| *** apuimedo has quit IRC | 00:00 | |
| *** raddaoui has quit IRC | 00:01 | |
| *** apuimedo has joined #openstack-ansible | 00:02 | |
| *** jmccrory has quit IRC | 00:04 | |
| *** jmccrory- is now known as jmccrory | 00:04 | |
| *** apuimedo has quit IRC | 00:10 | |
| *** apuimedo has joined #openstack-ansible | 00:11 | |
| *** cemmason has joined #openstack-ansible | 00:14 | |
| *** cemmason1 has quit IRC | 00:15 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 00:21 | |
| *** sdake has joined #openstack-ansible | 00:24 | |
| *** apuimedo has quit IRC | 00:30 | |
| *** cemmason has quit IRC | 00:31 | |
| *** apuimedo has joined #openstack-ansible | 00:31 | |
| *** cemmason has joined #openstack-ansible | 00:31 | |
| *** cemmason1 has joined #openstack-ansible | 00:35 | |
| *** cemmason has quit IRC | 00:36 | |
| *** apuimedo has quit IRC | 00:36 | |
| jmccrory | odyssey4me cloudnull could gate-openstack-ansible-dsvm-commit be added to the independent roles' merge gate list? | 00:37 |
|---|---|---|
| *** apuimedo has joined #openstack-ansible | 00:37 | |
| *** Mudpuppy has joined #openstack-ansible | 00:41 | |
| *** Mudpuppy has quit IRC | 00:42 | |
| *** apuimedo has quit IRC | 00:42 | |
| *** Mudpuppy has joined #openstack-ansible | 00:42 | |
| *** apuimedo has joined #openstack-ansible | 00:43 | |
| *** Mudpuppy has quit IRC | 00:43 | |
| *** Mudpuppy has joined #openstack-ansible | 00:44 | |
| *** spotz_zzz is now known as spotz | 00:50 | |
| *** Mudpuppy has quit IRC | 00:50 | |
| *** Mudpuppy has joined #openstack-ansible | 00:50 | |
| *** apuimedo has quit IRC | 00:52 | |
| *** apuimedo has joined #openstack-ansible | 00:52 | |
| *** cemmason1 has quit IRC | 00:56 | |
| *** cemmason has joined #openstack-ansible | 00:56 | |
| *** tlian has quit IRC | 00:57 | |
| *** apuimedo has quit IRC | 00:57 | |
| *** apuimedo has joined #openstack-ansible | 00:58 | |
| *** cemmason1 has joined #openstack-ansible | 00:59 | |
| *** cemmason has quit IRC | 00:59 | |
| *** eil397 has quit IRC | 01:02 | |
| *** apuimedo has quit IRC | 01:03 | |
| *** Mudpuppy has quit IRC | 01:04 | |
| *** tlian has joined #openstack-ansible | 01:04 | |
| *** apuimedo has joined #openstack-ansible | 01:04 | |
| *** Mudpuppy has joined #openstack-ansible | 01:04 | |
| *** Mudpuppy has quit IRC | 01:05 | |
| *** Mudpuppy has joined #openstack-ansible | 01:10 | |
| *** apuimedo has quit IRC | 01:11 | |
| *** apuimedo has joined #openstack-ansible | 01:12 | |
| *** jguy_ has joined #openstack-ansible | 01:12 | |
| *** tlian has quit IRC | 01:13 | |
| *** tlian has joined #openstack-ansible | 01:13 | |
| *** cemmason has joined #openstack-ansible | 01:15 | |
| *** cemmason1 has quit IRC | 01:15 | |
| *** cemmason has quit IRC | 01:18 | |
| *** cemmason has joined #openstack-ansible | 01:18 | |
| *** apuimedo has quit IRC | 01:19 | |
| *** apuimedo has joined #openstack-ansible | 01:20 | |
| openstackgerrit | wangyouwei proposed openstack/openstack-ansible: Trivial: Remove vim header from source files https://review.openstack.org/266624 | 01:23 |
| *** eil397 has joined #openstack-ansible | 01:23 | |
| *** apuimedo has quit IRC | 01:24 | |
| *** apuimedo has joined #openstack-ansible | 01:25 | |
| *** CheKoLyN has joined #openstack-ansible | 01:29 | |
| *** tlian has quit IRC | 01:30 | |
| *** apuimedo has quit IRC | 01:34 | |
| *** apuimedo has joined #openstack-ansible | 01:36 | |
| *** baker_ has quit IRC | 01:37 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add testing for dynamic_inventory.py https://review.openstack.org/242225 | 01:38 |
| *** apuimedo has quit IRC | 01:41 | |
| *** apuimedo has joined #openstack-ansible | 01:41 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add testing for dynamic_inventory.py https://review.openstack.org/242225 | 01:42 |
| *** cemmason has quit IRC | 01:44 | |
| *** cemmason has joined #openstack-ansible | 01:45 | |
| Mudpuppy | stand back, I'm bashing a bug lol | 01:57 |
| spotz | I apologize in advance for mudpuppy's bashing:) | 01:58 |
| *** eil397 has quit IRC | 01:59 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add testing for dynamic_inventory.py https://review.openstack.org/242225 | 01:59 |
| *** apuimedo has quit IRC | 02:00 | |
| *** apuimedo has joined #openstack-ansible | 02:01 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add testing for dynamic_inventory.py https://review.openstack.org/242225 | 02:04 |
| palendae | 4th time's the charm | 02:04 |
| spotz | of 5th or 6th...:) | 02:06 |
| *** apuimedo has quit IRC | 02:08 | |
| *** apuimedo has joined #openstack-ansible | 02:10 | |
| *** weezS has quit IRC | 02:11 | |
| *** cemmason has quit IRC | 02:14 | |
| *** cemmason has joined #openstack-ansible | 02:14 | |
| *** apuimedo has quit IRC | 02:16 | |
| *** apuimedo has joined #openstack-ansible | 02:17 | |
| *** apuimedo has quit IRC | 02:22 | |
| *** apuimedo has joined #openstack-ansible | 02:22 | |
| *** apuimedo has quit IRC | 02:27 | |
| *** apuimedo has joined #openstack-ansible | 02:27 | |
| *** jguy_ has quit IRC | 02:29 | |
| openstackgerrit | Dennis DeMarco proposed openstack/openstack-ansible: Correct CLI command for install verification https://review.openstack.org/266641 | 02:33 |
| Mudpuppy | YESSSS | 02:34 |
| spotz | I need a drink.... :) | 02:35 |
| *** cemmason has quit IRC | 02:42 | |
| *** Mudpuppy has quit IRC | 02:42 | |
| *** cemmason has joined #openstack-ansible | 02:43 | |
| *** apuimedo has quit IRC | 02:44 | |
| *** spotz is now known as spotz_zzz | 02:44 | |
| *** apuimedo has joined #openstack-ansible | 02:45 | |
| *** cemmason1 has joined #openstack-ansible | 02:46 | |
| *** cemmason has quit IRC | 02:47 | |
| *** cemmason1 has quit IRC | 02:49 | |
| *** cemmason has joined #openstack-ansible | 02:49 | |
| *** cemmason1 has joined #openstack-ansible | 02:53 | |
| *** cemmason has quit IRC | 02:53 | |
| *** apuimedo has quit IRC | 03:01 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 03:02 | |
| *** apuimedo has joined #openstack-ansible | 03:02 | |
| *** cemmason1 has quit IRC | 03:05 | |
| *** apuimedo has quit IRC | 03:07 | |
| *** apuimedo has joined #openstack-ansible | 03:08 | |
| *** cemmason has joined #openstack-ansible | 03:08 | |
| *** apuimedo has quit IRC | 03:15 | |
| *** apuimedo has joined #openstack-ansible | 03:16 | |
| *** woodard has quit IRC | 03:22 | |
| *** jguy_ has joined #openstack-ansible | 03:23 | |
| *** baker has joined #openstack-ansible | 03:28 | |
| *** apuimedo has quit IRC | 03:30 | |
| *** apuimedo has joined #openstack-ansible | 03:31 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 03:32 | |
| *** apuimedo has quit IRC | 03:35 | |
| *** apuimedo has joined #openstack-ansible | 03:36 | |
| *** baker_ has joined #openstack-ansible | 03:36 | |
| *** baker has quit IRC | 03:39 | |
| *** apuimedo has quit IRC | 03:43 | |
| *** apuimedo has joined #openstack-ansible | 03:44 | |
| *** apuimedo has quit IRC | 03:49 | |
| *** apuimedo has joined #openstack-ansible | 03:50 | |
| *** apuimedo has quit IRC | 03:57 | |
| *** apuimedo has joined #openstack-ansible | 03:57 | |
| *** apuimedo has quit IRC | 04:02 | |
| *** apuimedo has joined #openstack-ansible | 04:03 | |
| *** jguy_ has quit IRC | 04:05 | |
| *** apuimedo has quit IRC | 04:08 | |
| *** apuimedo has joined #openstack-ansible | 04:09 | |
| *** baker_ has quit IRC | 04:12 | |
| *** metral is now known as metral_zzz | 04:12 | |
| *** apuimedo has quit IRC | 04:18 | |
| *** apuimedo has joined #openstack-ansible | 04:19 | |
| *** apuimedo has quit IRC | 04:23 | |
| *** woodard has joined #openstack-ansible | 04:24 | |
| *** apuimedo has joined #openstack-ansible | 04:24 | |
| *** woodard has quit IRC | 04:28 | |
| *** apuimedo has quit IRC | 04:29 | |
| *** apuimedo has joined #openstack-ansible | 04:30 | |
| *** prometheanfire has quit IRC | 04:32 | |
| *** shausy has joined #openstack-ansible | 04:32 | |
| *** prometheanfire has joined #openstack-ansible | 04:33 | |
| *** apuimedo has quit IRC | 04:42 | |
| *** apuimedo has joined #openstack-ansible | 04:42 | |
| *** sdake_ has joined #openstack-ansible | 04:47 | |
| *** apuimedo has quit IRC | 04:47 | |
| *** apuimedo has joined #openstack-ansible | 04:48 | |
| *** sdake has quit IRC | 04:48 | |
| *** woodard has joined #openstack-ansible | 04:49 | |
| *** apuimedo has quit IRC | 04:52 | |
| *** sdake_ has quit IRC | 04:53 | |
| *** apuimedo has joined #openstack-ansible | 04:53 | |
| *** sdake has joined #openstack-ansible | 04:55 | |
| *** apuimedo has quit IRC | 05:04 | |
| *** apuimedo has joined #openstack-ansible | 05:05 | |
| *** raddaoui has joined #openstack-ansible | 05:06 | |
| *** apuimedo has quit IRC | 05:10 | |
| *** apuimedo has joined #openstack-ansible | 05:11 | |
| *** sdake has quit IRC | 05:16 | |
| *** apuimedo has quit IRC | 05:23 | |
| *** apuimedo has joined #openstack-ansible | 05:24 | |
| *** apuimedo has quit IRC | 05:28 | |
| *** apuimedo has joined #openstack-ansible | 05:30 | |
| *** metral_zzz is now known as metral | 05:33 | |
| *** shausy has quit IRC | 05:37 | |
| *** apuimedo has quit IRC | 05:37 | |
| *** apuimedo has joined #openstack-ansible | 05:38 | |
| *** apuimedo has quit IRC | 05:45 | |
| *** apuimedo has joined #openstack-ansible | 05:45 | |
| *** apuimedo has quit IRC | 05:56 | |
| *** apuimedo has joined #openstack-ansible | 05:56 | |
| *** apuimedo has quit IRC | 06:01 | |
| *** apuimedo has joined #openstack-ansible | 06:01 | |
| *** apuimedo has quit IRC | 06:06 | |
| *** apuimedo has joined #openstack-ansible | 06:08 | |
| *** javeriak has joined #openstack-ansible | 06:08 | |
| *** javeriak has quit IRC | 06:10 | |
| *** javeriak has joined #openstack-ansible | 06:10 | |
| *** apuimedo has quit IRC | 06:13 | |
| *** apuimedo has joined #openstack-ansible | 06:14 | |
| *** michaelgugino has quit IRC | 06:14 | |
| *** javeriak_ has joined #openstack-ansible | 06:14 | |
| *** CheKoLyN has quit IRC | 06:15 | |
| *** javeriak has quit IRC | 06:15 | |
| *** apuimedo has quit IRC | 06:18 | |
| *** apuimedo has joined #openstack-ansible | 06:20 | |
| *** apuimedo has quit IRC | 06:24 | |
| *** apuimedo has joined #openstack-ansible | 06:25 | |
| *** markvoelker has quit IRC | 06:27 | |
| *** apuimedo has quit IRC | 06:30 | |
| *** apuimedo has joined #openstack-ansible | 06:30 | |
| *** raddaoui has quit IRC | 06:31 | |
| *** apuimedo has quit IRC | 06:35 | |
| *** apuimedo has joined #openstack-ansible | 06:37 | |
| *** apuimedo has quit IRC | 06:41 | |
| *** apuimedo has joined #openstack-ansible | 06:41 | |
| *** phiche has joined #openstack-ansible | 06:45 | |
| *** apuimedo has quit IRC | 06:52 | |
| *** apuimedo has joined #openstack-ansible | 06:53 | |
| *** apuimedo has quit IRC | 07:03 | |
| *** apuimedo has joined #openstack-ansible | 07:03 | |
| *** robot_ has joined #openstack-ansible | 07:06 | |
| robot_ | Hi All | 07:06 |
| *** robot_ is now known as Guest52855 | 07:06 | |
| *** phiche1 has joined #openstack-ansible | 07:06 | |
| Guest52855 | if i create virtualenv inside any ansible container, and inn I connct it with other containers?stall some packages there, how ca | 07:07 |
| Guest52855 | if i create virtualenv inside any ansible container, and install some packages there, how can I connct it with other containers? | 07:07 |
| *** apuimedo has quit IRC | 07:08 | |
| *** apuimedo has joined #openstack-ansible | 07:08 | |
| *** phiche has quit IRC | 07:09 | |
| *** Guest52855 has quit IRC | 07:12 | |
| *** apuimedo has quit IRC | 07:25 | |
| *** apuimedo has joined #openstack-ansible | 07:26 | |
| *** markvoelker has joined #openstack-ansible | 07:28 | |
| *** apuimedo has quit IRC | 07:30 | |
| *** apuimedo has joined #openstack-ansible | 07:31 | |
| *** markvoelker has quit IRC | 07:33 | |
| *** apuimedo has quit IRC | 07:35 | |
| *** apuimedo has joined #openstack-ansible | 07:36 | |
| *** apuimedo has quit IRC | 07:47 | |
| *** woodard has quit IRC | 07:49 | |
| *** apuimedo has joined #openstack-ansible | 07:49 | |
| *** apuimedo has quit IRC | 07:53 | |
| *** apuimedo has joined #openstack-ansible | 07:54 | |
| *** mpavone has joined #openstack-ansible | 07:57 | |
| *** apuimedo has quit IRC | 07:59 | |
| *** adac has joined #openstack-ansible | 07:59 | |
| *** apuimedo has joined #openstack-ansible | 08:00 | |
| *** apuimedo has quit IRC | 08:05 | |
| *** apuimedo has joined #openstack-ansible | 08:06 | |
| *** apuimedo has quit IRC | 08:10 | |
| *** apuimedo has joined #openstack-ansible | 08:11 | |
| *** apuimedo has quit IRC | 08:20 | |
