Home
release-team@conference.openafs.org
Friday, October 19, 2018< ^ >
Room Configuration
Room Occupants

GMT+0
[02:18:39] mbarbosa joins the room
[02:20:50] mbarbosa leaves the room
[12:56:08] wiesand joins the room
[13:00:08] <kaduk@jabber.openafs.org/barnowl> greetings
[13:00:34] <kaduk@jabber.openafs.org/barnowl> Europe hits DST ... next weekend?
[13:02:32] <wiesand> hi ben - no, sunday the 28th
[13:03:10] <wiesand> in the us it's a week later, right?
[13:03:12] <kaduk@jabber.openafs.org/barnowl> Right, not the one in two days, but the following week
[13:03:15] <kaduk@jabber.openafs.org/barnowl> right
[13:03:46] meffie joins the room
[13:03:51] <meffie> good day
[13:04:24] <wiesand> so we'll have one meeting with a time skew - it shouldn't be a problem to keep the slot in us local time
[13:04:31] <wiesand> hello mike
[13:05:46] <wiesand> nb europe is getting serious about doing away with setting the clocks forth and back twice a year
[13:06:12] <meffie> yay! can i move to europe?
[13:07:00] <wiesand> shouldn't be too hard
[13:07:07] <kaduk@jabber.openafs.org/barnowl> I hear that immigration levels are a constant matter of discussion at
the highest political tiers
[13:07:18] <wiesand> but maybe wait until the dream has become reality
[13:07:34] <meffie> actually, my daughter is just back from europe, and hoping to relocate there permanently.
[13:07:52] <wiesand> yes, but that's not immigration by skilled it professionals from the 1st world
[13:08:46] <wiesand> we do have a real challenge originating from the chaos in northern africa and the middle east
[13:08:47] <meffie> sadly we've been in meetings all week, so not much progress on our gerrits
[13:09:11] <kaduk@jabber.openafs.org/barnowl> Ah, meetings: the great bane of productivity ;)
[13:09:18] <meffie> :)
[13:09:36] <wiesand> i'm still recovering (defeated the cavalry, now fighting the infantry…)
[13:10:03] <meffie> i hope you are feeling better
[13:10:03] <wiesand> so no news from me either
[13:10:24] <wiesand> better: yes. well: not yet
[13:10:50] <kaduk@jabber.openafs.org/barnowl> I know the feeling
[13:11:24] <wiesand> i tend to get serious sinus troubles *after* a flu/cold
[13:12:11] <wiesand> avoiding that is my #1 priority right now
[13:13:40] <wiesand> so what did i miss in openafs space while spending my time on herbal teas etc. past week?
[13:14:20] <meffie> hmm, one person tried to use -setpag ?
[13:14:46] <meffie> should we add a warning message or rip out aklog -setpag?
[13:15:27] <wiesand> i think it's the same guy who claims that 1.6.x aklog -setpag works if it's built right (with some keyring library iirc)
[13:15:44] <wiesand> looks like that no longer works with 1.8.x?
[13:17:08] <kaduk@jabber.openafs.org/barnowl> I thought 1.8.x had already ripped out the bulk of that code, but it
looks like it's still there, huh
[13:17:13] <wiesand> right, from a mail by him on june the 3rd: "The PAG's now are now done via keyctl . The configure of at least
openafss-1.6.21.1 checks this , look for KEYCTL_SESSION_TO_PARENT around line 23282
But the generated Makefile in src/aklog does not include -lkeyutils , there i change to
   AKLIBS = -L/opt/krb5/lib -lkrb5 -lk5crypto -lcom_err -lkrb5support -lkeyutils -lcrypt -lresolv
aklog -setpag then gets a token with a PAG (check with: keyctl show )
"
[13:19:37] mbarbosa joins the room
[13:19:38] <wiesand> trying that is still on my todo list
[13:19:54] <kaduk@jabber.openafs.org/barnowl> Adding -lkeyutils shouldn't change any behavior if the thing was
linking fine without it, IIUC
[13:20:39] <wiesand> i don't claim i understand what's going on, but i don't believe he's a fool
[13:21:01] <meffie> i've always was under the understanding that aklog -setpag was fundamentally broken. i dont know of anyone that uses it/depends on it. maybe we should have -setpag make a child process like pagsh?
[13:21:42] <kaduk@jabber.openafs.org/barnowl> I think aklog -setpag making a child process is too big of a behavior
change to be wise.
[13:22:20] <wiesand> +1
[13:23:23] <wiesand> obviously, rainer is using it in some way ;-)
[13:23:46] <wiesand> and i know of at least one site that was bitten when it stopped working
[13:24:28] <wiesand> but i don't think we should give this feature a high priority either
[13:24:59] <meffie> ok
[13:26:56] <wiesand> so, nothing on 1.6.x from my side this week again (shame on me)
[13:27:04] <wiesand> anyone?
[13:27:20] <wiesand> or on to 1.8.x? master?
[13:28:08] <meffie> is there anything you need for 1.6.24pre ?
[13:28:10] <kaduk@jabber.openafs.org/barnowl> Not much on those fronts, either, I think.
Did we talk much about the claim that getcwd was back, last week?
[13:29:25] <wiesand> mike: yes. spare time, health, motivation
[13:29:54] <wiesand> ben: i seem to recall mark wanted to look into it
[13:30:15] <wiesand> (easy to blame since not present…)
[13:30:48] <meffie> i'm looking at mark, let me poke him :)
[13:31:23] <meffie> poked
[13:32:13] <wiesand> my recollection could well be wrong (quite a few feverish dreams lately)
[13:32:33] <meffie> there was an issue related to apparmor
[13:33:33] <wiesand> ah, right, that's what mark wanted to look into
[13:33:50] <meffie> reported by Andreas Ladanyi on openafs-info
[13:34:39] <wiesand> the connection is that besides apparmor getcwd could be another facet of the issue
[13:35:35] <meffie> should i make a rt ticket on grand.central.org to track it?
[13:35:48] <kaduk@jabber.openafs.org/barnowl> Sure
[13:35:52] <meffie> ok
[13:37:07] <wiesand> my belief is that there is an issue, but it's not widespread/frequent (probably needs a fairly specific trigger)
[13:38:13] <meffie> ah good
[13:42:55] <kaduk@jabber.openafs.org/barnowl> We seem to be slowing down -- are we done for the day but just don't
know it yet?
[13:43:34] <wiesand> i was distracted for ~3 minutes, but yes
[13:44:17] <wiesand> adjourn?
[13:46:05] <wiesand> adjourn. thanks a lot everyone, cu next week!
[13:46:13] <meffie> thanks
[13:46:34] meffie leaves the room
[13:46:41] wiesand leaves the room
[16:16:50] mvita joins the room
[16:40:00] mvita leaves the room
[16:46:02] mbarbosa leaves the room