Home
release-team@conference.openafs.org
Friday, July 5, 2019< ^ >
Room Configuration
Room Occupants

GMT+0
[11:56:31] mbarbosa joins the room
[14:54:39] meffie joins the room
[14:58:40] <mvita> present
[15:00:33] <kaduk@jabber.openafs.org/barnowl> greetings
[15:01:30] <meffie> good day.
[15:02:33] wiesand joins the room
[15:02:43] <wiesand> Hi
[15:03:35] <wiesand> on 1.8.4pre1:
[15:04:23] <wiesand> 1) I'm working on deploying a test build with everything except 13675 included
[15:05:37] <wiesand> 2) Cheyenne just pulled up 13659 to 13675 - how urgent is that one?
[15:06:28] <wiesand> 3) I have a few nits on 13643 (NEWS) - Mike: Thanks a lot.! All: comments much appreciated
[15:06:47] <wiesand> And that's about it
[15:07:55] <kaduk@jabber.openafs.org/barnowl> It looks like omitting 13659 could leave a multithreaded server
process writing its logs to random places in the filesystem (most
likely relative to PWD).
[15:09:40] <meffie> yes, or just fail to reopen the log?
[15:10:11] <kaduk@jabber.openafs.org/barnowl> Or that, I suppose.  I didn't look too far.
[15:10:55] <wiesand> ok I guess we want it
[15:11:24] <meffie> yes, it's a bug introduced in 1.8.0.
[15:11:38] <wiesand> review review…
[15:13:25] <meffie> ok
[15:13:33] <mvita> aye-aye
[15:14:43] <wiesand> ok, looks like we'll include both 13674 and 13675
[15:15:13] <wiesand> then another minor update to NEWS, and we should be done
[15:15:52] <wiesand> anything else on the stable series?
[15:15:52] <kaduk@jabber.openafs.org/barnowl> Yay
[15:17:14] <wiesand> Mike, thanks for starting NEWS for 1.9.0 in 13673
[15:17:24] <kaduk@jabber.openafs.org/barnowl> Yes, thank you!
[15:18:44] <meffie> no problem. i used the holiday to write a script to find commits.
[15:18:52] <wiesand> 13599 is still on the 1.8.4 wish list and seems kind of ready?
[15:19:21] <kaduk@jabber.openafs.org/barnowl> It was waiting on 13598
[15:19:35] <meffie> yes, i think it is ready.
[15:19:50] <kaduk@jabber.openafs.org/barnowl> So thanks for reminding me to ask Mike and/or Mark about translating
EDOM to KTC_NOENT
[15:20:08] <kaduk@jabber.openafs.org/barnowl> I don't think we need to, but wanted to check before merging 13598
[15:20:11] <mvita> I don't know off the top of my head - looking at it now
[15:20:44] <mvita> I have to catch up w/ Mike's commits first
[15:21:46] <meffie> thanks mvita
[15:22:02] <mvita> I'll have to do that after the meeting, I'm just trying to keep up at the moment
[15:23:30] <wiesand> ok, on to 1.9.0/master?
[15:24:01] <kaduk@jabber.openafs.org/barnowl> Not a whole lot of note, though I did merge a few small things
yesterday.
[15:24:01] <meffie> ok
[15:24:29] <kaduk@jabber.openafs.org/barnowl> Mostly I think the plan is to take Mike's script's output and turn it
into an expositive form, then make a 1.9.0
[15:24:41] <kaduk@jabber.openafs.org/barnowl> I don't think there was anything else in flight that we needed to wait
for.
[15:24:44] <meffie> i will be doing that today, i hop.
[15:24:48] <meffie> hope
[15:25:39] <kaduk@jabber.openafs.org/barnowl> Oh, cool
[15:25:59] <meffie> seems like a fun thing to do on a friday after a holiday :)
[15:26:16] <kaduk@jabber.openafs.org/barnowl> Fairly low mental energy, sure
[15:26:47] <meffie> oh, note the new fix from andrew: 13672
[15:27:04] <kaduk@jabber.openafs.org/barnowl> I have that open in a tab to look at, yes
[15:27:51] <meffie> this fixes a bad situation when the system is under memory pressure. the only recourse is to reboot.