| *** apuimedo has joined #openstack-ansible | 08:21 | |
| *** apuimedo has quit IRC | 08:26 | |
| *** apuimedo has joined #openstack-ansible | 08:27 | |
| *** admin0 has joined #openstack-ansible | 08:29 | |
| *** apuimedo has quit IRC | 08:31 | |
| *** apuimedo has joined #openstack-ansible | 08:32 | |
| *** apuimedo has quit IRC | 08:36 | |
| *** apuimedo has joined #openstack-ansible | 08:37 | |
| *** apuimedo has quit IRC | 08:42 | |
| *** apuimedo has joined #openstack-ansible | 08:43 | |
| *** javeriak_ has quit IRC | 08:51 | |
| *** markvoelker has joined #openstack-ansible | 08:54 | |
| *** markvoelker has quit IRC | 08:59 | |
| *** apuimedo has quit IRC | 09:02 | |
| *** apuimedo has joined #openstack-ansible | 09:02 | |
| evrardjp | hello everyone | 09:21 |
| *** apuimedo has quit IRC | 09:22 | |
| *** apuimedo has joined #openstack-ansible | 09:23 | |
| odyssey4me | jmccrory it could, but that's not desirable - the whole point of the independent gating per role is for quicker turnaround... the parent repo with the plays in it will have multiple use-case tests which do integration of the roles and test that they all work together, whereas the role functional test is a minimal functional test | 09:30 |
| *** javeriak has joined #openstack-ansible | 09:31 | |
| *** apuimedo has quit IRC | 09:33 | |
| *** apuimedo has joined #openstack-ansible | 09:34 | |
| *** apuimedo has quit IRC | 09:38 | |
| *** apuimedo has joined #openstack-ansible | 09:40 | |
| *** electrofelix has joined #openstack-ansible | 09:41 | |
| odyssey4me | mattt hughsaunders what do you think about the approach in https://review.openstack.org/266530 - it's either that concise approach, or adding two tasks for the module load and load at boot | 09:47 |
| odyssey4me | I've prepped a revised patch, but I don't want to push it unless the current patch isn't accepted | 09:47 |
| *** apuimedo has quit IRC | 09:49 | |
| mattt | odyssey4me: why would you have a kernel 3.18? | 09:49 |
| javeriak | hey odyssey4me | 09:49 |
| odyssey4me | mattt kernel >3.18 is the new defaults for Ubuntu 14.04 and is therefore popping up all over the place | 09:51 |
| *** apuimedo has joined #openstack-ansible | 09:51 | |
| odyssey4me | without that kernel module, stuff doesn't work | 09:51 |
| hughsaunders | odyssey4me: that template may leave you with an empty list item, will that cause problems? | 09:51 |
| javeriak | pep8 seems to have failed on my commit; cant really see anything in the pipeline output, https://review.openstack.org/#/c/265946/ | 09:51 |
| odyssey4me | hughsaunders it seems to work just fine, even if it does create a blank line in /etc/modules | 09:52 |
| hughsaunders | odyssey4me: ok | 09:52 |
| odyssey4me | mattt hughsaunders the explanation for why br_netfilter is needed is in https://review.openstack.org/266021 - my patch is to cater for cases where the kernel is <3.18 | 09:52 |
| odyssey4me | javeriak that's odd - it's throwing errors in the playbooks/library/* files - I wonder why it's suddenly starting to do that | 09:54 |
| javeriak | odyssey4me maybe its a one off thing? dont know why pep8 would do that, anyway should i recheck? | 09:55 |
| odyssey4me | javeriak it's odd because that should be happening for every single review | 09:55 |
| javeriak | yes it should; but i cant really tell what exactly its throwing the error on from the output | 09:56 |
| odyssey4me | javeriak ah, it appears that it is happening for all reviews | 09:56 |
| odyssey4me | javeriak here's an example of the error: http://logs.openstack.org/46/265946/1/gate/gate-openstack-ansible-pep8/54c9c1c/console.html#_2016-01-12_21_24_21_247 | 09:57 |
| odyssey4me | ie W503 line break before binary operator | 09:57 |
| odyssey4me | there are several e402's too | 09:57 |
| javeriak | ah okay | 09:57 |
| javeriak | do you guys use any pep8 checking scripts on your commits? otherwise i can share mine if you like | 09:58 |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible-openstack_hosts: Only implement br_netfilter module for kernels above version 3.18 https://review.openstack.org/266530 | 09:58 |
| javeriak | for the good of the community :) | 09:58 |
| odyssey4me | javeriak the pep8 check can be done by executing 'tox -e pep8' on your workstation - that'll execute exactly the same check as in the gate | 09:59 |
| javeriak | also odyssey4me, let me know when u have a few minutes, id like to discuss a package versioning strategy i need to implement for our bits in OSAD | 09:59 |
| javeriak | oh okay cool, looks like some patch missed doing that check | 09:59 |
| odyssey4me | javeriak I think flake8 might have been updated, let me check | 10:00 |
| mattt | odyssey4me: yeah not sure of a better approach really | 10:00 |
| odyssey4me | javeriak pep8 went from 1.5.7 -> 1.7.0, that's why this is suddenly coming up | 10:02 |
| odyssey4me | mattt so the only other approach is to add two tasks that are specifically catering for that module, each with their own 'when' statement | 10:03 |
| odyssey4me | mattt hughsaunders well, that's the only other approach I can think of | 10:04 |
| javeriak | and the newer version implements stricter checks? that should be effecting a whole lot of stuff i imagine; do we need to comply with the new checks? do you need my help cleaning it up? | 10:04 |
| odyssey4me | javeriak yeah, we'll need to either decide that the checks aren't relevant to us and therefore skip them, or we need to clean up the code | 10:04 |
| odyssey4me | javeriak if you could help clean up that'd be fantastic - it'll all have to be done in one patch though | 10:05 |
| javeriak | sure, i can take a stab at it tonight; will tox -e pep8 only run on unmerged changes or the whole directory? | 10:06 |
| javeriak | you can decide meanwhile whats relevent or not? | 10:06 |
| hughsaunders | odyssey4me: could make it a list of dicts, with a name and min_version key. But seems overkill for one item | 10:07 |
| odyssey4me | tox -e pep8 will run whatever you have | 10:07 |
| odyssey4me | hughsaunders yeah, I thought of creating a dict with a module, and a condition for it being loaded... but it is very heavy handed for one item | 10:07 |
| odyssey4me | javeriak I'll do a clean up patch now quickly | 10:08 |
| javeriak | alright thanks | 10:08 |
| hughsaunders | odyssey4me: I agree. I fixed the commit message and +2d #rebel | 10:09 |
| odyssey4me | thanks hughsaunders - now we just need mattt to make a call :) | 10:09 |
| mattt | can you use {{ item }} in when? | 10:10 |
| mattt | if so you could guard against it being an empty list item | 10:10 |
| *** electrofelix has quit IRC | 10:13 | |
| odyssey4me | mattt it appears not | 10:13 |
| odyssey4me | mattt the only issue with the blank line is that it carries into /etc/modules... it doesn't seem to have any side effects | 10:13 |
| *** apuimedo has quit IRC | 10:14 | |
| *** apuimedo has joined #openstack-ansible | 10:15 | |
| odyssey4me | hmm, maybe a ternery on state would work | 10:15 |
| *** electrofelix has joined #openstack-ansible | 10:20 | |
| odyssey4me | mattt if I add the following to the lineinfile then it won't end up adding a blank line | 10:23 |
| odyssey4me | state: "{{ (item == '') | ternary('absent','present') }}" | 10:23 |
| odyssey4me | mattt hughsaunders thoughts on that idea? | 10:23 |
| mattt | odyssey4me: dunno, maybe it's not worht it for a blank line, i don't think that file cares about whitespace | 10:26 |
| *** apuimedo has quit IRC | 10:32 | |
| *** apuimedo has joined #openstack-ansible | 10:32 | |
| hughsaunders | odyssey4me: I'm ok with either | 10:35 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266832 | 10:41 |
| *** apuimedo has quit IRC | 10:44 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-galera_server: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266834 | 10:44 |
| *** apuimedo has joined #openstack-ansible | 10:45 | |
| *** permalac has joined #openstack-ansible | 10:46 | |
| *** apuimedo has quit IRC | 10:52 | |
| odyssey4me | mattt https://review.openstack.org/266530 is a gate unblocker - should we revert the patch that broke it, or vote this patch through? | 10:53 |
| *** apuimedo has joined #openstack-ansible | 10:53 | |
| *** markvoelker has joined #openstack-ansible | 10:55 | |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible: Use http request to check cinder api availability https://review.openstack.org/256414 | 10:57 |
| jmccrory | odyssey4me when: item != '' seems to work | 10:58 |
| *** markvoelker has quit IRC | 11:00 | |
| odyssey4me | jmccrory oh? can you pastebin your tasks that work? it didn't work for me | 11:01 |
| mattt | odyssey4me: how did that other commit gate then? | 11:03 |
| mattt | odyssey4me: i don't think it's worth reverting the other patch, since that just breaks stuff for others :) | 11:03 |
| odyssey4me | mattt it used a nodepool image which had the right kernel version | 11:03 |
| mattt | oh man infra need some consistency :( | 11:03 |
| odyssey4me | unfortunately it seems that not all nodepool images are created equal | 11:03 |
| jmccrory | http://paste.openstack.org/show/483728/ | 11:05 |
| *** apuimedo has quit IRC | 11:05 | |
| *** apuimedo has joined #openstack-ansible | 11:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Only implement br_netfilter module for kernels above version 3.18 https://review.openstack.org/266530 | 11:09 |
| *** alextricity25 has quit IRC | 11:09 | |
| *** hughsaunders has quit IRC | 11:10 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the nova ssh keys https://review.openstack.org/266839 | 11:10 |
| odyssey4me | heh, nice jmccrory :) | 11:11 |
| *** alextricity25 has joined #openstack-ansible | 11:11 | |
| *** apuimedo has quit IRC | 11:11 | |
| *** apuimedo has joined #openstack-ansible | 11:12 | |
| *** hughsaunders has joined #openstack-ansible | 11:12 | |
| *** apuimedo has quit IRC | 11:16 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the swift ssh keys https://review.openstack.org/266840 | 11:16 |
| *** apuimedo has joined #openstack-ansible | 11:17 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: [WIP] Redeploy venvs https://review.openstack.org/261293 | 11:20 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the keystone ssh keys https://review.openstack.org/266842 | 11:20 |
| *** apuimedo has quit IRC | 11:22 | |
| *** apuimedo has joined #openstack-ansible | 11:22 | |
| odyssey4me | jmccrory ping? is Michael around, or are you burning the midnight oil alone? | 11:26 |
| mattt | odyssey4me: hopefully the change to https://review.openstack.org/261293 is what you had in mind :) | 11:26 |
| jmccrory | 3:30 am here, having trouble sleeping. i think 6:30 for michael so he might be on with an hour or two | 11:27 |
| *** apuimedo has quit IRC | 11:27 | |
| *** apuimedo has joined #openstack-ansible | 11:28 | |
| *** sdake has joined #openstack-ansible | 11:29 | |
| *** jguy_ has joined #openstack-ansible | 11:29 | |
| odyssey4me | jmccrory I'm just thinking that the module backport is important, otherwise new deployments on updated kernels will fail... but obviously the backport is a combo of two patches so it's a little complex. | 11:31 |
| odyssey4me | jmccrory but we can wait for michael - he can learn how best to do it :) | 11:32 |
| jmccrory | wouldn't it just be your most recent patch that gets backported? | 11:32 |
| odyssey4me | jmccrory ah, you're quite right... although it won't be a clean pick | 11:33 |
| odyssey4me | ideally I think I'd like the change content from https://review.openstack.org/266530 but with more of the commit message content from https://review.openstack.org/266021 | 11:33 |
| *** sdake has quit IRC | 11:34 | |
| odyssey4me | mattt I've commented in the patch. | 11:34 |
| jmccrory | hmm well i'll shoot him an email to get on it on as soon as he does get in the office there :) | 11:34 |
| odyssey4me | mattt essentially if you do a sha256sum on the repo server, instead of a sha1sum, then you have direct content to feed to the get_url module instead of jumping through the hoops of doing a local sum manually | 11:35 |
| odyssey4me | hughsaunders mattt can we get votes through on https://review.openstack.org/266834 please to unblock that role's gate | 11:36 |
| hughsaunders | lgtm | 11:37 |
| mattt | odyssey4me: updated the review, but basically if we use get_url's sha256sum then we end up downloading the venv, which is precisely what we're trying to avoid | 11:38 |
| mattt | odyssey4me: passing sha256sum is to verify integrity of the downloaded file, which is not what we want (tho could do) | 11:39 |
