[15:24] <bdmurray> cjwatson: the reporter of bug 770041 replied are you still looking for someone to test it?
[15:24] <ubot2> Launchpad bug 770041 in debian-installer "natty server installation hang" [Undecided,New] https://launchpad.net/bugs/770041
[20:01] <cjwatson> bdmurray: I've been away all day, let me catch up
[20:11] <cjwatson> bdmurray: followed up
[20:19] <bdmurray> cjwatson: there was some discussion about it in #ubuntu-testing and jibel asked about set -ex and they said they were sure they set it
[20:20] <cjwatson> bdmurray: comment 10 - it might actually need to be set in the previous script
[20:20] <cjwatson> bdmurray: looks like I may have missed the user in question now, though?
[20:21] <bdmurray> cjwatson: ah yes it looks like they left #u-t
[20:22] <cjwatson> I sort of have half a suspicion, but it's not strong enough to make a late upload on the strength of my guesswork, and I don't yet understand why it wouldn't be consistently reproducible
[20:22] <bdmurray> from what I understand I don't think may people would run into it
[20:23] <cjwatson> I don't feel I have enough information to make a judgement on that, not knowing what causes it, TBH
[20:24] <cjwatson> I know it doesn't come up in trivial install-then-install-again cases, but for all I know there's some really common configuration that triggers it
[20:28] <cjwatson> I wonder if the fact that an LV snapshot is present might be relevant
[20:29] <cjwatson> but I won't be able to run a test until tomorrow morning, when I'm in London
[20:29] <bdmurray> cjwatson: how would you go about setting up the test?  I could give it a try.
[20:35] <cjwatson> I guess do an install, create a snapshot, try an install on top; TBH I don't actually use LVM snapshots myself but I think the command is lvcreate -s
[20:35] <cjwatson> (with a bunch of options, see its man page)
[20:37] <cjwatson> I don't see an obvious reason in the kernel why the bit I'm thinking of would fail for LVM snapshots, though