Home
release-team@conference.openafs.org
Friday, November 9, 2018< ^ >
Room Configuration
Room Occupants

GMT+0
[11:56:02] mbarbosa joins the room
[13:01:10] wiesand joins the room
[13:55:41] meffie joins the room
[13:55:50] meffie leaves the room
[13:56:18] meffie joins the room
[14:00:15] mvita joins the room
[14:00:51] <mvita> greetings
[14:00:54] <wiesand> Good morning
[14:01:27] <wiesand> I thought you were still on DST
[14:01:49] <mvita> no, the clocks changed last Sunda
[14:01:52] <mvita> Sunday
[14:02:17] <meffie> good day
[14:02:24] <mvita> (I hate my stand-in keyboard- I hope to get a new one today)
[14:02:43] <wiesand> I spent the better part of the last hour on deleting spam tickets in rt - the are becoming a real problem
[14:03:10] <meffie> yuck. i delete them about once a week too.
[14:04:25] <wiesand> I'm afraid that was about all my openafs-related activity this week.
[14:04:49] <mvita> HA!
[14:05:02] <meffie> i did not get much done this week, due to the election here.
[14:05:03] <mvita> But you fought the good fight.
[14:05:11] <mvita> ditto
[14:06:28] <wiesand> our Windows clients developed a BSOD problem lately, and it seems to be related to the AFS client
[14:06:33] <wiesand> BugCheck 50, {ffff8b8270218000, 0, fffff8022019d520, 0}
Probably caused by : AFSRedirLib.sys ( AFSRedirLib+273fc )
[14:07:10] <mvita> uh oh
[14:08:22] <wiesand> alas, it's not even easy to reproduce
[14:09:11] <meffie> are there stack traces?
[14:09:43] <mvita> or…. CORES?
[14:09:52] <meffie> minidumps?
[14:10:01] <mvita> yeah
[14:11:14] <wiesand> dunno, what is c:\windows\memory.dmp ? they're using some "dumpchk.exe" to examine them… windoze is not my turf at all
[14:11:49] <meffie> it's been too long for me to remember...
[14:12:02] <mvita> me too
[14:12:08] <mvita> but we can find out
[14:12:56] <wiesand> i'll advise the windows folks to send mail to -info if we don't make progress on this
[14:13:03] <mvita> teh google says:   C:/Windows/Minidump/*.dmp
[14:13:38] <mvita> full memory dump in C:/Windows/MEMORY.dmp
[14:13:58] <wiesand> so we even have full dumps
[14:14:01] <mvita> if they are enabled
[14:14:27] <wiesand> now if we had a support contract with someone ;-)
[14:14:50] <meffie> those are overrated :)
[14:15:00] <mvita> HEH
[14:15:19] <meffie> lol
[14:15:29] <mvita> <bites tongue clean through>
[14:16:27] <meffie> at any rate, sorry about the crash issue.
[14:17:02] <wiesand> it could be elated to our db servers changing the dns domain
[14:18:32] <wiesand> the old fqdn still resolves to the same ip address as before, but reverse lookups return the new fqdn
[14:18:50] <wiesand> linux clients obviously don't care
[14:19:16] <meffie> was there a windows update recently?
[14:19:43] <wiesand> not if they're telling the truth ;-)
[14:20:55] <meffie> heh
[14:21:02] <wiesand> maybe the windows clients need that data no longer leaked after the last security update
[14:22:37] <meffie> is there a ticket or any public info?
[14:22:39] <wiesand> anyway, back on topic? any news on that little linux problem we have since rc1?
[14:22:55] <wiesand> mike: not yet
[14:22:56] <mvita> Ah, wait
[14:23:05] <mvita> I think you are on to something with the DNS change
[14:23:20] <meffie> really?
[14:23:26] <mvita> CellServDB contains host names that are ignored by Unix client but used by Windows client
[14:23:39] <mvita> the ones after the #
[14:23:51] <meffie> correct.
[14:24:40] <mvita> anyway, it's worth an experiment
[14:25:22] <wiesand> I told them to change the CSDB (or remove it - the AFSDB records are correct)
[14:25:32] <wiesand> haven't hear back yet
[14:25:45] <wiesand> heard
[14:27:56] <wiesand> but fs listcells output looks fine
[14:31:04] <wiesand> did i miss anything spectacular going on in gerrit this week?