| odyssey4me | mattt but if there is already a file, doesn't it check the existing file against the sha256sum? | 11:42 |
| odyssey4me | and then download a fresh one if it doesn't match | 11:42 |
| mattt | odyssey4me: no | 11:43 |
| mattt | "If a SHA-256 checksum is passed to this parameter, the digest of the destination file will be calculated after it is downloaded to ensure its integrity and verify that the transfer completed successfully. This option is deprecated. Use 'checksum'." | 11:43 |
| mattt | odyssey4me: you're not passing sha256sum to see if the file has chaned, but to ensure what is downloaded is valid | 11:44 |
| mattt | *changed | 11:44 |
| mattt | so with sha256sum you cannot avoid the download which will happen every time | 11:44 |
| *** apuimedo has quit IRC | 11:44 | |
| mattt | unless i'm reading this doc incorrectly :) | 11:45 |
| *** apuimedo has joined #openstack-ansible | 11:45 | |
| odyssey4me | mattt I'm not convinced that the doc is right, as we use the sha256sum for the rabbit downloads and I'm fairly certain that it does a local sha256sum if the dest file is present already and only downloads if the sha256sum doesn't match the value its given | 11:48 |
| *** apuimedo has quit IRC | 11:50 | |
| *** apuimedo has joined #openstack-ansible | 11:50 | |
| *** jguy_ has quit IRC | 11:54 | |
| *** markvoelker has joined #openstack-ansible | 11:56 | |
| *** apuimedo has quit IRC | 11:59 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266832 | 12:00 |
| *** apuimedo has joined #openstack-ansible | 12:00 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the keystone ssh keys https://review.openstack.org/266842 | 12:00 |
| *** markvoelker has quit IRC | 12:00 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the swift ssh keys https://review.openstack.org/266840 | 12:01 |
| mattt | odyssey4me: i think it downloads it every time, see https://gist.githubusercontent.com/mattt416/8010b18ad02d2e43164c/raw/43a3fdeb4f2a3d4d6d6622314a3b5753f52c3156/gistfile1.txt | 12:01 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the nova ssh keys https://review.openstack.org/266839 | 12:02 |
| mattt | odyssey4me: specifically, look at the "src" | 12:02 |
| *** javeriak has quit IRC | 12:04 | |
| odyssey4me | mattt if that is true, then it's pathetic... I do notice that 'attempts: 0' though | 12:04 |
| odyssey4me | lemme try something | 12:05 |
| odyssey4me | mattt heh, the rabbitmq play deletes the local deb every time it runs :/ | 12:09 |
| *** mlima_ has joined #openstack-ansible | 12:10 | |
| mlima_ | Hello | 12:10 |
| mlima_ | I ran ansible-playbook -i inventory/all-in-one -e @/etc/kolla/globals.yml -e @/etc/kolla/passwords.yml site.yml and I have this output now TASK [memcached : Starting memcached container] fatal: [localhost]: FAILED! => {"changed": false, "changes": ["{\"status\":\"Pulling repository localhost:5010/kollaglue/centos-source-memcached\"}\r\n", "{\"errorDetail\":{\"message\":\"Error while pulling image. Somebody knows why it ha | 12:10 |
| mattt | mlima_: are you in the right channel? :) | 12:12 |
| mattt | odyssey4me: well jeez :P | 12:12 |
| odyssey4me | mlima_ I think you'll need to visit #kolla :) | 12:12 |
| mlima_ | Ok | 12:13 |
| mlima_ | Sorry | 12:13 |
| *** mlima_ has left #openstack-ansible | 12:13 | |
| *** mlima_ has quit IRC | 12:13 | |
| *** apuimedo has quit IRC | 12:16 | |
| *** apuimedo has joined #openstack-ansible | 12:18 | |
| mattt | odyssey4me: let me test that w/ a standalone test to be sure | 12:19 |
| odyssey4me | mattt yeah, I'm doing the same | 12:20 |
| *** admin0 has quit IRC | 12:21 | |
| mattt | odyssey4me: yeah, if you specify sha256sum it still seems to download it | 12:26 |
| *** sdake has joined #openstack-ansible | 12:27 | |
| odyssey4me | mattt worse, regardless of whether the local file matches the checksum or not, if there's a local file with the dest name, it doesn't bother doing anything at all | 12:28 |
| mattt | odyssey4me: hence me adding force: true :) | 12:29 |
| odyssey4me | force: true was exactly what I was hoping to avoid | 12:30 |
| mattt | if you add force: true it will compare checksums of the downloaded file to what is already there, and replaces if different | 12:30 |
| *** apuimedo has quit IRC | 12:30 | |
| odyssey4me | because force: true does definitely always re-download the file | 12:30 |
| mattt | odyssey4me: right, but did you see the conditions i added to the task? | 12:30 |
| mattt | the task doesn't get run unless the local and remote checksums differ | 12:30 |
| odyssey4me | ok, let me do a fresh review pass now that I'm more certain of the behaviour here | 12:30 |
| mattt | okee | 12:30 |
| mattt | odyssey4me: also if you are busy this doesn't need to be done now | 12:31 |
| *** apuimedo has joined #openstack-ansible | 12:31 | |
| odyssey4me | mattt I'll trade you - if you can review https://review.openstack.org/266834 and https://review.openstack.org/266832 then I'll review https://review.openstack.org/261293 :) | 12:33 |
| mattt | sure :) | 12:34 |
| *** apuimedo has quit IRC | 12:43 | |
| *** woodard has joined #openstack-ansible | 12:43 | |
| *** apuimedo has joined #openstack-ansible | 12:44 | |
| odyssey4me | mattt I think that's good to go. I'd suggest changing the subject line, rebasing it on top of https://review.openstack.org/266832 (so that it passes all gate checks) then we can get some more feedback before you do the same for the other roles. | 12:44 |
| odyssey4me | I'd like to actually submit a patch upstream to Ansible to change this behaviour. If the local file checksum and the provided checksum don't match, it should automatically download a replacement. | 12:45 |
| odyssey4me | mattt ah, in fact if I read the 'checksum' (v2.0+) description correctly, it does what I want it to do | 12:46 |
| mattt | odyssey4me: in those reviews, why did we put the ansible imports at the bottom of the file? | 12:47 |
| odyssey4me | mattt I have no idea | 12:48 |
| *** admin0 has joined #openstack-ansible | 12:48 | |
| odyssey4me | I think that's a pattern provided in some of the examples ansible provides | 12:48 |
| mattt | odyssey4me: yeah that new functionality sounds right | 12:48 |
| *** apuimedo has quit IRC | 12:49 | |
| *** woodard has quit IRC | 12:49 | |
| mattt | odyssey4me: going to add a TODO also to remind us to change that when we move to ansible 2 | 12:49 |
| *** apuimedo has joined #openstack-ansible | 12:51 | |
| odyssey4me | mattt good plan | 12:51 |
| *** jguy_ has joined #openstack-ansible | 12:53 | |
| odyssey4me | blast, there's one I haven't seen with ansible > 1.6: http://logs.openstack.org/34/266834/1/gate/gate-openstack-ansible-galera_server-dsvm-ansible-functional-trusty/bdecaf6/console.html#_2016-01-13_12_51_03_071 | 12:54 |
| mattt | :( | 12:55 |
| *** apuimedo has quit IRC | 12:55 | |
| *** jguy_ has quit IRC | 12:56 | |
| *** apuimedo has joined #openstack-ansible | 12:56 | |
| *** markvoelker has joined #openstack-ansible | 12:57 | |
| *** admin0 has quit IRC | 12:58 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-galera_server: Ensure that ansible_processor_vcpus always has a default value https://review.openstack.org/266897 | 12:58 |
| mattt | blah looks like there are more neutron db migration issues :-/ | 13:01 |
| *** markvoelker has quit IRC | 13:02 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Allow re-deployment of venvs https://review.openstack.org/261293 | 13:07 |
| *** jguy_ has joined #openstack-ansible | 13:09 | |
| *** mgoddard has quit IRC | 13:12 | |
| *** mgoddard has joined #openstack-ansible | 13:13 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the keystone ssh keys https://review.openstack.org/266842 | 13:14 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the nova ssh keys https://review.openstack.org/266839 | 13:15 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use slurp to collect the swift ssh keys https://review.openstack.org/266840 | 13:16 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-galera_server: Ensure that ansible_processor_vcpus always has a default value https://review.openstack.org/266897 | 13:17 |
| *** admin0 has joined #openstack-ansible | 13:21 | |
| *** javeriak has joined #openstack-ansible | 13:23 | |
| *** javeriak has quit IRC | 13:23 | |
| *** javeriak has joined #openstack-ansible | 13:24 | |
| *** pbelamge_ has joined #openstack-ansible | 13:26 | |
| *** dslev has joined #openstack-ansible | 13:27 | |
| *** javeriak_ has joined #openstack-ansible | 13:30 | |
| *** javeriak has quit IRC | 13:31 | |
| *** markvoelker has joined #openstack-ansible | 13:31 | |
| *** apuimedo has quit IRC | 13:31 | |
| mhayden | mornin' | 13:32 |
| *** apuimedo has joined #openstack-ansible | 13:33 | |
| odyssey4me | mattt I assume that you're talking about http://logs.openstack.org/32/266832/2/check/gate-openstack-ansible-dsvm-commit/21613fb/console.html#_2016-01-13_13_26_06_765 ? | 13:35 |
| *** admin0 has quit IRC | 13:37 | |
| mattt | odyssey4me: yeah :( | 13:40 |
| *** javeriak has joined #openstack-ansible | 13:40 | |
| *** metral is now known as metral_zzz | 13:40 | |
| mattt | odyssey4me: bit confused why we're hitting a bug there tho, has anything changed w/ neutron in the last 24 hours ? | 13:40 |
| *** javeriak_ has quit IRC | 13:41 | |
| *** MCoLo has joined #openstack-ansible | 13:41 | |
| *** admin0 has joined #openstack-ansible | 13:41 | |
| odyssey4me | mattt I've rechecked to see if it pops up again. We're still at the same SHA, so we shouldn't be hitting any issues related to upstream changes. | 13:42 |
| mattt | odyssey4me: yah and we had stuff passing yesterday ... i have an AIO booted yesterday which runs fine | 13:42 |
| mattt | going to spin up an AIO now to check | 13:42 |
| *** apuimedo has quit IRC | 13:47 | |
| *** apuimedo has joined #openstack-ansible | 13:48 | |
| *** apuimedo has quit IRC | 13:55 | |
| *** MCoLo has quit IRC | 13:55 | |
| *** dslev has quit IRC | 13:55 | |
| *** apuimedo has joined #openstack-ansible | 13:56 | |
| *** javeriak_ has joined #openstack-ansible | 13:57 | |
| *** javeriak has quit IRC | 13:59 | |
| *** tlian has joined #openstack-ansible | 14:00 | |
| evrardjp | hello again | 14:02 |
| *** apuimedo has quit IRC | 14:05 | |
| *** apuimedo has joined #openstack-ansible | 14:06 | |
| *** admin0 has quit IRC | 14:08 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266834 | 14:08 |
| *** dslev has joined #openstack-ansible | 14:10 | |
| *** apuimedo has quit IRC | 14:12 | |
| *** apuimedo has joined #openstack-ansible | 14:13 | |
| *** woodard has joined #openstack-ansible | 14:15 | |
| *** woodard has quit IRC | 14:16 | |
| *** woodard has joined #openstack-ansible | 14:16 | |
| *** pai15 has joined #openstack-ansible | 14:17 | |
| *** apuimedo has quit IRC | 14:25 | |
| *** apuimedo has joined #openstack-ansible | 14:26 | |
| *** dslev has quit IRC | 14:29 | |
| *** admin0 has joined #openstack-ansible | 14:33 | |
| odyssey4me | mattt hughsaunders can we get this reviewed to further reduce gate failures: https://review.openstack.org/266897 | 14:38 |
| odyssey4me | o/ evrardjp | 14:38 |
| * mhayden ventures back over to the security midcycle | 14:39 | |
| *** apuimedo has quit IRC | 14:39 | |
| *** apuimedo has joined #openstack-ansible | 14:41 | |
| evrardjp | mhayden: that seems interesting :p | 14:43 |
| odyssey4me | mattt you're right, we now have a new neutron migration problem - https://review.openstack.org/266839 , https://review.openstack.org/266840 and https://review.openstack.org/266842 all failed as a results of it | 14:45 |
| mattt | odyssey4me: yeah got an instance here that failed also, troubleshooting now | 14:45 |
| *** apuimedo has quit IRC | 14:45 | |
| *** apuimedo has joined #openstack-ansible | 14:47 | |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 14:47 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 14:50 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 14:51 |
| *** apuimedo has quit IRC | 14:52 | |
| mattt | odyssey4me: ah! | 14:52 |
| mhayden | uh oh, the security folks are in our midst | 14:52 |
| odyssey4me | mattt ? | 14:52 |
| odyssey4me | mhayden everyone is welcome :) | 14:53 |
| mattt | odyssey4me: it's breaking on something plumgrid related | 14:53 |
| *** apuimedo has joined #openstack-ansible | 14:53 | |
| odyssey4me | mattt ah, I recall javeriak_ mentioning something about packages installing that perhaps shouldn't be | 14:53 |
| mattt | odyssey4me: yeah i think that's it | 14:54 |
