[00:32] <AndChat467409> Hey, I am getting the "Alert! /dev/disk/by-uuid/######### does not exist. Dropping to shell!" After trying to install windows on a empty part of my drive. Now nothing will boot, not even live CDs. Anybody know what's up? I'm beginning to think if bricked my computer.
[00:41] <lamont> has anyone seen a situation where traffic passing through a linux firewall (precise in this case) fails for win8 clients talking to a variety of far-side destinations, while it works (1) through a cable modem, and (2) if handed to a squid on the lan which then goes through the very firewall.  traffic looks sane on the firewall, haven't caught traffic yet in the working case
[00:41] <lamont> firewall nats, of course
[03:26] <infinity> lamont: I saw that on my friend's network for a while, but every attempt at diagnosis never allowed us to make any sense of it.
[03:26] <infinity> lamont: And upgrading to trusty made it go away, so we gave up caring.
[08:47]  * apw yawns theatrically
[08:53]  * ogra_ throws a single bee
[09:41] <niluje> hi
[09:41] <niluje> how can I debug this kernel panic? http://blackhole.brmzkw.info/2014-05-20/panic.txt
[09:41] <niluje> # uname -a
[09:41] <niluje> Linux vtydc2-p10cp-c01b16n12 3.14.4 #1 SMP Tue May 13 16:27:33 CEST 2014 armv7l armv7l armv7l GNU/Linux
[09:41] <niluje> # mount
[09:41] <niluje> /dev/nbd0 on / type ext4 (rw)
[09:41] <niluje> # ps auxw | grep nbd-client
[09:41] <niluje> root      1041  0.3  0.0   1312    72 ?        Ss   00:00   0:00 /sbin/nbd-client 172.16.1.22 4448 /dev/nbd0 -persist
[09:42] <niluje> I get this kernel panic when running : mkdir /root/toto && chmod 777 /root/toto && strace bonnie++ -d /root/toto -r 2048 -u nobody
[09:45] <apw> 3.14.4 ... does it occur with the 3.15 series ?
[09:48] <xnox> niluje: do you have a picture?
[09:48]  * xnox has kernel panics on ext4 with 3.15 package
[09:49] <apw> xnox, yeah i had a look at your picture, you mentioned the fs is dirty ?  is that just i hard powered off, or is it in serious trouble
[09:49] <apw> xnox, niluje's is http://blackhole.brmzkw.info/2014-05-20/panic.txt (3.14.4)
[09:50] <niluje> apw: didn't try with 3.15
[09:51] <xnox> apw: it's as dirty as it gets under systemd. As far as i can tell, because we are missing whole bunch of init-scripts (removed as not needed under upstart) systemd just hold the fs open and kills the power.
[09:51] <apw> xnox, quality error handling for the win
[09:51] <xnox> thus on next boot, either systemd/mountall do the "reboot required" thing and auto-reboot
[09:51] <xnox> then you get fsck
[09:51] <xnox> and then can finally boot again.
[09:52] <xnox> .... or not boot with 3.15.
[09:52] <apw> you are asking to not have a root doing that, i hope you don't have data on there you care about
[09:52] <niluje> (my real question is "how can I debug this?" and not "what is the problem?" :-))
[09:53] <xnox> i have little clue about niluje jibberish in the log.
[09:53] <xnox> but last line indicates that a reboot is required....
[09:53] <apw> niluje, well firstly i would read the source for n_tty_receive_buf_common to see what it might be memcpying, and then i'd be asking (as indeed i did) what versions do not have this issue
[09:53] <xnox> niluje: is this on powerpc? or an ARM board?
[09:54] <apw> niluje, to see what might have changed there to trigger this
[09:54] <niluje> xnox: an ARM board
[09:55] <niluje> apw: ok, I'm gonna try to setup a kernel 3.14 with the minimal set of options then I'll try to see if 3.15 fixes the problem or not
[09:55] <niluje> thanks
[10:18] <apw> niluje, i had a quick look at changes to n_tty since 3.14.4 but cannot see anything interesting, there is the output panic CVE thing but nothing input related
[10:19] <apw> niluje, unless in some what that could be the other side perhaps of the transaction which corrupted the tty
[10:20] <niluje> apw: I'm currently trying with a 3.2.34
[10:27] <apw> henrix, do i assume that netfilter bpf fix for the "lastest two cves" are coming down the stable route ?
[10:28] <henrix> apw: yep, that's correct
[10:28] <apw> henrix, great, then i can ignore them :)
[11:23] <zequence> infinity: So, only precise and quantal needed to be updated this time?
[11:27] <zequence> infinity: Reason why I'm asking is that the saucy bug report was handled differently from the rest, making the new unversioned one a duplicate of the original one
[11:27] <zequence> Never mind, I updated saucy. Seems it was like the rest
[11:49] <zequence> Well, a new ABI for saucy, in fact. Anyway, all uploaded now.
[11:52] <infinity> zequence: A new ABI for saucy?   That seems unlikely.
[11:53] <mlankhorst> infinity: can you review xorg-lts-trusty in precise?
[11:53] <infinity> zequence: Err, oh.  You rebased against the wrong one, that's why. :/
[11:54] <infinity> zequence: 22.38 is the one we're trying to catch up with.  But if you want to skip it and go straight to 23.39, that works too.
[11:54] <zequence> infinity: I don't rebase manually, so I don't get to control that :)
[11:55] <infinity> Oh.  I assume you picked a tag to rebase against or something.
[11:55] <apw> by default it updates against the latest master tag
[11:55] <apw> though you can "fix" that manually
[11:55] <infinity> Kay, well, nevermind then.  Easy enough to just let lowlatency skip the mess that this last cycle was and skip ahead to next cycle. :P
[11:56] <zequence> Cool. Less brainwork for me.
[11:57] <infinity> mlankhorst: I can after I've napped a bit.  I'm on the wrong end of night right now.
[11:58] <mlankhorst> ok
[13:00] <GortiZ> I've question about bug #1318531
[13:02] <GortiZ> It has been marked "triaged" without any comment and tagged as "cherry-pick reverse-bisect-done".
[13:02] <GortiZ> Does it means that the fix will be backported on the current LTS kernel (3.13) so that in future I'll only need to update to the latest KUbuntu 14.04 to make it work on my pc?
[13:09] <apw> oi ubottu
[13:13] <apw> GortiZ, no idea if penalvch is going to continue the hunt, but the next logical step is reverting that commit against tip and testing there
[15:01] <jsalisbury> **
[15:01] <jsalisbury> ** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting
[15:01] <jsalisbury> **
[15:35] <lamont> infinity: that is good to hear.  I have a trusty upgrade in my week.
[16:16] <GortiZ> apw: that commit doesn't need to be reverted but to be integrated (the newer works the older doesn't) anyway thanks for the answer. I'll wait and if nothing new happens I'll ask to penalvch :) (I was just wondering which of the two things falls in the "Triaged" definition)
[16:17] <apw> Triaged just means there is enough info for someone to progress it
[16:55] <jsalisbury> ##
[16:55] <jsalisbury> ## Kernel team meeting in 5 minutes
[16:55] <jsalisbury> ##