- Friday, November 16, 2018

release-team@conference.openafs.org

Friday, November 16, 2018< ^ >

Room Configuration

Room Occupants

GMT+0
[04:28:39] mvita joins the room
[09:06:12] mvita leaves the room
[13:27:14] meffie joins the room
[14:00:40] wiesand joins the room
[14:01:21] <meffie> good afternoon.
[14:01:45] <wiesand> good morning
[14:03:03] <kaduk@jabber.openafs.org/barnowl> greetings
[14:03:22] <wiesand> another week gone by w/o much action here :-(
[14:03:25] <kaduk@jabber.openafs.org/barnowl> I've been under the weather this week and have gotten nothing done for
openafs
[14:03:25] <wiesand> hello ben
[14:03:46] <wiesand> I think there's one major topic to discuss today
[14:04:30] <wiesand> gerrit 13392 & company
[14:05:44] <meffie> ah, yes mark has been busy with those.
[14:05:55] <wiesand> removing afs_osi_SetTime() seems fine on master and 1.8.x
[14:06:16] <wiesand> but what do we do with 1.6.x - where we haven't ripped out -settime support yet
[14:06:17] <kaduk@jabber.openafs.org/barnowl> Yup, it sounds reasonable
[14:06:36] <wiesand> rip it out now?
[14:07:10] <wiesand> or waste effort on something that's broken on linux anyway?
[14:07:32] <kaduk@jabber.openafs.org/barnowl> Probably not rip out the whole thing on all platforms.
I'd consider making no-op stubs for linux 4.20+ but otherwise
minimizing changes.
[14:07:53] <meffie> good idea.
[14:09:04] <wiesand> sounds reasonable, but it still has to be done
[14:09:25] <wiesand> and it's one more configure test
[14:09:35] <meffie> i'll talk to mark about it.
[14:09:46] <wiesand> thanks
[14:10:36] <meffie> or maybe marcio or i could do the stubs, while mark looks at the next linux issue
[14:11:09] <wiesand> the gplonly ting?
[14:11:29] <meffie> \
[14:11:36] <meffie> yes
[14:11:59] <wiesand> thanks to mark for working on that stuff
[14:12:38] mvita joins the room
[14:12:52] <wiesand> nb our windows bsods have become much more infrequent and harder to reproduce, but they're still there
[14:12:53] <meffie> he said he was looking into that next after this current_kernel_time issue.
[14:13:00] <mvita> I'm here
[14:13:05] <wiesand> hello mark
[14:13:14] <mvita> the gpl thing was fixed in Linux
[14:13:19] <meffie> WOOT!
[14:13:19] <mvita> we don't have to do anything
[14:13:25] <wiesand> hooray
[14:13:32] <meffie> take the rest of the day off!
[14:13:46] <meffie> birch beer is on me.
[14:13:52] <mvita> with what I submitted on master, we build on 4.20-rc2
[14:14:04] <mvita> however, I have not tested it, just built it
[14:14:04] <wiesand> great, thanks
[14:14:21] <meffie> great news
[14:14:30] <wiesand> mark, do you have an opinion on what to do for 1.6?
[14:15:00] <meffie> ben suggested no-op stubs for settime.
[14:15:27] <mvita> yes, that sounds fine
[14:15:43] <mvita> they are already stubs for some of the other platforms
[14:15:44] <wiesand> (only on Linux w/o the old interfaces, right?)
[14:16:04] <mvita> yes, just the minimum required work for 1.6.x
[14:16:47] <wiesand> i think it's the minimum required change, but more work?
[14:16:55] <mvita> there are some more gerrits coming for 4.20 but they are not required - just getting rid of our use of some deprecated APIs
[14:18:06] <mvita> sorry, I'm not sure what you're asking Stephan
[14:18:21] <wiesand> OT: I read that Ben saved TLS in Bangkok last week :)
[14:19:14] <wiesand> mark: I think the minimum amount of work would be to rip out -settime (maybe on Linux only)
[14:19:17] <kaduk@jabber.openafs.org/barnowl> saved TLS from what?
[14:19:50] <wiesand> getting a backdoor
[14:22:36] <wiesand> back on topic, does anyone have an F29 VM handy to triage RT #134833 ?
[14:22:57] <mvita> not handy
[14:23:09] <mvita> but we can build it
[14:23:14] <mvita> we have the technology
[14:23:16] <mvita> ;-)
[14:24:02] <mvita> hrrm - rt.central.org is sluggish?
[14:24:13] <kaduk@jabber.openafs.org/barnowl> Indeed, though it did load for me.
[14:24:35] <mvita> SPAM ATTACK
[14:25:08] <meffie> it is slow here too.
[14:25:16] <mvita> <twiddles thumbs?
[14:25:28] <kaduk@jabber.openafs.org/barnowl> I ran into a problem with the systemd start script in the openafs-
client-1.8.1.1-1.fc29.x86_64.rpm I built from the source rpm.
For some reason, the line:
ExecStartPre=/sbin/modprobe openafs
in file openafs-client.service fails to load the kernel module, I had
to change it to:
ExecStartPre=/usr/bin/bash -c "/usr/sbin/insmod
/usr/lib/modules/$(/usr/bin/unam
e -r)/extra/openafs.ko.xz"
to make it load. This may be some idiosyncrasy of my computer but I
thought I'd pass this on in case it's not.
[14:26:10] <mvita> odd
[14:26:43] <mvita> finally got to login panel on rt
[14:29:28] <meffie> rt still not loading for me. probably everyone closing tickets because things have been fixes.
[14:29:33] <meffie> fixed
[14:29:42] <wiesand> that must be it ;-)
[14:30:03] <meffie> ;)
[14:30:49] <wiesand> I'm afraid I don't have more topics for today.
[14:31:21] <wiesand> The prereleases should probably wait until we have Linux 4.20 covered
[14:32:30] <mvita> sounds reasonable
[14:33:41] <meffie> sounds good.
[14:35:15] <wiesand> fine, thanks
[14:35:28] <wiesand> more to discuss today?
[14:35:40] <mvita> nothing here…
[14:35:42] <meffie> i dont have anything new to report on the buildbot front. been tied up on changes for the bosserver-no-client-dirs topic.
[14:36:58] <wiesand> and with Ben having been under the weather, we're probably finished for today (get well soon)
[14:37:01] <meffie> that should not block 1.8.3
[14:37:10] <kaduk@jabber.openafs.org/barnowl> Yeah, I don't have anything else for today.
[14:37:24] <kaduk@jabber.openafs.org/barnowl> (and thanks for the well-wishes!)
[14:37:24] <wiesand> mike: understood and agreed
[14:37:42] <wiesand> Let's adjourn then. Thanks a lot everyone!
[14:37:44] <mvita> sorry you're not feeling well, Ben
[14:37:46] <meffie> get well soon.
[14:37:55] <mvita> adieu
[14:38:11] <wiesand> Tschüß
[14:38:25] wiesand leaves the room
[14:41:16] meffie leaves the room
[14:41:54] mvita leaves the room
[20:19:22] mvita joins the room
[21:06:46] mvita leaves the room