| mattt | odyssey4me: it breaks neutron migrations, has nothing to do w/ the neutron migrations module thankfully | 14:54 |
| javeriak_ | odyssey4me the networking-plumgrid seems to be going into the venv regardless of plumgrid being enabled or not | 14:54 |
| javeriak_ | and if that gets installed when the venv is extracted the it will replace the ml2 totally | 14:55 |
| odyssey4me | thanks javeriak_ - it's time we figured out why :) | 14:55 |
| javeriak_ | odyssey4me on a similar note; i wanted to ask how i can manage the versions for it if wanted the version to be configurable through an extrenal parameter? | 14:56 |
| odyssey4me | javeriak_ as-in you'd like a more up to date version of the package on the repo server? | 14:57 |
| mattt | javeriak_: it makes sense we install it in the venv | 14:57 |
| javeriak_ | odyssey4me the easiest solution i can think of atm is taking it out from the venv enviroment entirely and installing from a particular upstream everytime | 14:57 |
| mattt | javeriak_: because the venvs are pre-built right, so they have to cater to all deployment scenarios | 14:57 |
| odyssey4me | javeriak_ yes, but that breaks the concept of repeatability and reliability - you never know which package you're getting | 14:58 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 14:58 |
| odyssey4me | javeriak_ this is why wheel versions are fixed per tag | 14:58 |
| odyssey4me | mattt yes, but it should only install into the venv if the plumgrid bits are enabled | 14:58 |
| mattt | odyssey4me: that doesn't make sense | 14:58 |
| *** michaelgugino has joined #openstack-ansible | 14:58 | |
| odyssey4me | mattt in this special case, it does - the plumgrid package overwrites parts of the neutron package | 14:59 |
| mattt | odyssey4me: so a plumgrid deploy cannot use the rpc-mirror for venvs? | 14:59 |
| mattt | javeriak_: on a side note, how come that package doesn't work with migrations? | 14:59 |
| *** Mudpuppy has joined #openstack-ansible | 14:59 | |
| *** Mudpuppy has quit IRC | 15:00 | |
| odyssey4me | ^ yeah, javeriak_ that will actually cause reliability issues too | 15:00 |
| *** Mudpuppy has joined #openstack-ansible | 15:00 | |
| odyssey4me | mattt my suggestion is that the plumgrid wheel is installed into the venv on-demand if it exists, over the top of whatever came from the repo venv package | 15:01 |
| javeriak_ | odyssey4me mattt; i understand guys; the resoning for venvs does make sense; but heres my scenario. i now have different versions of this package for kilo/liberty etc; we release minor versions every 2-3 weeks that go with a platform release; so i would typically want to be able to use a particular OSAD tag with any version of this package that i parameterize? | 15:01 |
| mattt | odyssey4me: that's messy, does the repo building stuff have the ability to not install those bits ? | 15:02 |
| odyssey4me | javeriak_ we typically tag every 2 weeks anyway, so if a consumer wishes the newer package, they just need to update to the later tag | 15:02 |
| odyssey4me | javeriak_ the only reason we haven't tagged for a month now is due to the holidays | 15:03 |
| *** alextricity_h has joined #openstack-ansible | 15:03 | |
| *** alextricity_h has quit IRC | 15:03 | |
| odyssey4me | mattt that's what I'm checking now - I have the feeling that it installs ".*_pip_packages" | 15:03 |
| mattt | me too :( | 15:03 |
| mattt | but if we lock down that version it should help? since this gated fine which means we have a working version of it somewhere :) | 15:03 |
| javeriak_ | mattt it does work with migrations | 15:04 |
| odyssey4me | javeriak_ I have the feeling that it may work for migrations of all stable releases of openstack, but not for Mitaka just yet... and we're gating on Mitaka | 15:04 |
| javeriak_ | guys question; will the rpc-mirror keep all versions of an upstream package? | 15:04 |
| odyssey4me | that's where we're seeing the breakdown | 15:05 |
| javeriak_ | yea we dont have a mitaka upgrade path yet | 15:05 |
| odyssey4me | javeriak_ the rpc-mirror keeps all versions of packages included in each release | 15:05 |
| mattt | javeriak_: we'll figure this out, it's no worries :) | 15:05 |
| odyssey4me | javeriak_ the mirror has package versions going back to icehouse :p | 15:05 |
| javeriak_ | mattt sure, let me knw if i can help look | 15:06 |
| mattt | javeriak_: on liberty/mitaka there should be two migration branches, "expand" and "contract" ... it looks like this plugin's migrations aren't separated | 15:06 |
| mattt | javeriak_: so when you run neutron-db-manage upgrade, it fails to find the necessary migration branch directories | 15:07 |
| michaelgugino | joining the convo late here, but the default rpc mirror in the docs does not have many of the point releases. This seems to work fine for everything but the pip wheel packages. | 15:07 |
| javeriak_ | matt right; can you guys disable including this package into the mitaka venev until i get that sorted? | 15:08 |
| michaelgugino | liberty is currently pointing to 12.0.3 (or was yesterday), and there is only 12.0.0 in the mirror. | 15:08 |
| mattt | javeriak_: look at https://github.com/openstack/neutron-lbaas/tree/master/neutron_lbaas/db/migration/alembic_migrations/versions versus https://github.com/openstack/networking-plumgrid/tree/master/networking_plumgrid/neutron/plugins/db/migration/alembic_migrations/versions | 15:08 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 15:08 |
| mattt | javeriak_: yeah sure | 15:09 |
| mattt | javeriak_: i am confused how this gated successfully tho | 15:09 |
| *** CheKoLyN has joined #openstack-ansible | 15:10 | |
| javeriak_ | matt yea i see what you mean | 15:10 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 15:11 |
| javeriak_ | and yes when enabled; it should have stopped all your neutron calls | 15:11 |
| openstackgerrit | Dennis DeMarco proposed openstack/openstack-ansible: Fixed output to match updated command https://review.openstack.org/266971 | 15:11 |
| odyssey4me | michaelgugino yes, none of the liberty tags are on the mirror as RPC does not yet deploy Liberty... I'm looking at trying to publish OSA packages/venvs onto tarballs.openstack.org as part of the automated gating process so that the project has a place to publish and so that all publishing is automated | 15:12 |
| michaelgugino | sounds good. | 15:13 |
| *** gparaskevas has joined #openstack-ansible | 15:13 | |
| odyssey4me | javeriak_ the inclusion of that package in the venv is a mistake anyway, so I'd like to ensure that it's not included by default and that it's only included when plumgrid config options are enabled | 15:13 |
| *** apuimedo has quit IRC | 15:14 | |
| *** apuimedo has joined #openstack-ansible | 15:16 | |
| mattt | brb for a few minutes | 15:16 |
| javeriak_ | odyssey4me, are you seeing this in all branches? or just master | 15:17 |
| javeriak_ | i looked through the recent venv changes and didnt see any check for inclusion on kilo | 15:17 |
| odyssey4me | javeriak_ the failure is only in master - but the inclusion of the package in the venv by default is likely in both master and liberty | 15:17 |
| odyssey4me | we only have venvs in master/liberty | 15:18 |
| javeriak_ | odyssey4me sorry, yea i meant liberty* | 15:18 |
| openstackgerrit | Travis McPeak proposed openstack/openstack-ansible-security: Adding Vagrant setup for deploying security-ansible https://review.openstack.org/266954 | 15:18 |
| javeriak_ | odyssey4me, so make it include only a specific version of this package in the venv, where do i specify it, in the requirements file? | 15:19 |
| javeriak_ | to * make it inlcude | 15:19 |
| *** baker has joined #openstack-ansible | 15:19 | |
| javeriak_ | and i assume that the rpc-mirror keeps them all from upstream | 15:20 |
| odyssey4me | javeriak_ for a particular environment you shouldn't have to do anything - if you repo-build it'll grab the latest allowed | 15:20 |
| *** spotz_zzz is now known as spotz | 15:20 | |
| odyssey4me | RPC uses the repo-clone and does not do a repo-build for production environments, so RPC always gets a static set of packages | 15:20 |
| javeriak_ | odyssey4me, nope but i dont want latest; id like a specific liberty version depending on a parameter i define in my external playbooks, is that possible? | 15:21 |
| *** KLevenstein has joined #openstack-ansible | 15:21 | |
| *** yaya has joined #openstack-ansible | 15:22 | |
| javeriak_ | like for example ive got 2.3, 2.4 and 2.5 up on pypi and id like my deployment to install any one of these with RPC 12 depending on what a parameter i choose | 15:22 |
| odyssey4me | javeriak_ ah, if you specifically want to cap it then you can change https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/defaults/main.yml#L348 to look something like https://github.com/openstack/openstack-ansible/blob/12.0.0/playbooks/roles/os_horizon/defaults/main.yml#L162 | 15:22 |
| odyssey4me | well, that'll work for any normal pip requirements parameters | 15:23 |
| odyssey4me | cloudnull ping? | 15:24 |
| javeriak_ | odyssey4me can i do this: http://pastebin.com/9eeYyfFY | 15:25 |
| *** apuimedo has quit IRC | 15:26 | |
| javeriak_ | odyssey4me, with "version" coming from somewhere else | 15:26 |
| *** pbelamge_ has quit IRC | 15:26 | |
| *** apuimedo has joined #openstack-ansible | 15:27 | |
| odyssey4me | javeriak_ it'll need to be something more like http://pastebin.com/bVE4ZgjJ | 15:28 |
| openstackgerrit | Dennis DeMarco proposed openstack/openstack-ansible: Correct CLI command for install verification https://review.openstack.org/266641 | 15:28 |
| odyssey4me | javeriak_ the variable will need to be namespaced, and you need to cater for the potential of there being no variable | 15:28 |
| javeriak_ | odyssey4me yea perfect thats what i had in mind. so can plumgrid_package_version come from a external yml and then that effects what it build into the venv | 15:30 |
| odyssey4me | javeriak_ correct | 15:30 |
| odyssey4me | but if that value is not provided, it'll just grab whatever version is available on pypi | 15:31 |
| *** galstrom_zzz is now known as galstrom | 15:31 | |
| javeriak_ | odyssey4me cool, let me give that a try | 15:31 |
| mattt | javeriak_ odyssey4me : why wouldn't we just lock down the version with the rest of the software we build? | 15:31 |
| *** apuimedo has quit IRC | 15:32 | |
| mattt | because we're inevitably going to run into further issues leaving networking-plumgrid uncapped | 15:32 |
| odyssey4me | javeriak_ another option is to place the package with its constraints into global-requirement-pins.txt | 15:33 |
| javeriak_ | guys afk, brb in a bit | 15:33 |
| odyssey4me | but that would require a requirements.txt format, and it seems to me that you're wanting to specify a package version per install | 15:33 |
| *** apuimedo has joined #openstack-ansible | 15:33 | |
| *** pegmanm has quit IRC | 15:33 | |
| *** pegmanm_ has joined #openstack-ansible | 15:34 | |
| odyssey4me | mattt I can't seem to find out why it is that the venv has the package | 15:34 |
| odyssey4me | mattt the package ony gets installed https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/tasks/plumgrid_config.yml#L26-L41 | 15:34 |
| mattt | odyssey4me: i would have thought we'd cap it in https://github.com/openstack/openstack-ansible/blob/master/playbooks/defaults/repo_packages/openstack_services.yml with the other neutron plugins? | 15:35 |
| odyssey4me | that task list only happens from here: https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/tasks/neutron_post_install.yml#L143-L145 | 15:35 |
| odyssey4me | mattt it seems that plumgrid wants the control of the version in their hands, rather than whatever's current | 15:36 |
| *** pai15 has quit IRC | 15:36 | |
| *** phalmos has joined #openstack-ansible | 15:36 | |
| *** pai15 has joined #openstack-ansible | 15:36 | |
| mattt | odyssey4me: yeah that makes sense | 15:37 |
| mattt | odyssey4me: maybe change the neutron_plumgrid_pip_packages var so it doesn't get slurped up by the venv builder | 15:37 |
| mattt | ? | 15:37 |
| odyssey4me | mattt I can't seem to find how it gets slurped by the venv builder | 15:37 |
| *** pegmanm has joined #openstack-ansible | 15:38 | |
| odyssey4me | mattt as far as I can see, the reference from which to build the package list comes from https://github.com/openstack/openstack-ansible/blob/master/playbooks/repo-build.yml#L24-L27 | 15:38 |
| *** pegmanm_ has quit IRC | 15:38 | |
| odyssey4me | that uses the lookup that looks for these variable names: https://github.com/openstack/openstack-ansible/blob/master/playbooks/plugins/lookups/py_pkgs.py#L42-L49 | 15:39 |
