| openstackgerrit | Mike Fedosin proposed openstack/mistral master: Add Glare action pack https://review.openstack.org/473898 | 00:34 |
|---|---|---|
| *** zhurong has joined #openstack-mistral | 00:39 | |
| openstackgerrit | Mike Fedosin proposed openstack/mistral master: Use recommended function to setup auth middleware in devstack https://review.openstack.org/473796 | 01:07 |
| *** liujiong has joined #openstack-mistral | 01:59 | |
| liujiong | Hi, mistral team, I got trouble with mistral dabase, can you help me? | 03:00 |
| liujiong | 2017-06-15 10:57:37.745 181972 INFO mistral.api.controllers.v2.workflow [-] Fetch workflows. marker=None, limit=None, sort_keys=[u'created_at'], sort_dirs=[u'asc'], fields=[], filters={} | 03:00 |
| liujiong | 2017-06-15 10:57:37.786 181972 DEBUG oslo_db.sqlalchemy.engines [-] Parent process 181924 forked (181972) with an open database connection, which is being discarded and recreated. checkout | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool [-] Exception during reset or similar | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool Traceback (most recent call last): | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool File "/usr/lib64/python2.7/site-packages/sqlalchemy/pool.py", line 636, in _finalize_fairy | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool fairy._reset(pool) | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool File "/usr/lib64/python2.7/site-packages/sqlalchemy/pool.py", line 776, in _reset | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool pool._dialect.do_rollback(self) | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool File "/usr/lib64/python2.7/site-packages/sqlalchemy/dialects/mysql/base.py", line 2526, in do_rollback | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool dbapi_connection.rollback() | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool OperationalError: (2006, 'MySQL server has gone away') | 03:00 |
| liujiong | 2017-06-15 10:57:37.791 181972 ERROR sqlalchemy.pool.QueuePool | 03:00 |
| *** catintheroof has joined #openstack-mistral | 03:15 | |
| *** fultonj has quit IRC | 03:28 | |
| *** fultonj has joined #openstack-mistral | 03:40 | |
| *** catintheroof has quit IRC | 03:47 | |
| *** fultonj has quit IRC | 04:04 | |
| *** fultonj has joined #openstack-mistral | 04:06 | |
| *** gkadam has joined #openstack-mistral | 04:21 | |
| *** zhurong has quit IRC | 04:28 | |
| *** zhurong has joined #openstack-mistral | 04:40 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral master: Make sure that the field "state_info" trimmed as expected https://review.openstack.org/474195 | 04:40 |
| *** ttpl has quit IRC | 04:41 | |
| rakhmerov | liujiong: hi, please paste all logs and stack traces to paste.openstack.org | 04:45 |
| rakhmerov | can you show your mistral config? | 04:45 |
| *** ttpl has joined #openstack-mistral | 04:48 | |
| *** ttpl has quit IRC | 04:52 | |
| openstackgerrit | Shubhendu Poothia proposed openstack/mistral master: Set access_policy for messaging's dispatcher https://review.openstack.org/474451 | 04:53 |
| *** ttpl has joined #openstack-mistral | 04:54 | |
| *** jaosorior has quit IRC | 05:00 | |
| *** ttpl has quit IRC | 05:22 | |
| openstackgerrit | Sharat Sharma proposed openstack/python-mistralclient master: Add pagination options for workflows and actions https://review.openstack.org/458515 | 05:41 |
| *** sharatss has joined #openstack-mistral | 05:41 | |
| *** dtantsur|afk is now known as dtantsur | 05:49 | |
| openstackgerrit | Sharat Sharma proposed openstack/python-mistralclient master: Add pagination options for workflows and actions https://review.openstack.org/458515 | 06:54 |
| *** liujiong_lj has joined #openstack-mistral | 06:57 | |
| *** liujiong has quit IRC | 06:58 | |
| liujiong_lj | rakhmerov: thank you! I figured it out - the logs I posted wasn't the root cause of the problem, I added new actions to mapping.json but forgot to populate database. | 06:59 |
| rakhmerov | ok ) | 06:59 |
| liujiong_lj | rakhmerov: but seems mistral will have database issue like posted when it's set with HA. | 07:00 |