[14:31:30] <mvita> don't think so
[14:31:38] <mvita> I added a few little things
[14:32:42] <meffie> 35
36 Recently submitted for branch 'master':
37
38 13375 afs: refactor afs_GetDownDSlot()
39 13376 bozo: mark failed bnodes as stopped/shutdown
40 13377 vlserver: Introduce SDT probes
41 13374 ptserver: Check for -restricted in SPR_Delete
42 13373 ptserver: Fix AccessOK -restricted for SYSADMINID
43
[14:34:26] <wiesand> I'll try to get some homework done over the weekend.
[14:34:45] <meffie> and 13364 should be on that list
[14:37:11] <meffie> i'm currently working the patches to support 12586 (bozo: do not create client directories and symlinks)
[14:37:39] <wiesand> yes, these make sense to me
[14:40:30] <wiesand> have you switched over to the new buildbot master yet?
[14:41:38] <meffie> not yet, but may do that today.
[14:42:04] <wiesand> that's why it looks so familiar…
[14:42:06] <meffie> (democracy had a higher priority)
[14:42:15] <wiesand> sure
[14:42:32] <meffie> the new one is running, if you want to see it.
[14:42:59] <wiesand> where?
[14:43:15] <meffie> http://buildbot.openafs.org:8011/ <http://buildbot.openafs.org:8011/#/>
[14:43:24] <meffie> it is very different
[14:44:34] <meffie> so there will be some getting used to the new ui.
[14:44:52] <wiesand> yes' i'll notice the switch easily ;-)
[14:45:08] <meffie> yes.
[14:45:36] <meffie> i need to document the new setup on wiki.openafs.org, before i cut over.
[14:46:13] <meffie> then it should be a matter of just changing the port numbers and restarting.
[14:47:31] <meffie> sorry for the delay.
[14:47:44] <wiesand> no problem… are those F28 build failures something to be concerned about?
[14:49:22] <meffie> yes
[14:49:36] <meffie> but that is not a buildbot problem if i recall
[14:50:27] <meffie> we do not currently build on that platform
[14:50:47] <meffie> with warnings treated as errors that is.
[14:52:33] <wiesand> I see. Something about snprintf formats. At least it hasn't become harder to spot the problem :)
[14:52:46] <meffie> ok, good!
[14:53:15] <wiesand> We'll get used to it. And it pleases the eye a bit more than the old one.
[14:53:57] <meffie> it does seem to be much better interface, but it is different.
[14:54:01] <meffie> thanks.
[14:54:14] <wiesand> Anything else to discuss today?
[14:54:21] <mvita> linux
[14:54:35] <mvita> I am working on 2 issues for 4.20
[14:54:54] <mvita> one with current_time, one with GPL symbols
[14:55:05] <mvita> I hope to have something for the time issue today
[14:55:07] <wiesand> I had asked for that - probably got lost between the windows BSODs
[14:55:29] <wiesand> Thanks a lot for working on those!
[14:57:45] <wiesand> The GPL issue was about some paravirt function we don't use but pull in for some reason?
[14:58:02] <mvita> I think so
[14:58:36] <mvita> a symbol
[14:58:47] <mvita> for pv_ops
[14:58:55] <wiesand> At least this doesn't sound like a major threat. Right, pv_ops.
[15:00:16] <wiesand> More topics for today?
[15:00:49] <meffie> none here. have a good weekent
[15:00:51] <meffie> d
[15:01:17] <mvita> Stephan, is the bug check code x50 or decimal 50?
[15:01:20] <mvita> do you know?
[15:01:40] <wiesand> no idea, but I can find out
[15:01:51] <mvita> (typically they are expressed n hex)
[15:02:13] <mvita> which would mean this:  https://docs.microsoft.com/en-us/windows-hardware/drivers/debugger/bug-check-0x50--page-fault-in-nonpaged-area
[15:02:52] <mvita> all the bugcheck codes issued by the redirector itself are negative numbers, as far as I can tell in a quick look
[15:04:06] <wiesand> thanks for the pointer - some of the affected systems have ECC RAM though, so i't unlikely to be the main memory
[15:04:41] <wiesand> "antivirus software can also trigger this problem"… interesting
[15:04:53] <wiesand> of course those signatures change all the imte
[15:04:56] <wiesand> time
[15:06:24] <wiesand> Are we done for today?
[15:08:54] <meffie> yes, have a good weekend.
[15:08:56] meffie leaves the room
[15:09:17] <wiesand> Thanks a lot (especially for your patience with me).
[15:09:20] wiesand leaves the room
[15:33:16] mvita leaves the room
[15:39:22] meffie joins the room
[16:49:42] meffie leaves the room
[20:04:54] mvita joins the room
[22:02:30] mbarbosa leaves the room
[22:22:59] mvita leaves the room