openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 00:29 |
---|---|---|
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Create ubuntu/fedora test for pip-and-virtualenv https://review.openstack.org/407769 | 00:31 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add install-types as pip-and-virtualenv dep https://review.openstack.org/408248 | 00:31 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Fix pip-and-virtualenv to work with python3 https://review.openstack.org/407723 | 00:31 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Move pip-and-virtualenv source install to 10- https://review.openstack.org/408277 | 00:31 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Allow package-installs to parse python_version https://review.openstack.org/408293 | 00:31 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Allow package-installs to parse python_version https://review.openstack.org/408293 | 00:39 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Create ubuntu/fedora test for pip-and-virtualenv https://review.openstack.org/407769 | 00:43 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add install-types as pip-and-virtualenv dep https://review.openstack.org/408248 | 00:43 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Fix pip-and-virtualenv to work with python3 https://review.openstack.org/407723 | 00:43 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Move pip-and-virtualenv source install to 10- https://review.openstack.org/408277 | 00:43 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 00:48 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Create ubuntu/fedora test for pip-and-virtualenv https://review.openstack.org/407769 | 00:51 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add install-types as pip-and-virtualenv dep https://review.openstack.org/408248 | 00:51 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Fix pip-and-virtualenv to work with python3 https://review.openstack.org/407723 | 00:51 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Move pip-and-virtualenv source install to 10- https://review.openstack.org/408277 | 00:51 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Allow package-installs to parse python_version https://review.openstack.org/408293 | 00:51 |
*** jschwarz has quit IRC | 01:25 | |
*** jschwarz has joined #openstack-dib | 01:27 | |
openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add get_elements; preserve backwards compat for expand_dependencies https://review.openstack.org/407386 | 01:57 |
ianw | tfw you pull a thread and everything untangles | 01:58 |
greghaynes | Yes, see that py3 vs 2 deal | 02:37 |
greghaynes | I thought it was going to be one patch | 02:37 |
openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add get_elements; preserve backwards compat for expand_dependencies https://review.openstack.org/407386 | 03:15 |
*** jaianshu has joined #openstack-dib | 04:23 | |
*** jaianshu_ has joined #openstack-dib | 04:45 | |
*** jaianshu has quit IRC | 04:48 | |
openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add get_elements; preserve backwards compat for expand_dependencies https://review.openstack.org/407386 | 04:52 |
*** chhavi__ has joined #openstack-dib | 04:57 | |
*** sudipto has joined #openstack-dib | 05:04 | |
*** sudipto_ has joined #openstack-dib | 05:04 | |
openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Put MKFS_OPTS after filesystem type https://review.openstack.org/408408 | 05:05 |