| *** pegmanm has quit IRC | 15:39 | |
| spotz | If anyone gets a chance look at https://review.openstack.org/#/c/266641/. It's a quicky I don't want to review as I helped with it | 15:39 |
| odyssey4me | which then builds a package list like so: https://github.com/openstack/openstack-ansible/blob/master/playbooks/plugins/lookups/py_pkgs.py#L350-L365 | 15:39 |
| *** pai15 has quit IRC | 15:39 | |
| *** pai15 has joined #openstack-ansible | 15:40 | |
| mattt | yep | 15:40 |
| mattt | so we just have to change the variable so it doesn't get picked up | 15:40 |
| mattt | but then the package won't be on the repo server, which would be a problem | 15:40 |
| *** fawadkhaliq has joined #openstack-ansible | 15:40 | |
| *** pegmanm has joined #openstack-ansible | 15:41 | |
| odyssey4me | mattt but neutron_plumgrid_pip_packages is not included as a key | 15:41 |
| odyssey4me | so I'm guessing that it uses a wildcard somehow, which I'm not seeing | 15:42 |
| mattt | yeah wildcard | 15:42 |
| mattt | since each role has different vards | 15:42 |
| mattt | *vars | 15:42 |
| *** yaya has quit IRC | 15:42 | |
| javeriak_ | fawadkhaliq ping | 15:42 |
| mattt | odyssey4me: what about changing var namespace to not include neutron? | 15:43 |
| mattt | that way it will end up repo server but hopefully not get installed into venv? | 15:43 |
| mattt | (where's cloudnull when you need him :P ) | 15:43 |
| *** mgoddard_ has joined #openstack-ansible | 15:45 | |
| odyssey4me | mattt I'm thinking more along the lines of the package list collection resulting in a flat list, but also a list which only has the <servicename>_pip_packages list | 15:45 |
| fawadkhaliq | hola javeriak_ | 15:46 |
| fawadkhaliq | whatsup | 15:46 |
| odyssey4me | the flat list should be used for the repo build, but the smaller list should be used for the venv | 15:46 |
| *** mgoddard has quit IRC | 15:48 | |
| mattt | odyssey4me: yeah that sounds sensible | 15:49 |
| *** alextricity25 has quit IRC | 15:51 | |
| javeriak_ | hey fawadkhaliq matt had a point about the plumgrid plugin migrations not being separated; was hoping you could shed some light; we are running into problems with neutron-db-manage upgrade, it fails to find the necessary migration branch directories | 15:52 |
| *** alextricity25 has joined #openstack-ansible | 15:52 | |
| cloudnull | morning | 15:52 |
| cloudnull | mattt: whats up ? | 15:53 |
| fawadkhaliq | that's weird. anything different we are doing in osad than DevStack script for same migration? All seems to work and pass testing and the gate. | 15:53 |
| cloudnull | odyssey4me: pong | 15:53 |
| odyssey4me | o/ cloudnull | 15:53 |
| *** alextricity25 has quit IRC | 15:54 | |
| odyssey4me | we have a special case which we need to find a suitable solution to | 15:54 |
| mattt | fawadkhaliq: don't think so, it's the upstream neutron-db-manage command that's failing on networking-plumgrid | 15:54 |
| *** alextricity25 has joined #openstack-ansible | 15:54 | |
| *** alextricty25 has joined #openstack-ansible | 15:54 | |
| fawadkhaliq | mattt: can you please point me to the failure you are looking at? | 15:54 |
| mattt | fawadkhaliq: sure thing | 15:54 |
| fawadkhaliq | mattt: thanks! | 15:55 |
| odyssey4me | cloudnull https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/defaults/main.yml#L347 is being included in the os_neutron venv... and that package overwrites some of the contents of the neutron wheel | 15:55 |
| mattt | fawadkhaliq: https://gist.github.com/mattt416/812c7fe2c14458eae66f | 15:55 |
| fawadkhaliq | mattt: thanks, looking at it | 15:56 |
| *** alextricty25 has quit IRC | 15:56 | |
| odyssey4me | cloudnull the problem is, of course, that the venv package install list is basically all ".*_pip_packages" found in the role, rather than just 'neutron_pip_packages' | 15:56 |
| cloudnull | correct | 15:56 |
| fawadkhaliq | mattt: javeriak_: question, why run PLUMgrid plugin migration with ML2? --config-file /etc/neutron/plugins/ml2/ml2_conf.ini | 15:56 |
| cloudnull | its pulls in everything _pip_packages | 15:56 |
| mattt | fawadkhaliq: look at https://github.com/openstack/neutron-lbaas/tree/master/neutron_lbaas/db/migration/alembic_migrations/versions versus https://github.com/openstack/networking-plumgrid/tree/master/networking_plumgrid/neutron/plugins/db/migration/alembic_migrations/versions | 15:56 |
| odyssey4me | cloudnull so I was about to look into how we could effectively have the list of packages to build on the repo server, and the venv package list being made of just the <role>_pip_package lists | 15:57 |
| odyssey4me | cloudnull I figure that it needs an update to the py_pkgs lookup, so I was about to give that a go | 15:57 |
| fawadkhaliq | mattt: yup, having expand/contract is not a requirement AFAIK. But let's narrow down the issue. | 15:57 |
| cloudnull | would you want that plumgrid package when using plumgrid within the venv ? | 15:57 |
| mattt | fawadkhaliq: yeah, that's a side issue we're cleaning up now, the package should not be installed at the moment | 15:57 |
| *** alextricity_h has joined #openstack-ansible | 15:57 | |
| odyssey4me | cloudnull my thinking is that if plumgrid is enabled, then the package can be installed into the venv after the pact | 15:58 |
| odyssey4me | *fact | 15:58 |
| fawadkhaliq | mattt: the package to supposed to be installed when we have core_plugin as PLUMgrid. | 15:58 |
| javeriak_ | odyssey4me cloudnull, like i see, it extracts and installs everything in the venv, and you would only want that plumgrid package when plumgrid is enabled otherwise its going to replace ml2 entirely | 15:58 |
| fawadkhaliq | mattt: definitely not with Ml2 | 15:58 |
| fawadkhaliq | s/package to/package is/ | 15:59 |
| mattt | fawadkhaliq: sure, but the migrations should run either way right? | 15:59 |
| cloudnull | couldnt you simply do a conditional include like we've proposed with the kernel modiule here: https://review.openstack.org/#/c/266530/5/defaults/main.yml | 15:59 |
| *** gparaskevas has quit IRC | 15:59 | |
| mattt | cloudnull: and if someone uses rpc-repo ? | 15:59 |
| alextricity_h | Hey..does anybody use the ZNC IRC bouncer? It should allow you to connect from multiple clients, right? | 16:01 |
| fawadkhaliq | mattt: not necessarily, one of the points of plugin decomposition is that this is all separate. In the run you shared, we have conflicts and seems like Ml2 conf has some requirement that is not there. | 16:01 |
| odyssey4me | cloudnull yeah, the package always needs to be on the repo, but it mustn't always be in the venv | 16:01 |
| alextricity_h | I left my client connected back at work, and am having trouble connecting to the bouncer from home. | 16:02 |
| cloudnull | the package will be there. being that its been built, however so long as we're carrying the package within the role, and the package cant coexist with neutron in a normal deployment, we're always going to have an issue like this | 16:02 |
| fawadkhaliq | mattt: so no there is no guarantee that plumgrid packaging will work and migration will work with a conf file from a different plugin. | 16:02 |
| mattt | fawadkhaliq: i see! | 16:02 |
| fawadkhaliq | mattt: javeriak_: it should fail in a better way. | 16:02 |
| openstackgerrit | Dennis DeMarco proposed openstack/openstack-ansible: Correct CLI command for install verification https://review.openstack.org/266641 | 16:03 |
| mattt | fawadkhaliq: so getting that package removed from the venv is all that way need to do for now | 16:03 |
| javeriak_ | cloudnull correct, the package cannot coexist in an ml2 installation | 16:03 |
| odyssey4me | cloudnull yep, which is why I thought it better that the packages installed into the venv should be more selective | 16:03 |
| fawadkhaliq | mattt: yes, go for it. having in the venv *only* when testing plumgrid bits makes sense to me. | 16:03 |
| cloudnull | thats fair. | 16:03 |
| *** apuimedo has quit IRC | 16:03 | |
| fawadkhaliq | mattt: you might run into the same issue with other plugins. | 16:04 |
| cloudnull | im missing some context here though, what causes this to break ? | 16:04 |
| odyssey4me | cloudnull with the plumgrid networking package in the venv, it overwrites some of the standard neutron stuff, which in turn causes the migrations to break | 16:04 |
| cloudnull | so the plumgrid package replaces files within neutron | 16:05 |
| odyssey4me | cloudnull yes | 16:05 |
| fawadkhaliq | odyssey4me: cloudnull: no :-) | 16:05 |
| fawadkhaliq | odyssey4me: cloudnull: let me elaborate a bit | 16:05 |
| *** apuimedo has joined #openstack-ansible | 16:05 | |
| fawadkhaliq | odyssey4me: cloudnull: neutron-db-manage when run with specific plugin configuration, it would be required to run DB migration for that OR the ones which are installed in the python entry points for db migrations. there could be multiple. Now, having a core plugin A and then using DB migration from core plugin B will break things. | 16:07 |
| odyssey4me | cloudnull ok, to be clearer, it causing http://logs.openstack.org/32/266832/2/check/gate-openstack-ansible-dsvm-commit/025f25e/console.html#_2016-01-13_14_45_07_805 currently | 16:07 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 16:07 | |
| *** admin0 has quit IRC | 16:07 | |
| odyssey4me | fawadkhaliq mattt cloudnull oh, does that mean then that out neutron db migration is perhaps a little too eager in collecting potential migrations to be done? | 16:08 |
| cloudnull | so is this being caused because we are not being specific enough when we do the db migration ? | 16:08 |
| *** ddaskal has joined #openstack-ansible | 16:08 | |
| fawadkhaliq | odyssey4me: not really. cloudnull: that's correct. | 16:08 |
| cloudnull | so sounds like we need to fix https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/tasks/neutron_db_setup.yml#L57-L68 | 16:09 |
| Mudpuppy | This white space is my moby whale | 16:09 |
| odyssey4me | how can we be more specific to avoid this issue then? | 16:09 |
| cloudnull | to assume less | 16:09 |
| mattt | odyssey4me: yeah i'm confused here | 16:09 |
| mattt | cloudnull: "assume less"? | 16:10 |
| mattt | we're running migrations as you should | 16:10 |
| fawadkhaliq | cloudnull: odyssey4me mattt javeriak_. so the command to migrate | 16:10 |
| cloudnull | we're running [u'/openstack/venvs/neutron-master/bin/neutron-db-manage', u'--config-file', u'/etc/neutron/neutron.conf', u'--config-file', u'/etc/neutron/plugins/ml2/ml2_conf.ini', u'upgrade', u'--expand'] | 16:10 |
| *** weezS has joined #openstack-ansible | 16:10 | |
| cloudnull | sounds like we're missing something | 16:10 |
| fawadkhaliq | cloudnull: odyssey4me mattt javeriak_. has two issues: one if ml2 conf and the other is it is specifically asking for the expand migration. | 16:10 |
| fawadkhaliq | s/one if/one is/ | 16:10 |
| *** yaya has joined #openstack-ansible | 16:11 | |
| mattt | but all neutron and subprojest use expand/contract | 16:11 |
| mattt | why would we do it differently? | 16:11 |
| javeriak_ | cloudnull odyssey4me matt fawadkhaliq; issues remains, we shouldn't be packaging and installing the plumgrid package everytime | 16:12 |
| mattt | javeriak_: that to me sounds like the issue :P | 16:12 |
| mattt | but i'm not sure what happens when i actually deploy plumgrid, i have no idea if migrations will run, i'm guessing not ? | 16:12 |
| fawadkhaliq | mattt: neutron subprojects implement both, expand/contract and older one | 16:12 |
| mattt | fawadkhaliq: OH | 16:12 |
| mattt | fawadkhaliq: yeah, we do expand and contract, just at different times | 16:12 |
| mattt | fawadkhaliq: we do expand, shut down neutron-server, do contract, then bring it back up | 16:13 |
| fawadkhaliq | mattt: i am wondering explicit expand will not call the older implementations. | 16:13 |
| javeriak_ | mattt true, in that case; maybe we could disable migrations if plumgrid is enabled? fawadkhaliq | 16:13 |
| fawadkhaliq | javeriak_: disable migrations? where will my DBs go :-( | 16:13 |
| mattt | hehe | 16:13 |
| odyssey4me | javeriak_ ick, not good - the expand/contract actions are to handle db changes needed when moving between versions | 16:13 |
| cloudnull | maybe we go to the old method if plumgrid is enabled | 16:14 |
| fawadkhaliq | cloudnull: thats one option | 16:14 |
| javeriak_ | oh... :)... odyssey4me mattt so this is only problematic with mitaka | 16:14 |
| mattt | javeriak_: liberty also iirc | 16:14 |
| *** metral_zzz is now known as metral | 16:14 | |
| cloudnull | my guess would be that other neutron subprojects will have the same issues | 16:14 |
