:01] <eoli3n> because filename could change
[09:01] <eoli3n> there is no rules to be sure that autofs manage homes
[09:01] <zyga> having an autofs on /home feels like a good enough flag
[09:01] <eoli3n> yes
[09:02] <zyga> and /etc/auto.master.d/home, does that mention nfs?
[09:02] <eoli3n> is there anyworkaround to manually indicates to snapd that i use nfs homes ?
[09:02] <zyga> no, not at present
[09:02] <eoli3n> zyga see my first paste line
[09:02] <zyga> ah, I see now, thank you :)
[09:03] <zyga> I'm curious to ask about /net as well
[09:03] <zyga> the /net directory is not something that will work with snaps
[09:03] <zyga> is that a common convention?
[09:03] <eoli3n> that's nfs installed applications
[09:03] <zyga> I see
[09:07] <zyga> eoli3n: I've updated https://bugs.launchpad.net/snapd/+bug/1784774 and marked https://bugs.launchpad.net/snapd/+bug/1821193 as duplicate
[09:07] <ubottu> Launchpad bug 1784774 in snapd "snapd is not autofs aware and fails with nfs home dir" [Medium,Triaged]
[09:07] <ubottu> Launchpad bug 1784774 in snapd "duplicate for #1821193 snapd is not autofs aware and fails with nfs home dir" [Medium,Triaged]
[09:08] <eoli3n> thanks zyga
[09:27] <zyga> eoli3n: fixed
[09:28] <zyga> eoli3n: except github 500s
[09:30] <zyga> eoli3n: I'll open the pull request once github is operational
[10:11] <zyga> eoli3n: https://github.com/snapcore/snapd/pull/8936
[10:38] <sil2100> cpaelzer_: hey! Looking at the removal of php-horde* packages - just making sure, we also want to remove those that migrated to the release pocket already, right?
[10:38] <sil2100> Or does it make sense to leave them there as they migrated?
[10:39] <sil2100> Since it seems 16 of those migrated successfully
[11:40] <eoli3n> zyga: great, thanks a lot
[11:42] <zyga> :)
[12:43] <cpaelzer_> sil2100: you mean 1868281 I guess
=== cpaelzer_ is now known as cpaelzer
[12:44] <cpaelzer> yes I think that includes these
[12:44] <cpaelzer> I'll also ask bryce in standup today to be sure
[12:48] <rafaeldtinoco> cpaelzer: just as a fyi... if you're splitting the logical tags in versions because of me (when reviewing your big qemu merges), no need
[12:51] <sil2100> cpaelzer: yeah, removing all of those now, we can always re-sync if needed
[13:23] <kanashiro> hi o/ today I am starting my +1 maint shift and I started updating the "Things to pick" wiki page section, now I am going to check excuses and pick something to work on
[13:38] <sil2100> kanashiro: excellent o/
[13:39] <sil2100> kanashiro: I'm looking at the r-cran-httr test failures now - I see those regressed in the release pocket actually as well, but I want to check if they're reported upstream or not before I hint them
[13:41] <kanashiro> sil2100: I am checking puppetdb, it is stuck in proposed for a while and it hasn't migrated in Debian for almost one year. I'll probably file a RM bug against it
[13:49] <Unit193> ..If you were super ambitious, you could drop the upper limit in puppet-beaker. >_>
[14:02] <kanashiro> Unit193: I feel the Debian maintainer is not having too much time to work on the package, I do not want to fix it now and then need to handle other bugs on my own. Better sync it again when it is well maintained
[14:02] <kanashiro> btw it never landed in the release pocket
[14:04] <Unit193> I only mention as it holds up net-scp
[14:07] <kanashiro> Unit193: ah I see, did you try to relax the net-ssh version constraint in puppet-beaker?
[14:07] <Unit193> net-scp*, no bug I figure it's what's needed.
[14:07] <kanashiro> if not I can give it a shot
[14:12] <Unit193> It's easier when all affected packages are in pkg-ruby. :/
[14:14] <kanashiro> Unit193: agreed, one interested developer asked for help to work on puppet packages during the last debian ruby team irc meeting
[14:15] <kanashiro> they lack manpower apparently
[14:19] <Unit193> I mean, according to UDD same for -ruby...
[14:52] <kanashiro> Unit193: after relaxing versions of net-s{sh,scp} the test passed, I'll submit a patch to Debian and upload it to Ubuntu
[15:09] <Unit193> \o/
=== zyga-mbp is now known as zyga
=== heather1 is now known as Guest39331
|