16:00:59 <frankbrockners> #startmeeting OPNFV BGS Arno release readiness - daily checkin
16:00:59 <collabot> Meeting started Thu May 14 16:00:59 2015 UTC.  The chair is frankbrockners. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:59 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:00:59 <collabot> The meeting name has been set to 'opnfv_bgs_arno_release_readiness___daily_checkin'
16:01:05 <frankbrockners> #info Frank Brockners
16:01:58 <fdegir> #info Fatih Degirmenci
16:02:08 <bryan_att> #info Bryan Sullivan
16:02:12 <pbandzi> #info Peter Bandzi
16:02:45 <morgan_orange> #info Morgan Richomme
16:02:48 <frankbrockners> hey folks, ... likely going to be a brief meeting today... most folks enjoy a public holiday
16:03:09 <[2]JonasB> #info Jonas Bjurel
16:03:14 <bryan_att> nice to have them ;-)
16:03:50 <trozet> #info Tim Rozet
16:04:50 <frankbrockners> ok - let's get started
16:05:11 <frankbrockners> #info daily updates on functest and automatic deploys on POD1 and POD2
16:05:26 <frankbrockners> Tim is pressed for time... - Tim could you go first?
16:05:49 <trozet> #info have a puppet module that will configure the external network
16:06:02 <trozet> #info need to do more testing on it to make sure its solid then will push a commit to gerrit
16:06:30 <trozet> thats it from me
16:07:10 <frankbrockners> #info automatic deploy on POD2: Build #27 failed, #28 was successful. Any idea what was the issue? (log says some issues with vagrant)
16:07:42 <trozet> #info #27 failed because the nodes did not pxe boot
16:08:01 <trozet> #info I didnt have a chance to determine why
16:08:20 <trozet> #info but if we see it again someone should go look at the remote console and see what happened
16:08:51 <trozet> #info in the log you see "rebuilding node", and then you see it waiting to try to ssh back to the node
16:09:12 <trozet> #info it will keep retrying to ssh to the node until linux is installed and the node comes up, if this does not happen within a timeout the deploy fails
16:10:02 <trozet> so if we see it again we should look at that node and see why the pxe failed on the remote console
16:10:09 <trozet> thats it from me, i got to run
16:10:20 <frankbrockners> thanks trozet - #29 is in progress - let's see what the result is going to be
16:10:38 <frankbrockners> anything else on automatic deploy? fdegir?
16:10:50 <fdegir> nope
16:10:54 <frankbrockners> cool
16:11:13 <frankbrockners> [2]JonasB: Quick update on POD1?
16:11:16 <[2]JonasB> #info No updates from my side
16:11:26 * frankbrockners seems that [1]JonasB got incremented...
16:11:42 <frankbrockners> thanks [2]JonasB
16:11:51 <frankbrockners> #info updates on functest
16:12:01 <frankbrockners> morgan_orange - any news?
16:12:23 <morgan_orange> #info POD 2 #29 shall perform the build/deploy/run
16:12:28 <morgan_orange> wait and see
16:12:35 <frankbrockners> including test?
16:12:36 <fdegir> #info Once the ongoing deployment #29 is successful, functest job will be triggered
16:12:43 <frankbrockners> cool!
16:12:59 <morgan_orange> last night it failed on deploy but it would not have work anyway, but it was a way to see how jobs are connected
16:13:10 <morgan_orange> #info change of job commited
16:13:27 <morgan_orange> #info (manual) functest on POD1 started
16:13:48 <frankbrockners> any early results?
16:13:59 <morgan_orange> #info functest/Pod1 vPing => TimeOut (bothe VM properly booted, network OK, but Ping not seen)
16:14:14 <morgan_orange> #info Tempest several tests run
16:14:29 <morgan_orange> #link https://wiki.opnfv.org/r1_tempest
16:14:47 <morgan_orange> #info first run showed 23 errors on 100 tests
16:15:06 <morgan_orange> # 4 root causes (3 common with Tests on POD2)
16:15:15 <morgan_orange> #info 4 root causes (3 common with Tests on POD2)
16:16:23 <morgan_orange> #info 3 common causes:  Quota exceeded, Multiple possible networks found, need more than 1 value to unpack
16:16:46 <morgan_orange> #info first bench scenario in progress
16:17:09 <morgan_orange> for ODL I do not know if pbandzi had time to perform some test on POD 1 or 2
16:17:12 <[2]JonasB> How does this compare with the tempest runs on Ericsson Lab?
16:17:42 <pbandzi> #info ODL no updates now
16:17:51 <morgan_orange> as far as I remember (need Jose for confirmation) there were 15 errors on E/// POD
16:18:00 <morgan_orange> but need to compare to see the difference
16:18:29 <[2]JonasB> Would be good to have a correlation to better understand wha could differ between the env.
16:18:37 <morgan_orange> and if OK we could also configure a nightly build/deploy/test on POD1
16:19:18 <fdegir> #info deployment #29 on POD2 failed
16:19:42 <fdegir> so no functest
16:19:59 <frankbrockners> :-(
16:20:09 <fdegir> I think we should disable automatic runs on pod2 somtrozet can take a look
16:20:34 <frankbrockners> fdegir: agreed
16:21:03 <fdegir> will send an email so trozet gets the info since he's left already
16:21:16 <frankbrockners> given the experience with POD2 autodeploy - should we try POD1 autodeploy the next night? Would at least give us some indication.
16:21:32 <frankbrockners> thanks fdegir
16:22:06 <morgan_orange> that is all for me
16:22:14 <[2]JonasB> frankbrockners: Depends on if Jose is planning to work on it tomorrow, which I think he did
16:22:48 <fdegir1> yes
16:22:50 <frankbrockners> [2]JonasB: good point. morgan_orange: do you know?
16:23:08 <morgan_orange> no
16:23:36 <morgan_orange> so wait for his green light
16:23:46 <frankbrockners> ok - then let's keep with the original plan to switch POD1 to autodeploy on Monday... We can still redecide tomorrow.
16:23:48 <morgan_orange> for POD 1
16:24:18 <frankbrockners> ok... anything else to cover for today? Do we have anyone from docs?
16:24:47 <frankbrockners> ... if not, we can call it a day.
16:24:53 <bryan_att> I'm still trying to get connected into the toolchain so I I can help. Minimal support so far.
16:24:55 <[2]JonasB> Thanks:-)
16:25:15 <bryan_att> As soon as I can get the toolchain working I'll help move the docs forward.
16:25:16 <[2]JonasB> quit
16:25:20 <bryan_att> as I can
16:25:27 <frankbrockners> bryan_att - could you check with Iben?
16:25:49 <frankbrockners> he might be around and he doesn't live in a holiday country...
16:26:09 <bryan_att> He's on the list. I was told by Victor to check with the LF help desk.
16:26:29 <frankbrockners> yup - ping aricg
16:26:39 <frankbrockners> he's on IRC
16:26:49 <bryan_att> I need to be able to commit without the overhead of gerrit reviews etc in order to test and improve the doc toolchain.
16:27:00 <bryan_att> i'll ping aricg
16:27:11 <frankbrockners> well - you'll need to go through gerrit...
16:27:24 <frankbrockners> you can't push directly.
16:27:49 <frankbrockners> ok... looks like we're done with status updates.
16:27:50 <bryan_att> if it's automatic then fine - but if I have to get votes from all committers for each change as I test/develop it wont happen
16:27:58 <frankbrockners> thanks everyone
16:28:03 <frankbrockners> #endmeeting