=== slyon is now known as slyon_ [09:16] sil2100: hi! Now that groovy is open, can you please add it to Bileto? [09:16] I guess it will be reverting https://git.launchpad.net/bileto/commit/?id=48137fa785297961c2a3c090ed557ef2393aa268 or something like that :) [09:19] bahaha that commit [09:24] Yeah, darn lru cache [09:24] mitya57: sure ;p === sil2100 changed the topic of #ubuntu-devel to: Archive: Open | 20.04 Released! | Devel of Ubuntu (not support) | Build failures: http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of Trusty-Focal | If you can't send messages here, authenticate to NickServ first | Patch Pilots: [09:26] I'm really thinking of removing the lru_cache for that one, but then again this only needs to be done twice a year, so oh well [09:32] sil2100: can't you like make it so it rechecks every hour or so, like a timeout? [09:32] that'd be enough [09:33] heck, 24 hours would be enough [10:03] You could use cachetools, which lets you specify a cache key [10:03] Then additionally key on the day number or something like that. [10:03] (or maybe (year, month, day) to avoid regressing on the same day each month actually) [10:04] Oh [10:04] cachetools has a TTL cache [10:04] Even easier :) [10:07] is there a canonical way to see why a package was removed from focal (not python2 related, i think)? [10:07] popey: LP publishing history [10:08] is that accessible via some url nonsense? [10:08] There are links in the UI ... [10:08] (assuming you mean a source package) [10:08] https://launchpad.net/ubuntu/+source/spek [10:08] i am there [10:08] aha, top right [10:09] "View full publishing history", top right [10:09] ta [10:09] I was in a maze of corridors [10:09] So reflected from https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=941338 [10:09] Debian bug 941338 in ftp.debian.org "RM: spek -- RoQA; dead upstream; unmaintained; low popcon; RC-buggy" [Normal,Open] [10:09] perfecto, thank you! [10:17] Laney, msttcorefonts fails to install, PermissionError: [Errno 13] Permission denied: '/usr/lib/msttcorefonts/update-ms-fonts' [10:17] LocutusOfBorg, ^ [10:18] -rw-r--r-- 1 root root 6155 Apr 26 18:51 /usr/lib/msttcorefonts/update-ms-fonts [10:20] sil2100: it works now, thanks! [10:43] LocutusOfBorg: Looks like protobuf FTBFS on i386 [10:57] Laney, do you have an hit for that? [10:57] ricotz, on my way [11:00] LocutusOfBorg: Probably don't build that elpa-... package on i386 [11:00] I was trying to figure out if we can add something to that i386 whitelist [11:00] dh-elpa should be installable on i386... [11:00] but yes, your solution works if the first is not viable [11:01] that means building emacs [11:02] building an OS on i386 seems a good thing to do, right? :) [11:07] I have a patch ongoing testbuild on my ppa [11:12] ricotz, its in queue/uploaded, it was a missing +x [11:30] LocutusOfBorg, I figured that much ;) [14:01] hello everybody [14:05] ahasenack: I have been suffering a bug in samba 4.11.6; I have seen yo tried to get the 4.11.7 version here: https://launchpad.net/~ahasenack/+archive/ubuntu/samba-4117/+packages [14:05] santa_: what bug? Do you have a link? [14:05] so I tried to finish the work here: https://launchpad.net/~panfaust/+archive/ubuntu/samba-4117 [14:06] ahasenack: yes, to summarize: some files are shown as folders so I can't access it, let me find the bug report... [14:06] ah, that one [14:06] I just added a comment to it this morning [14:08] yeah, that one [14:08] https://bugs.launchpad.net/ubuntu/+source/samba/+bug/1872476 [14:08] Launchpad bug 1872476 in samba (Ubuntu) "Shared files are shown as folders" [Undecided,New] [14:10] ahasenack: FYI when working for my company I had to set these two in smb.conf [14:11] client min protocol = NT1 [14:11] client max protocol = NT1 [14:12] santa_: we won't do that in the package, that is enabling the deprecated and insecure SMB1 protocol version [14:12] not just enabling, but forcing it [14:13] ahasenack: I know I'm not asking you to do that in the package, but perhaps you could use that kind of configuration to reproduce the bug? [14:13] have you tried smbclient with -m? [14:16] ahasenack: the few tests I did were with kde's plasma and smb:// [14:16] santa_: ah, ok [14:17] santa_: yeah, forcing NT1 showed the bug for me [14:17] in the gui [14:17] ok, so you should be able to test the 4.11.7 packages and just check that the bug doesn't happen there [14:18] smbclient works fine still (with -m NT1) [14:18] santa_: you confirmed that it works in 4.11.7? [14:18] ahasenack: yes, with kde's kio smb:// [14:18] * ahasenack checks the release notes at https://www.samba.org/samba/history/samba-4.11.7.html [14:18] I almost added that to focal, but something else more urgent got in the way [14:19] yeah, I have seen you just didn't have time to update the ldb symbols files ;) [14:19] I might be able to grab the fix [14:21] santa_: can you please add to the bug that 4.11.7 works, and that to reproduce the bug one has to really force NT1? (both max and min) [14:28] ahasenack: ack, I will do that. how this could be fixed for focal? just patching 4.11.6 or having 4.11.7 as SRU? [14:28] preferably patching 4.11.6 [14:28] 4.11.7, even though it has only bug fixes, would be harder as an sru [14:29] allright, a co-worker suggested me to dig into this https://bugzilla.samba.org/show_bug.cgi?id=14101 [14:29] Error: Could not parse XML returned by bugzilla.samba.org: mismatched tag: line 100, column 4 (https://bugzilla.samba.org/show_bug.cgi?id=14101&ctype=xml) [14:31] santa_: that looks promising, but doesn't explain why 4.11.7 works [14:32] ahasenack: so, I presume you don't have yet a patch candidate? needless to say I will be very glad to prepare a fixed package for a focal SRU if you want to sponsor it [14:33] I was checking the 11.7 release notes and was going to try https://bugzilla.samba.org/show_bug.cgi?id=14285, which was in the 4.11.7 release notes list, even though doesn't look too related [14:33] Error: Could not parse XML returned by bugzilla.samba.org: mismatched tag: line 100, column 4 (https://bugzilla.samba.org/show_bug.cgi?id=14285&ctype=xml) [14:34] your link looks more related [14:36] ack, so we just need to find which commit fixed this [14:38] santa_: #14101 was claimed to be fixed in 4.11.6, maybe 4.11.7 reverted it without a note in the release notes, let me check the code [14:39] #14101 being the revert [14:39] well, fix and revert [14:53] ahasenack: commented @ the bug, I will try to find out which commit fixed it, feel free to investigate in parallel, and thanks for your time :) [14:53] santa_: I have a feeling it's not fixed in 4.11.7, did you build that yourself and tested? [14:53] ahasenack: yes, I did a quick test and the problem goes away [14:54] santa_: because fedora actually patched 4.11.7 [14:55] ahasenack: interesting, maybe it worked by coincidence, let me check again then... [14:58] santa_: https://gitlab.com/samba-team/devel/samba/-/commit/39c910fd9cba3caf7414274b678b9eee33d7e20b is what fedora used in 4.11.7 [15:00] Hi, I'm trying to install Ubuntu 20.04 over the network automatically. I have followed https://discourse.ubuntu.com/t/netbooting-the-live-server-installer/14510 for this, which works. The focal-live ISO is downloaded over the network in initramfs, then mounted and started. However, I now have the problem that the new autoinstall config seems to be [15:00] ignored when the PXE loader specifies `autoinstall ds=nocloud-net;s=` as kernel parmeter as shown at https://wiki.ubuntu.com/FoundationsTeam/AutomatedServerInstalls/QuickStart point 6. The system starts with the usual systemd-messages but then `cloud-init: Cloud init v. 20.1-10-g71af48df-0ubuntu5 running 'init-local' at [15:00] ` shows up and execution is held until the network is ready (which I think is because the confguration was not loaded). After a long time the ISO starts again but this time the manual installation overview is shown.Could someone explain to me how I have to provide the configuration file on a network boot or how I have to specify it as a [15:00] kernel parameter in the PXELoader? === xnox is now known as xnox|sessions [15:17] ahasenack: ok I have retested with 4.11.7 and I can see now one file shown as a directory. so indeed, not fixed yet in 4.11.7 [15:17] ok, thanks for double checking === heliocastro is now known as helio|afk === ben_r_ is now known as ben_r === diddledan1 is now known as diddledan