*** narayrak has joined #openstack-dib | 05:17 | |
*** pmannidi has quit IRC | 05:47 | |
*** pmannidi has joined #openstack-dib | 06:29 | |
*** owalsh has joined #openstack-dib | 08:15 | |
*** jaianshu__ has joined #openstack-dib | 08:20 | |
*** jaianshu_ has quit IRC | 08:23 | |
*** jaianshu_ has joined #openstack-dib | 08:27 | |
*** jaianshu__ has quit IRC | 08:30 | |
*** jaianshu has joined #openstack-dib | 08:30 | |
*** jaianshu_ has quit IRC | 08:33 | |
*** hashar has joined #openstack-dib | 08:37 | |
*** strigazi_AFK is now known as strigazi | 08:39 | |
*** pmannidi has quit IRC | 09:28 | |
*** pmannidi has joined #openstack-dib | 09:30 | |
*** pmannidi has quit IRC | 09:39 | |
*** pmannidi has joined #openstack-dib | 09:41 | |
*** hogepodge has quit IRC | 09:45 | |
*** hogepodge has joined #openstack-dib | 09:46 | |
*** chhavi__ has quit IRC | 09:48 | |
*** pmannidi has quit IRC | 09:50 | |
*** owalsh has quit IRC | 10:02 | |
*** chhavi__ has joined #openstack-dib | 10:51 | |
*** owalsh has joined #openstack-dib | 10:54 | |
*** strigazi is now known as strigazi_AFK | 10:59 | |
*** hogepodge has quit IRC | 11:01 | |
*** chhavi__ has quit IRC | 11:12 | |
*** hogepodge has joined #openstack-dib | 11:21 | |
*** hogepodge has quit IRC | 11:26 | |
*** hogepodge has joined #openstack-dib | 11:38 | |
*** hogepodge has quit IRC | 11:48 | |
*** sudipto_ has quit IRC | 11:51 | |
*** sudipto has quit IRC | 11:51 | |
*** hogepodge has joined #openstack-dib | 12:00 | |
*** chhavi__ has joined #openstack-dib | 12:03 | |
*** jaianshu has quit IRC | 12:21 | |
openstackgerrit | Andrey Shestakov proposed openstack/diskimage-builder: Revert "Fix ironic-python-agent image not loading vfat mod" https://review.openstack.org/407955 | 12:45 |
*** narayrak has quit IRC | 13:37 | |
*** jaianshu has joined #openstack-dib | 13:47 | |
*** strigazi_AFK is now known as strigazi | 13:49 | |
*** jaianshu has quit IRC | 14:19 | |
*** sudipto_ has joined #openstack-dib | 14:31 | |
*** sudipto has joined #openstack-dib | 14:31 | |
*** sudipto_ has quit IRC | 15:25 | |
*** sudipto has quit IRC | 15:25 | |
*** owalsh has quit IRC | 16:05 | |
*** chhavi__ has quit IRC | 16:25 | |
*** hashar is now known as hasharCall | 17:02 | |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 17:27 |
*** hasharCall is now known as hashar | 18:07 | |
*** hashar has quit IRC | 18:16 | |
*** sudipto_ has joined #openstack-dib | 18:24 | |
*** sudipto has joined #openstack-dib | 18:24 | |
*** sudipto_ has quit IRC | 18:36 | |
*** sudipto has quit IRC | 18:36 | |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 19:04 |
*** strigazi is now known as strigazi_AFK | 19:26 | |
greghaynes | http://logs.openstack.org/88/408288/8/check/gate-dib-dsvm-functests-ubuntu-xenial/9c7656f/console.html#_2016-12-08_19_23_48_471594 | 19:52 |
greghaynes | thats a very odd failure | 19:52 |
ianw | hmm, yep | 19:57 |
ianw | greghyanes: repeatable? | 19:57 |
pabelanger | Ummm | 19:57 |
pabelanger | 2016-12-08 19:47:43,817 INFO nodepool.image.build.fedora-24: mount: none already mounted or /opt/dib_tmp/dib_build.bIPidWaE/mnt/proc busy | 19:57 |
pabelanger | we just got that | 19:58 |
pabelanger | on nb02.o.o | 19:58 |
pabelanger | weird | 19:58 |
pabelanger | working theory is we manually killed DIB, and the next build hit the existing mount | 20:00 |
greghaynes | ianw: I guess so? | 20:01 |
greghaynes | pabelanger: yea this seems like either a xenial or new dib bug | 20:01 |
greghaynes | im guessing the latter | 20:02 |
ianw | pabelanger: seems unlikely if the same thing is hitting in dib CI | 20:02 |
ianw | any interest in dib meeting? | 20:02 |
greghaynes | oh, today? | 20:03 |
greghaynes | if we were giong to I think itd be yesterday | 20:03 |
greghaynes | the odd thing about this bug is clearly it got through our CI, so something else must have changed | 20:04 |
ianw | ok, it's mounted in http://logs.openstack.org/88/408288/8/check/gate-dib-dsvm-functests-ubuntu-xenial/9c7656f/console.html#_2016-12-08_19_23_29_068819 | 20:07 |
ianw | then the next occurrence is another mount at http://logs.openstack.org/88/408288/8/check/gate-dib-dsvm-functests-ubuntu-xenial/9c7656f/console.html#_2016-12-08_19_23_48_466972 | 20:07 |
ianw | ahhh, here we go! http://logs.openstack.org/88/408288/8/check/gate-dib-dsvm-functests-ubuntu-xenial/9c7656f/console.html#_2016-12-08_19_23_48_325844 | 20:08 |