[15:28:58] <wiesand> it can't be all that common in the wild though - not a 1.8.4 blocker I think
[15:29:08] <meffie> it was hit at a site with a heavily loaded client (running java apps)
[15:29:23] <mvita> repeatedly
[15:29:47] <kaduk@jabber.openafs.org/barnowl> Oh, java.
[15:29:54] <mvita> heh
[15:29:55] <meffie> yeah, it has been a serious problem for those users.
[15:31:00] <meffie> it is not a new bug, so i dont think it is a blocker for 1.8.4.
[15:31:25] <wiesand> just saying that if it was easy to trigger with just memory pressure, my users would all the time ;-)
[15:32:01] <meffie> yeah, i agree, it's not too easy to trigger. do you see the OOM killer running often?
[15:32:50] <wiesand> occasionally, but at that point things have gone way wrong and a reboot seems expedient in any case
[15:34:37] <wiesand> more topics to discuss today?
[15:34:59] <kaduk@jabber.openafs.org/barnowl> I see Marcio submitted OS X 10.15 support
[15:35:17] <kaduk@jabber.openafs.org/barnowl> but I forget where in the macOS release cycle we are (i.e., if we are
playing catch-up)
[15:36:27] <kaduk@jabber.openafs.org/barnowl> Also, I rebased 10286 for the start of the bozo pthread conversion.
I didn't feel a need to rebase the whole series at once, though, and
reviewing that can probably run as a background-level task
[15:36:35] <wiesand> I think 10.15's ETA is "fall"
[15:37:06] <kaduk@jabber.openafs.org/barnowl> Okay.  The codename didn't sound familiar, but i couldn't rule out
that I've just been so focused on other things I didn't know it was
out.
[15:37:45] <kaduk@jabber.openafs.org/barnowl> It would be good to get another set of eyes on the CVE-2018-7168
patches, too.
[15:37:46] <wiesand> no blocker, but good we have something
[15:38:23] <meffie> which patches for CVE?
[15:38:41] <kaduk@jabber.openafs.org/barnowl> this is the thing where auristor clients sent negative acls as
positive acls
[15:39:14] <kaduk@jabber.openafs.org/barnowl> so we basically burned the RPC codepoint and allocated another; this
is just having openafs use the new RPC codepoint with some appropriate
fallback/knobs.
[15:39:30] <wiesand> 12942 and 13647
[15:39:31] <meffie> ah, yes.
[15:40:02] <meffie> thanks.
[15:40:09] <kaduk@jabber.openafs.org/barnowl> I don't mean to sound like I'm champing at the bit, but have you heard
anything from Andrew about when rxgk-phase2 will get pushed?  I think
I remember hearing "about a month" in Pittsburgh, which is totally
fine.
[15:40:47] <meffie> it's on his list of things todo, not sure on timing yet.
[15:40:56] <kaduk@jabber.openafs.org/barnowl> That works for me; thanks!
[15:42:00] <meffie> did we send an invite / creds to yadav ? he asked to join this meeting (2 per month)
[15:42:24] <kaduk@jabber.openafs.org/barnowl> I still have a todo to generate those creds; I have to remember what
command I used on the server to do it :(
[15:42:40] <meffie> ah, ok, thanks
[15:43:21] <meffie> that's all the topics i had.
[15:43:25] <kaduk@jabber.openafs.org/barnowl> But I might as well try to do it now, so I don't risk looking even
more irresponsible than  I already do
[15:43:30] <kaduk@jabber.openafs.org/barnowl> I don't think I have other topics for today; I just need to do more
"review review review"ing
[15:44:28] <wiesand> seems we're finished for today then - adjourn?
[15:44:41] <meffie> second. have a good weekend.
[15:45:01] <wiesand> Thanks a lot everybody!
[15:45:04] <kaduk@jabber.openafs.org/barnowl> yes, have a good weekend, and thanks!
[15:45:18] <wiesand> Off to water the garden…
[15:45:29] mvita leaves the room
[15:48:29] wiesand leaves the room
[15:59:06] meffie leaves the room
[17:12:15] mbarbosa leaves the room
[17:46:46] mbarbosa joins the room
[20:03:52] mbarbosa leaves the room
[20:37:09] mbarbosa joins the room
[21:23:22] mbarbosa leaves the room