| odyssey4me | javeriak_ for now, it might come up in other versions at a later stage | 16:14 |
| mattt | cloudnull: they don't | 16:14 |
| mattt | cloudnull: we have no issues w/ fwaas, lbaas, etc. | 16:14 |
| cloudnull | are you sure | 16:14 |
| mattt | yes i'm sure | 16:14 |
| cloudnull | like mediocura | 16:14 |
| fawadkhaliq | cloudnull: javeriak_ mattt. does only upgrade not trigger expand/contracts as well? I am not sure about it, I will have to check | 16:14 |
| odyssey4me | yeah, I'm not fond of putting a conditional in based on which plugin is used - that'll end up with a mess | 16:14 |
| cloudnull | cisco nexus | 16:14 |
| cloudnull | etc | 16:14 |
| mattt | i don't even knopw what mediocura is | 16:15 |
| mattt | fawadkhaliq: i think it does both yeah | 16:15 |
| cloudnull | mattt: http://www.midokura.com/midonet/openstack/ | 16:15 |
| mattt | cloudnull: do we deploy any of those other bits? i've not tested those at any point | 16:15 |
| cloudnull | *midokura | 16:15 |
| odyssey4me | it seems to me that we should be more selective in installing the packages in the first place | 16:15 |
| fawadkhaliq | mattt: so any specific reason for doing it explicitly? | 16:15 |
| mattt | fawadkhaliq: contract migrations should happen when your neutron-server is offline, so splitting them allows you to shut down neutron-server for only the contract migrations | 16:16 |
| cloudnull | mattt: point being that plumgrid is a neutron plugin and subproject and id suspect that other plugins and providers will have the same issues. | 16:16 |
| mattt | cloudnull: it's possible but i can only comment on stuff i've worked on | 16:16 |
| cloudnull | so have a framework to handle that would make sense | 16:16 |
| fawadkhaliq | mattt: thats right. | 16:17 |
| mattt | cloudnull: i'm not sure i understand | 16:17 |
| mattt | cloudnull: it's neutron-db-manage that is failing, that's now something we write | 16:17 |
| odyssey4me | I would agree on that, but it seems to be that we now have unnecessary db tables being created and contracted/expanded just because we have the packages installed... is that right? | 16:17 |
| openstackgerrit | Dennis DeMarco proposed openstack/openstack-ansible: Correct CLI command for install verification https://review.openstack.org/266641 | 16:17 |
| mattt | cloudnull: if it was our custom neutron module failing then i'd certainly think we'd need to fix | 16:17 |
| Mudpuppy | die white space die | 16:17 |
| fawadkhaliq | odyssey4me: I agree with you | 16:17 |
| odyssey4me | fawadkhaliq in that case, we need to ensure that the package is only installed when required - not always installed | 16:18 |
| fawadkhaliq | odyssey4me: absolutely | 16:18 |
| javeriak_ | odyssey4me, also we not enabling plumgrid in our current osa gates, there is no CI for it | 16:19 |
| odyssey4me | javeriak_ yes | 16:19 |
| mattt | fawadkhaliq: so neutron-db-manage upgrade heads works | 16:19 |
| javeriak_ | so you guys shouldnt see any issue, beause you're not testing enabling it | 16:19 |
| fawadkhaliq | odyssey4me: also that fixes the issue we are seeing. For PLUMgrid plugin, you will have something like neutron-db-manage upgrade heads --conf neutron.conf | 16:19 |
| odyssey4me | this is why I'm in favour of ensuring that the venv only has what it needs | 16:19 |
| cloudnull | mattt: the problem im trying to point out is, if we enable plumgrid the db wont migrate with our current tasks | 16:19 |
| fawadkhaliq | mattt: that's correct. same as DevStack | 16:19 |
| openstackgerrit | Eric Brown proposed openstack/openstack-ansible-security: Bump minimum required version of Ansible https://review.openstack.org/267009 | 16:20 |
| cloudnull | so simply removing the package from the base install wont "fix" the problem. | 16:20 |
| mattt | cloudnull: it won't no | 16:20 |
| odyssey4me | cloudnull that may be true, but it will unblock the gate and the subsequent issue can be worked out afterwards | 16:21 |
| fawadkhaliq | cloudnull: if we still do not have a solution, we could explore migrating to expand/contract but would need some time and backporting to Liberty :-( | 16:21 |
| odyssey4me | unless the suggestion here is to have a fallback task that tries the old migration method if the newer methods fail? | 16:22 |
| mattt | cloudnull fawadkhaliq : we could do a upgrade heads if database is empty | 16:22 |
| mattt | cloudnull fawadkhaliq : but not sure how future migrations would get applied if they're not in contract/expand branches :( | 16:22 |
| *** sdake_ has joined #openstack-ansible | 16:22 | |
| odyssey4me | but cloudnull with the package installed in the venv, we have unused db tables around... so that's also a problem | 16:22 |
| cloudnull | i'd be against simply unblocking the gate and "getting to it later". odyssey4me i think we need to work out how we migrate when these other plugins are being used. | 16:22 |
| cloudnull | odyssey4me: agreeded | 16:23 |
| mattt | cloudnull: yeah we will need to solve both | 16:23 |
| odyssey4me | so we effectivelyhave two issues - one is the unused tables being created due to the package being in the venv unnecessarily, the other is that the db migrations don't all work using the new method | 16:23 |
| cloudnull | it should be extracted. but we've also found an issue with a plugin we want to have support for | 16:23 |
| * odyssey4me nods | 16:23 | |
| fawadkhaliq | mattt: cloudnull: okay guys, let me explore migrating to expand/contract. | 16:24 |
| fawadkhaliq | mattt: for future, it should not break but it you never know what combination we are trying later. | 16:24 |
| mattt | fawadkhaliq: if there is something easier for us to implement i'm all ears | 16:24 |
| *** sdake has quit IRC | 16:25 | |
| cloudnull | fawadkhaliq: I'd suggest that, however i dont think your the only provider not using expand/contract so as a deployment project we need to work out how to deploy in this scenario. | 16:25 |
| *** sdake_ is now known as sdake | 16:25 | |
| fawadkhaliq | mattt: for now, would be great if we have db-manage for plumgrid specified as supported with the current version and in parallel, the expand/contract schema can be worked on. Dont want to block javeriak_ here | 16:26 |
| fawadkhaliq | cloudnull: exactly | 16:26 |
| odyssey4me | so mattt will you figure out how to also include the old migration method | 16:26 |
| fawadkhaliq | mattt: one more thing I noticed.. | 16:27 |
| fawadkhaliq | mattt: https://github.com/openstack/neutron-lbaas/tree/master/neutron_lbaas/db/migration/alembic_migrations/versions here, the migration scripts outside expand directory | 16:28 |
| fawadkhaliq | mattt: does the current db-manage command does the db migration for them? | 16:28 |
| fawadkhaliq | mattt: with the expand/contract flag | 16:29 |
| mattt | fawadkhaliq: yeah i was thinking of that earlier | 16:29 |
| fawadkhaliq | mattt: I have not tried that. If you could send me over the exact commands you use, I could give it a try and share results. | 16:30 |
| mattt | fawadkhaliq: i want to make sure we are applying all the right migrations, i know i looked at this in detail a while back but it's been a while | 16:30 |
| fawadkhaliq | mattt: makes sense. Maybe you or javeriak_ could point me to the db migration commands we use, I can give it a shot on my dev env and see if the migrations under questions get picked up | 16:32 |
| *** jthorne has joined #openstack-ansible | 16:32 | |
| *** galstrom is now known as galstrom_zzz | 16:33 | |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible: Add missing kernel module br_netfilter to openstack_hosts role https://review.openstack.org/267013 | 16:33 |
| *** apuimedo has quit IRC | 16:33 | |
| michaelgugino | back porting br_netfilter change to liberty | 16:34 |
| *** mpavone has quit IRC | 16:35 | |
| *** apuimedo has joined #openstack-ansible | 16:35 | |
| *** pai15 has quit IRC | 16:36 | |
| cloudnull | odyssey4me: fawadkhaliq mattt is there an issue for this ? | 16:37 |
| cloudnull | in launchpad ? | 16:37 |
| odyssey4me | cloudnull nope, it only hit us midday and we've been trying to triage it | 16:37 |
| mattt | fawadkhaliq: it looks like running --expand runs all the old pre-liberty migrations | 16:37 |
| javeriak_ | odyssey4me coming back to my original versioning question, yes im looking to specify a package version per install, and a requirements file only allows you to specify ranges; | 16:38 |
| fawadkhaliq | mattt: okay, and misses everything post liberty which does not fall under expand? | 16:38 |
| *** pai15 has joined #openstack-ansible | 16:39 | |
| spotz | Oh that's the one mudpuppy ran into when we were building test environments | 16:39 |
| javeriak_ | odyssey4me so i'd like to parameterize the definition like showed and then hopefully that works for venv packaging as well; | 16:39 |
| fawadkhaliq | mattt: if it does not miss the newer ones as well, I could trick db-migration my adding a placeholder in the expand and keep the existing ones as is. | 16:40 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: [WIP] Fix neutron migrations with optional providers https://review.openstack.org/267019 | 16:40 |
| cloudnull | odyssey4me fawadkhaliq mattt ^ | 16:40 |
| fawadkhaliq | mattt: that would be quicker and will work without any issues and future migration would go into expand | 16:40 |
| javeriak_ | fawadkhaliq https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_neutron/tasks/neutron_db_setup.yml | 16:41 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Add missing kernel module br_netfilter to openstack_hosts role https://review.openstack.org/267013 | 16:41 |
| odyssey4me | michaelgugino thanks - I modified the commit message slightly | 16:41 |
| michaelgugino | sounds good. | 16:41 |
| fawadkhaliq | javeriak_: thanks. got it. I will give it a shot now. Stay tuned. | 16:42 |
| javeriak_ | cloudnull your patch takes the plumgrid out of the venv entirely? | 16:43 |
| palendae | michaelgugino: Were you seeing failures against master? I've got a gate test failing on getting br_netfilter because it was on 3.13 | 16:43 |
| javeriak_ | plumgrid* package | 16:43 |
| cloudnull | javeriak_: yes | 16:43 |
| cloudnull | but will install it in the venv if the plumgrid provider is used | 16:43 |
| cloudnull | and will index / build the package within the base repo | 16:44 |
| michaelgugino | I didn't see any failures here | 16:44 |
| javeriak_ | cloudnull i dont see that bit? as in where does the 'neutron_optional_plumgrid_pip_packages' become a part iof the main neutron-pip-packages for plugin=plumgrid? | 16:45 |
| *** adac has quit IRC | 16:45 | |
| javeriak_ | cloudnull, oh you mean it wont be packaged within the venv at alll from now on, only get installed if enabled | 16:45 |
| cloudnull | yes | 16:46 |
| cloudnull | https://review.openstack.org/#/c/267019/1/playbooks/roles/os_neutron/tasks/plumgrid_config.yml | 16:46 |
| cloudnull | but the package will be available in the repo-server | 16:46 |
| cloudnull | regardless | 16:46 |
| javeriak_ | cloudnull right, ive been proposing that all along :), it makes things easier for me to manage versioning wise | 16:46 |
| * cloudnull high-fives javeriak_ :) | 16:46 | |
| odyssey4me | cloudnull feedback on https://review.openstack.org/267019 | 16:46 |
| javeriak_ | cloudnull perfect! | 16:46 |
| odyssey4me | palendae how old was your clone? | 16:47 |
| palendae | odyssey4me: I had updated it last night. I'll double check | 16:47 |
| mattt | fawadkhaliq: sorry just looking at this here | 16:48 |
| palendae | But the patch was ancient, something I started in November and just recently picked back up | 16:48 |
| odyssey4me | palendae if you cloned master, and cloned before https://review.openstack.org/266530 merged, then I would suggest rebootstrapping ansible to pick up the newer roles | 16:48 |
| palendae | 6 hours ago, so yeah :p | 16:48 |
| palendae | Thanks | 16:48 |
| palendae | I'll rebase | 16:48 |
| mattt | fawadkhaliq: i'm seeing a slight discrepancy, but it looks like all old migrations are run when you --contract | 16:49 |
| odyssey4me | palendae if it's master, rebasing won't work - the role concerned is not in git - you need to bootstrap-ansible to collect the updated roles | 16:49 |
| palendae | Ah, right, reading comprehension. So just a recheck | 16:49 |
| *** phiche has joined #openstack-ansible | 16:49 | |
| palendae | And I have another change that needs to be pushed anyway | 16:49 |
| *** yatin has joined #openstack-ansible | 16:50 | |