ianw | i think this is a red herring | 20:08 |
ianw | we haven't cleaned up properly | 20:08 |
ianw | in fact, it seems that "die" did not die -- http://logs.openstack.org/88/408288/8/check/gate-dib-dsvm-functests-ubuntu-xenial/9c7656f/console.html#_2016-12-08_19_23_48_325844 | 20:09 |
openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Put MKFS_OPTS after filesystem type https://review.openstack.org/408408 | 20:23 |
greghaynes | ianw: ohhh | 20:47 |
greghaynes | ianw: yea die is pretty broken | 20:47 |
greghaynes | I want to say it only exits 1 if the previous command did or something, its not an actualy exit | 20:48 |
greghaynes | so yea, I think its two bugs - the die didnt die and were failing to cleanup for some reason | 20:48 |
greghaynes | pabelanger: ^ | 20:49 |
greghaynes | pabelanger: you might be right in that you killed something which tickled the second bug | 20:49 |
pabelanger | I think so, we did -9 disk-image-create to abort a diskimage build in nodepool | 20:51 |
pabelanger | but didn't check mount to see what was left over | 20:51 |
ianw | pabelanger: thing is we can leave old mounts behind, but we're always doing the mounts into new tmp dirs | 20:53 |
ianw | since the die didn't die, we continued on, assuming that we'd cleaned up and unmounted ... but we hadn't | 20:54 |
pabelanger | maybe add a signal handler? | 20:55 |
pabelanger | or what is the right way to abort a build | 20:56 |
ianw | pabelanger: well ... i would say don't to be honest :) | 20:56 |
pabelanger | ha | 20:56 |
ianw | so many points probably don't have asynchronous exit handlers | 20:56 |
greghaynes | if you kill dib-run-parts it might just work | 20:56 |
ianw | we do try to clean up, modulo bugs like this one | 20:57 |
greghaynes | I suspect we need to do a bit of polling here | 20:57 |
ianw | but not much of a guarantee | 20:57 |
ianw | https://git.openstack.org/cgit/openstack/diskimage-builder/tree/elements/yum-minimal/root.d/08-yum-chroot#n274 | 20:58 |
ianw | # TODO : move this into a exit trap; and reconsider how | 20:59 |
ianw | # this integrates with the global exit cleanup path. | 20:59 |
greghaynes | oh right | 20:59 |
ianw | hmm, i wonder if the die() is actually working, but it's the subshell/flock hiding this https://git.openstack.org/cgit/openstack/diskimage-builder/tree/elements/yum-minimal/root.d/08-yum-chroot#n138 | 20:59 |
greghaynes | I dont see 'sudo $_RPM --root ...' so die has to be working | 21:03 |
ianw | yeah, i think what happens here is the flock returns ok, so the exit code for the script is 0 | 21:11 |
greghaynes | nice | 21:11 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 22:32 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Make dib-python use the default python for distro https://review.openstack.org/408288 | 22:35 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Dont generate qcow2's in tests https://review.openstack.org/408840 | 22:49 |
*** pmannidi has joined #openstack-dib | 23:02 | |
greghaynes | ianw: we keep getting fedora mirror fails :/ | 23:04 |
greghaynes | http://logs.openstack.org/40/408840/1/check/gate-dib-dsvm-functests-ubuntu-trusty/2665ea4/console.html#_2016-12-08_23_01_34_281833 | 23:04 |
*** pmannidi has quit IRC | 23:25 | |
*** pmannidi has joined #openstack-dib | 23:27 | |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Allow disto-specific mirror settings https://review.openstack.org/408850 | 23:34 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add support for using infra's mirrors in tests https://review.openstack.org/408851 | 23:34 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Allow disto-specific mirror settings https://review.openstack.org/408850 | 23:47 |
openstackgerrit | greghaynes proposed openstack/diskimage-builder: Add support for using infra's mirrors in tests https://review.openstack.org/408851 | 23:50 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!