| palendae | Basically dsvm-commit on https://review.openstack.org/#/c/242225/8 failed | 16:50 |
| palendae | But I need to remove some of my vim cruft from the tox.ini | 16:50 |
| mattt | fawadkhaliq: sorry, --expand not --contract | 16:50 |
| *** phiche1 has quit IRC | 16:51 | |
| fawadkhaliq | mattt: cool. Let me know if I can pull something quick. Thanks. I will update. | 16:51 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: [WIP] Fix neutron migrations with optional providers https://review.openstack.org/267019 | 16:51 |
| fawadkhaliq | javeriak_: can you make progress in the meantime? | 16:51 |
| *** weezS has quit IRC | 16:51 | |
| fawadkhaliq | mattt: javeriak_ I will feel bad if you guys blocked because of me :-( | 16:52 |
| mattt | fawadkhaliq: don't worry, we're blocked pretty much every other day by something | 16:53 |
| cloudnull | ++ :) | 16:53 |
| javeriak_ | fawadkhaliq im good, cloudnull's patch for removing our package from venv will take care of the current issues in the gate. | 16:53 |
| fawadkhaliq | mattt: lol | 16:54 |
| mattt | cloudnull: sweet, that patch looks good at 30,000 feet | 16:56 |
| *** yatin has quit IRC | 16:57 | |
| *** mgoddard_ has quit IRC | 16:57 | |
| *** mgoddard has joined #openstack-ansible | 16:57 | |
| *** woodard has quit IRC | 16:57 | |
| mattt | fawadkhaliq: had a look at some other provider plugins, they all seem to have contract/expand dirs, but not tested them obviously | 16:58 |
| mattt | fawadkhaliq: and by all i mean the three random ones i chose :P | 16:58 |
| *** woodard has joined #openstack-ansible | 16:58 | |
| mattt | fawadkhaliq: actually found one that doesn't | 17:00 |
| fawadkhaliq | mattt: :-) I will move expand/contract very soon, sir! | 17:00 |
| javeriak_ | yea mattt, that one would be us.. :) | 17:01 |
| mattt | javeriak_: nah another one! :) | 17:01 |
| *** jthorne has quit IRC | 17:01 | |
| mattt | fawadkhaliq: i'm heading out shortly ... i'll be back online tomorrow God willing to look at this further :) if you can think of anything we can implement on this side to better guard us against this i'm all ears | 17:02 |
| mattt | but i think separating out migrations gives us more flexibility and allows us to shorten the time that neutron-server has to be down for running migrations | 17:02 |
| *** CheKoLyN has quit IRC | 17:02 | |
| fawadkhaliq | mattt: sure. will think about it and let you know. agree on the benefits. anyway, i will drop a note. later! | 17:04 |
| odyssey4me | cloudnull last tweak on https://review.openstack.org/267019 then I think the WIP can come out and we can see if it gates | 17:05 |
| *** Nepoc has quit IRC | 17:06 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: [WIP] Fix neutron migrations with optional providers https://review.openstack.org/267019 | 17:07 |
| *** galstrom_zzz is now known as galstrom | 17:07 | |
| cloudnull | odyssey4me: done | 17:07 |
| *** woodard has quit IRC | 17:07 | |
| *** woodard has joined #openstack-ansible | 17:07 | |
| *** Nepoc has joined #openstack-ansible | 17:07 | |
| odyssey4me | cloudnull mattt ok, so we have 3 issues to resolve here - one is the new pep8 violations (solved by https://review.openstack.org/266832 ), another is the incorrect venv package inclusion (solved by https://review.openstack.org/267019 ) and the other is the fixing up of the db migrations (fix ongoing) | 17:10 |
| odyssey4me | in order to get https://review.openstack.org/267019 merged we need https://review.openstack.org/266832, but they both need each other | 17:10 |
| odyssey4me | should we perhaps temporarily cap the pep8 version to allow https://review.openstack.org/267019 to pass, then uncap it in https://review.openstack.org/266832 ? | 17:11 |
| *** permalac has quit IRC | 17:12 | |
| *** apuimedo has quit IRC | 17:19 | |
| *** apuimedo has joined #openstack-ansible | 17:20 | |
| *** yatin has joined #openstack-ansible | 17:20 | |
| *** CheKoLyN has joined #openstack-ansible | 17:21 | |
| *** phiche has quit IRC | 17:22 | |
| *** yatin has quit IRC | 17:25 | |
| *** phiche has joined #openstack-ansible | 17:28 | |
| *** javeriak_ has quit IRC | 17:29 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Gate: Temporarily fix pep8 version to unblock gate checks https://review.openstack.org/267053 | 17:31 |
| *** michaelgugino has quit IRC | 17:31 | |
| openstackgerrit | Tom Cameron proposed openstack/openstack-ansible: set unique galera server-id https://review.openstack.org/267054 | 17:32 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267019 | 17:32 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267019 | 17:33 |
| *** phiche1 has joined #openstack-ansible | 17:34 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266832 | 17:34 |
| *** phiche has quit IRC | 17:35 | |
| *** galstrom is now known as galstrom_zzz | 17:35 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Revert "Gate: Temporarily fix pep8 version to unblock gate checks" https://review.openstack.org/267057 | 17:36 |
| *** adac has joined #openstack-ansible | 17:39 | |
| *** apuimedo has quit IRC | 17:41 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Re-deploy the Horizon venv if it mismatches the repo https://review.openstack.org/261293 | 17:42 |
| *** apuimedo has joined #openstack-ansible | 17:43 | |
| stevelle | odyssey4me: sorry, thought things were further unwound than they actually are, scrollback is catching me up | 17:46 |
| odyssey4me | stevelle yeah, I think we've managed to get it right - waiting for a positive gate result | 17:47 |
| spotz | Only half following the channel today, any of those coming through need reviewing? | 17:47 |
| stevelle | odyssey4me: I believe I have all the relevant reviews up now to monitor | 17:47 |
| *** apuimedo has quit IRC | 17:47 | |
| *** pai15 has quit IRC | 17:48 | |
| *** apuimedo has joined #openstack-ansible | 17:48 | |
| odyssey4me | spotz not to my knowledge | 17:49 |
| *** jthorne has joined #openstack-ansible | 17:51 | |
| *** javeriak has joined #openstack-ansible | 17:51 | |
| *** apuimedo has quit IRC | 17:53 | |
| *** electrofelix has quit IRC | 17:53 | |
| *** pai15 has joined #openstack-ansible | 17:54 | |
| *** apuimedo has joined #openstack-ansible | 17:54 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-security: Cap ansible < 2.0 for ansible-lint compatibility https://review.openstack.org/267067 | 17:57 |
| *** apuimedo has quit IRC | 18:01 | |
| *** KLevenstein has quit IRC | 18:01 | |
| *** apuimedo has joined #openstack-ansible | 18:02 | |
| *** KLevenstein has joined #openstack-ansible | 18:06 | |
| *** maximov has quit IRC | 18:08 | |
| *** meteorfox has quit IRC | 18:08 | |
| *** etoews has quit IRC | 18:08 | |
| *** serverascode has quit IRC | 18:08 | |
| *** rackertom has quit IRC | 18:08 | |
| *** apuimedo has quit IRC | 18:09 | |
| *** apuimedo has joined #openstack-ansible | 18:09 | |
| *** rackertom has joined #openstack-ansible | 18:10 | |
| *** yaya has quit IRC | 18:11 | |
| *** serverascode has joined #openstack-ansible | 18:16 | |
| *** admin0 has joined #openstack-ansible | 18:19 | |
| *** meteorfox has joined #openstack-ansible | 18:20 | |
| *** etoews has joined #openstack-ansible | 18:21 | |
| *** maximov has joined #openstack-ansible | 18:23 | |
| *** javeriak_ has joined #openstack-ansible | 18:29 | |
| *** ddaskal has quit IRC | 18:30 | |
| *** mgagne has quit IRC | 18:30 | |
| *** ddaskal has joined #openstack-ansible | 18:30 | |
| *** javeriak has quit IRC | 18:31 | |
| *** Guest82803 has joined #openstack-ansible | 18:31 | |
| *** Guest82803 is now known as mgagne | 18:32 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Set tox pep8 environment to use test-requirements.txt https://review.openstack.org/267083 | 18:32 |
| *** mgagne has quit IRC | 18:33 | |
| *** mgagne has joined #openstack-ansible | 18:33 | |
| *** apuimedo has quit IRC | 18:33 | |
| *** apuimedo has joined #openstack-ansible | 18:33 | |
| *** ddaskal has quit IRC | 18:34 | |
| *** apuimedo has quit IRC | 18:40 | |
| *** jaypipes has quit IRC | 18:41 | |
| *** apuimedo has joined #openstack-ansible | 18:42 | |
| *** ig0r_ has joined #openstack-ansible | 18:43 | |
| *** jaypipes has joined #openstack-ansible | 18:44 | |
| *** jthorne has quit IRC | 18:44 | |
| *** pai15 has quit IRC | 18:45 | |
| *** apuimedo has quit IRC | 18:49 | |
| *** javeriak has joined #openstack-ansible | 18:49 | |
| *** apuimedo has joined #openstack-ansible | 18:51 | |
| *** javeriak_ has quit IRC | 18:51 | |
| *** cemmason1 has joined #openstack-ansible | 18:54 | |
| *** cemmason has quit IRC | 18:57 | |
| *** mgoddard_ has joined #openstack-ansible | 19:01 | |
| *** mgoddard has quit IRC | 19:04 | |
| *** galstrom_zzz is now known as galstrom | 19:08 | |
| *** apuimedo has quit IRC | 19:10 | |
| *** apuimedo has joined #openstack-ansible | 19:11 | |
| openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Tidy up notification_driver implementation across projects https://review.openstack.org/266318 | 19:12 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Default values for enabling services in ceilometer https://review.openstack.org/267112 | 19:15 |
| *** apuimedo has quit IRC | 19:15 | |
| *** apuimedo has joined #openstack-ansible | 19:16 | |
| *** javeriak_ has joined #openstack-ansible | 19:19 | |
| *** scarlisle has joined #openstack-ansible | 19:19 | |
| *** javeriak has quit IRC | 19:20 | |
| *** yaya has joined #openstack-ansible | 19:23 | |
| *** ddaskal has joined #openstack-ansible | 19:24 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Default values for enabling services in ceilometer https://review.openstack.org/267112 | 19:25 |
| *** apuimedo has quit IRC | 19:25 | |
| *** apuimedo has joined #openstack-ansible | 19:27 | |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible: Remove unwanted veth pair creation from lxc_container_create role https://review.openstack.org/267126 | 19:30 |
| *** fawadkhaliq has quit IRC | 19:30 | |
| *** admin0 has quit IRC | 19:31 | |
| *** phiche1 has quit IRC | 19:32 | |
| *** michaelgugino has joined #openstack-ansible | 19:32 | |
| *** phiche has joined #openstack-ansible | 19:32 | |
| *** apuimedo has quit IRC | 19:34 | |
| *** CheKoLyN has quit IRC | 19:35 | |
| *** apuimedo has joined #openstack-ansible | 19:36 | |
| *** raddaoui has joined #openstack-ansible | 19:37 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Remove unwanted veth pair creation from lxc_container_create role https://review.openstack.org/267126 | 19:38 |
| mhayden | i might update my spec around SSL to use anchor | 19:41 |
| *** ssl_ has joined #openstack-ansible | 19:51 | |
| *** apuimedo has quit IRC | 19:55 | |
| *** mpavone has joined #openstack-ansible | 19:55 | |
| *** apuimedo has joined #openstack-ansible | 19:56 | |
| *** mpavone has quit IRC | 19:59 | |
| *** CheKoLyN has joined #openstack-ansible | 19:59 | |
| *** Bjoern has joined #openstack-ansible | 19:59 | |
| *** Mudpuppy has quit IRC | 20:00 | |
| *** Mudpuppy_ has joined #openstack-ansible | 20:00 | |
| *** Mudpuppy_ has quit IRC | 20:01 | |
| *** admin0 has joined #openstack-ansible | 20:02 | |
| *** Mudpuppy has joined #openstack-ansible | 20:02 | |
| *** mgoddard_ has quit IRC | 20:06 | |
| *** Bjoern has quit IRC | 20:06 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Added extra variable for upgrade https://review.openstack.org/267135 | 20:07 |
| *** apuimedo has quit IRC | 20:07 | |
| *** apuimedo has joined #openstack-ansible | 20:08 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: networking-plumgrid package version fix https://review.openstack.org/267136 | 20:09 |
| ssl_ | new to lxc and osad.. how do i restart the openstack service (e.g. nova scheduler) in AIO | 20:13 |
| palendae | ssl_: lxc-stop -n <container> && lxc-start -n <container> | 20:14 |
| palendae | Or accessing the container with SSH/lxc-attach and restarting the service like normal | 20:14 |
| *** apuimedo has quit IRC | 20:15 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Cap ansible < 2.0 for ansible-lint compatibility https://review.openstack.org/267067 | 20:15 |
| *** apuimedo has joined #openstack-ansible | 20:16 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Docs: overview cleanup https://review.openstack.org/265303 | 20:18 |
| cloudnull | ssl_: all of the openstack services had upstart init scripts | 20:25 |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: networking-plumgrid package version fix https://review.openstack.org/267146 | 20:25 |
| cloudnull | so if your on a compute node you can do `service nova-compute restart` | 20:25 |
| *** apuimedo has quit IRC | 20:26 | |
| *** apuimedo has joined #openstack-ansible | 20:27 | |
| cloudnull | javeriak_: this PR https://review.openstack.org/#/c/267136/ conflicts with https://review.openstack.org/#/c/267019/5 , can you make this a dependency ? | 20:27 |
| javeriak_ | cloudnull yep was just doing that | 20:27 |
| cloudnull | ha :) | 20:27 |
| * cloudnull goes back to my hole | 20:28 | |
| *** CheKoLyN has quit IRC | 20:30 | |
| odyssey4me | javeriak_ is it possible to set a range for each branch - eg for kilo it should be networking-plumgrid>=2015.1,<2016 ? | 20:31 |
| odyssey4me | or is it static for the entire lifespan of the kilo branch? | 20:31 |
| javeriak_ | odyssey4me it's going to be static versions that map between our platform releases : kilo/liberty etc | 20:32 |
| javeriak_ | odyssey4me im going to push the exact the versions through the plumgrid playbooks that run before neutron | 20:33 |
| odyssey4me | ok, in that case I've added a comment | 20:34 |
| odyssey4me | for the repo to have the right version, the package list should have the constraints | 20:34 |
| odyssey4me | if you want to override the constraints for an install, you simply override the package list in user_variables | 20:34 |
| odyssey4me | otherwise the install will fail because the repo doesn't have the right version | 20:35 |
| *** apuimedo has quit IRC | 20:36 | |
| javeriak_ | odyssey4me sure, how about i just do this networking-plumgrid=="{{networking_plumgrid_version}}" | 20:36 |
| javeriak_ | then i could change networking_plumgrid_version in my user vars | 20:36 |
| javeriak_ | let me test this though first | 20:36 |
| odyssey4me | javeriak_ the quotes will need to be on the outside, ie - "networking-plumgrid=={{networking_plumgrid_version}}" | 20:37 |
| javeriak_ | odyssey4me okay thanks | 20:37 |
| odyssey4me | that said, I'm not fond of creating yet another var | 20:37 |
| *** Mudpuppy has quit IRC | 20:38 | |
| *** apuimedo has joined #openstack-ansible | 20:38 | |
| javeriak_ | oh hmm... well then i could use the static name networking-plumgrid==2015.1.1.1 and then just override this in my user vars? | 20:38 |
| javeriak_ | odyssey4me does that look better? | 20:39 |
| *** Mudpuppy has joined #openstack-ansible | 20:40 | |
| openstackgerrit | Eric Brown proposed openstack/openstack-ansible-security: Bump minimum required version of Ansible https://review.openstack.org/267009 | 20:41 |
| *** Mudpuppy_ has joined #openstack-ansible | 20:41 | |
| *** Mudpuppy has quit IRC | 20:42 | |
| *** apuimedo has quit IRC | 20:42 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267019 | 20:43 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Add template option for the default horizon panel https://review.openstack.org/267155 | 20:43 |
| *** apuimedo has joined #openstack-ansible | 20:44 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266832 | 20:45 |
| *** ddaskal has quit IRC | 20:48 | |
| *** apuimedo has quit IRC | 20:48 | |
| *** ddaskal has joined #openstack-ansible | 20:49 | |
| *** CheKoLyN has joined #openstack-ansible | 20:49 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Revert the temp tox config change https://review.openstack.org/267163 | 20:49 |
| *** apuimedo has joined #openstack-ansible | 20:50 | |
| *** ddaskal has quit IRC | 20:53 | |
| *** apuimedo has quit IRC | 20:54 | |
| *** apuimedo has joined #openstack-ansible | 20:55 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267019 | 20:55 |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: networking-plumgrid package version fix https://review.openstack.org/267136 | 20:55 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267019 | 20:58 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: networking-plumgrid package version fix https://review.openstack.org/267136 | 20:59 |
| *** raddaoui has quit IRC | 21:00 | |
| *** apuimedo has quit IRC | 21:00 | |
| *** raddaoui has joined #openstack-ansible | 21:00 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/266832 | 21:01 |
| *** apuimedo has joined #openstack-ansible | 21:01 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Revert the temp tox config change https://review.openstack.org/267163 | 21:01 |
| *** adac has quit IRC | 21:02 | |
| *** apuimedo has quit IRC | 21:06 | |
| *** apuimedo has joined #openstack-ansible | 21:08 | |
| *** raddaoui has quit IRC | 21:11 | |
| *** ig0r_ has quit IRC | 21:14 | |
| *** apuimedo has quit IRC | 21:17 | |
| *** apuimedo has joined #openstack-ansible | 21:18 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: networking-plumgrid package version fix https://review.openstack.org/267146 | 21:21 |
| *** apuimedo has quit IRC | 21:22 | |
| *** apuimedo has joined #openstack-ansible | 21:23 | |
| *** ig0r_ has joined #openstack-ansible | 21:23 | |
| *** Mudpuppy_ is now known as Mudpuppy | 21:26 | |
| *** sdake has quit IRC | 21:26 | |
| jmccrory | has anyone see failures like this before? ERROR! failed to resolve remote temporary directory from ansible-tmp-1452720259.32-221209870114351 | 21:27 |
| jmccrory | intermittent on template tasks | 21:27 |
| *** apuimedo has quit IRC | 21:27 | |
| *** admin0 has quit IRC | 21:28 | |
| *** apuimedo has joined #openstack-ansible | 21:28 | |
| *** admin0 has joined #openstack-ansible | 21:28 | |
| *** raddaoui has joined #openstack-ansible | 21:31 | |
| *** ig0r_ has quit IRC | 21:31 | |
| *** javeriak_ has quit IRC | 21:32 | |
| mhayden | i have seen that happen from time to time | 21:34 |
| mhayden | i've usually specified -> remote_tmp = ../someplace | 21:34 |
| *** apuimedo has quit IRC | 21:40 | |
| *** apuimedo has joined #openstack-ansible | 21:42 | |
| *** jguy_ has quit IRC | 21:43 | |
| *** ig0r_ has joined #openstack-ansible | 21:44 | |
| *** ig0r_ has quit IRC | 21:45 | |
| *** apuimedo has quit IRC | 21:46 | |
| *** apuimedo has joined #openstack-ansible | 21:47 | |
| *** sdake has joined #openstack-ansible | 21:53 | |
| *** alextricity_h has quit IRC | 21:56 | |
| *** sdake_ has joined #openstack-ansible | 21:56 | |
| d34dh0r53 | can we get reviews on https://review.openstack.org/#/c/267135? | 21:56 |
| *** sdake has quit IRC | 21:57 | |
| *** ddaskal has joined #openstack-ansible | 21:58 | |
| *** raddaoui has quit IRC | 22:00 | |
| odyssey4me | d34dh0r53 does it work? | 22:06 |
| *** apuimedo has quit IRC | 22:06 | |
| *** apuimedo has joined #openstack-ansible | 22:07 | |
| *** Mudpuppy has quit IRC | 22:09 | |
| odyssey4me | andymccr FYI, these patches came in yesterday and merged into master today: https://review.openstack.org/264936 / https://review.openstack.org/266021 | 22:09 |
| odyssey4me | andymccr we then needed to do https://review.openstack.org/266530 as a follow up | 22:10 |
| odyssey4me | andymccr then finally, to get master working - https://review.openstack.org/267019 | 22:10 |
| odyssey4me | that last one's in the gate currently (the gate has a lng queue) | 22:10 |
| andymccr | odyssey4me: yeh i saw that PR I +2'd it but i figured it'd get changed anyway. i had just removed the br_netfilter bit to get it to work in the meantime | 22:10 |
| mhayden | that neutron one dinged me good | 22:10 |
| andymccr | i'll take a look at that last PR 267019 | 22:11 |
| *** apuimedo has quit IRC | 22:12 | |
| d34dh0r53 | odyssey4me: yeah | 22:13 |
| *** apuimedo has joined #openstack-ansible | 22:13 | |
| odyssey4me | d34dh0r53 I've added my vote +w, the gate is grumpy though | 22:14 |
| odyssey4me | hopefully we gat unwedge the gate in master today, then backport all the bits | 22:14 |
| *** galstrom is now known as galstrom_zzz | 22:14 | |
| d34dh0r53 | odyssey4me: thanks man! :) I'll keep an eye on the gate | 22:14 |
| odyssey4me | d34dh0r53 if you can review https://review.openstack.org/267019 and https://review.openstack.org/267136 to help that process, then that'll help | 22:15 |
| d34dh0r53 | yep, in a meeting but I'll look at them after | 22:15 |
| odyssey4me | I don't know if the first will need to go back all the way to kilo, but the second will | 22:15 |
| odyssey4me | urgh, the second review number was wrong - this is the one: https://review.openstack.org/266832 | 22:17 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/267200 | 22:18 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/267200 | 22:20 |
| *** apuimedo has quit IRC | 22:20 | |
| *** apuimedo has joined #openstack-ansible | 22:21 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/267200 | 22:23 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/267203 | 22:24 |
| *** apuimedo has quit IRC | 22:25 | |
| *** apuimedo has joined #openstack-ansible | 22:26 | |
| *** mattronix has quit IRC | 22:34 | |
| *** apuimedo has quit IRC | 22:35 | |
| *** apuimedo has joined #openstack-ansible | 22:37 | |
| *** jguy_ has joined #openstack-ansible | 22:40 | |
| *** mattronix has joined #openstack-ansible | 22:41 | |
| *** apuimedo has quit IRC | 22:41 | |
| *** apuimedo has joined #openstack-ansible | 22:43 | |
| *** apuimedo has quit IRC | 22:50 | |
| *** apuimedo has joined #openstack-ansible | 22:51 | |
| *** jguy_ has quit IRC | 22:52 | |
| stevelle | gogo 266832 | 22:53 |
| *** hybridpollo has joined #openstack-ansible | 22:54 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:55 | |
| *** yaya has left #openstack-ansible | 22:56 | |
| odyssey4me | stevelle yep, that and https://review.openstack.org/267019 are the ticket :) | 22:56 |
| *** jeh has quit IRC | 22:59 | |
| *** apuimedo has quit IRC | 23:03 | |
| *** apuimedo has joined #openstack-ansible | 23:04 | |
| *** spotz is now known as spotz_zzz | 23:04 | |
| odyssey4me | stevelle and we have success: https://review.openstack.org/266832 | 23:07 |
| odyssey4me | care to add your vote? | 23:07 |
| stevelle | 266832 through gate, just workflowed | 23:07 |
| stevelle | 267019 in smoketests | 23:07 |
| odyssey4me | stevelle done, my vote is in | 23:09 |
| stevelle | ditto | 23:10 |
| odyssey4me | thanks stevelle | 23:10 |
| stevelle | will let them merge and we can look at getting the backports through | 23:10 |
| *** CheKoLyN has quit IRC | 23:11 | |
| *** evanston_ has quit IRC | 23:14 | |
| *** michaelgugino has quit IRC | 23:16 | |
| *** phiche has quit IRC | 23:19 | |
| *** evanstoner has joined #openstack-ansible | 23:22 | |
| odyssey4me | stevelle can we also get this one through? https://review.openstack.org/267163 | 23:22 |
| odyssey4me | andymccr ^ | 23:22 |
| andymccr | looking | 23:22 |
| stevelle | odyssey4me: was watching it, ready to push the butan | 23:22 |
| andymccr | make it so. | 23:24 |
| *** baker has quit IRC | 23:29 | |
| odyssey4me | thanks andymccr | 23:29 |
| *** apuimedo has quit IRC | 23:30 | |
| *** ddaskal has quit IRC | 23:31 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible-galera_server: Merge bashate/pep8 lint jobs in common job https://review.openstack.org/257755 | 23:31 |
| *** apuimedo has joined #openstack-ansible | 23:31 | |
| *** ddaskal has joined #openstack-ansible | 23:32 | |
| *** baker has joined #openstack-ansible | 23:32 | |
| *** phalmos has quit IRC | 23:33 | |
| *** baker has quit IRC | 23:33 | |
| odyssey4me | if anyone has a gap to look into it, https://review.openstack.org/267200 is failing the ansible-lint test | 23:34 |
| *** spotz_zzz is now known as spotz | 23:34 | |
| *** ddaskal has quit IRC | 23:36 | |
| *** scarlisle has quit IRC | 23:37 | |
| *** baker has joined #openstack-ansible | 23:37 | |
| *** spotz is now known as spotz_zzz | 23:37 | |
| *** apuimedo has quit IRC | 23:38 | |
| *** KLevenstein has quit IRC | 23:39 | |
| *** apuimedo has joined #openstack-ansible | 23:39 | |
| jmccrory | odyssey4me looks like it installed ansible2 | 23:42 |
| jmccrory | which doesn't have ansible.util.parse_yaml | 23:43 |
| jmccrory | utils* | 23:43 |
| *** apuimedo has quit IRC | 23:44 | |
| *** apuimedo has joined #openstack-ansible | 23:45 | |
| odyssey4me | jmccrory and so it does seem, how the heck did it do that? | 23:49 |
| *** apuimedo has quit IRC | 23:49 | |
| *** apuimedo has joined #openstack-ansible | 23:50 | |
| *** baker has quit IRC | 23:52 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix neutron migrations with optional providers https://review.openstack.org/267233 | 23:53 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Resolve pep8 violations caught by pep8 v1.7.0 https://review.openstack.org/267203 | 23:56 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Revert the temp tox config change https://review.openstack.org/267234 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!