| rakhmerov | why? | 07:01 |
| liujiong_lj | I havn't digged into that but got such problem when we deploy mistral with HA. | 07:02 |
| liujiong_lj | The same issue also exists in Barbican project, it should depend on the way we use oslo.db | 07:03 |
| *** jaosorior has joined #openstack-mistral | 07:12 | |
| *** amoralej|off is now known as amoralej | 07:15 | |
| *** liujiong_lj has quit IRC | 07:22 | |
| *** liujiong has joined #openstack-mistral | 07:22 | |
| openstackgerrit | Sharat Sharma proposed openstack/python-mistralclient master: Add pagination options for workflows and actions https://review.openstack.org/458515 | 07:36 |
| *** shardy has joined #openstack-mistral | 07:36 | |
| *** gkadam is now known as gkadam-brb | 07:48 | |
| *** jpich has joined #openstack-mistral | 08:11 | |
| *** shardy has quit IRC | 08:31 | |
| *** dtantsur is now known as dtantsur|lunch | 08:33 | |
| *** liujiong has quit IRC | 08:43 | |
| *** liujiong has joined #openstack-mistral | 08:43 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral master: Make sure that the field "state_info" trimmed as expected https://review.openstack.org/474195 | 09:15 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral master: Make sure that the field "state_info" trimmed as expected https://review.openstack.org/474195 | 09:28 |
| *** sharatss has quit IRC | 09:36 | |
| *** d0ugal has quit IRC | 09:48 | |
| *** d0ugal has joined #openstack-mistral | 09:49 | |
| *** sharatss has joined #openstack-mistral | 09:54 | |
| *** dtantsur|lunch is now known as dtantsur | 09:56 | |
| openstackgerrit | Merged openstack/mistral master: Set access_policy for messaging's dispatcher https://review.openstack.org/474451 | 10:00 |
| therve | sharatss, Are you looking at https://review.openstack.org/#/c/455174/ failures? | 10:03 |
| therve | It was broken by the recent changes you made | 10:03 |
| sharatss | hi therve, i was making those changes given in review | 10:03 |
| therve | sharatss, Well yeah review are wrong sometimes :) | 10:04 |
| sharatss | therve, if u have got some time pls take a look at it | 10:04 |
| therve | test failures are not :D | 10:04 |
| therve | Yeah we can't rename api_key, it's a compatibility parameter | 10:04 |
| therve | The kwargs.pop('session') is not doing the same thing as before | 10:05 |
| sharatss | therve, yea.. i reverted the api_key though... | 10:05 |
| therve | sharatss, Not the tests | 10:05 |
| sharatss | therve, i am deeply involved in the cricket match :P | 10:06 |
| openstackgerrit | Thomas Herve proposed openstack/python-mistralclient master: Use keystoneauth plugins and session instead of keystoneclient https://review.openstack.org/455174 | 10:06 |
| therve | Good for you :D | 10:06 |
| sharatss | therve, you watching? | 10:06 |
| *** liujiong has quit IRC | 10:06 | |
| therve | Hum no :) | 10:06 |
| *** jkilpatr has quit IRC | 10:32 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral master: Make sure that the field "state_info" trimmed as expected https://review.openstack.org/474195 | 10:45 |
| *** jkilpatr has joined #openstack-mistral | 10:49 | |
| *** shardy has joined #openstack-mistral | 10:50 | |
| *** gkadam-brb is now known as gkadam | 11:00 | |
| *** amoralej is now known as amoralej|lunch | 11:02 | |
| rakhmerov | sharatss, therve: yeah, that was rather a question :) | 11:09 |
| rakhmerov | I mean renaming api_key | 11:09 |
| rakhmerov | I didn't know for sure | 11:09 |
| sharatss | rakhmerov, i tried to rename but failed badly :D | 11:09 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral master: Make sure that the field "state_info" trimmed as expected https://review.openstack.org/474195 | 11:29 |
| *** chlong has joined #openstack-mistral | 11:32 | |
| *** amoralej|lunch is now known as amoralej | 12:17 | |
| *** zhurong has quit IRC | 12:26 | |
| *** zhurong has joined #openstack-mistral | 12:28 | |
| *** zhurong has quit IRC | 12:36 | |
| fultonj | https://review.openstack.org/#/c/469644/12/workbooks/ceph-ansible.yaml@64 | 12:36 |
| fultonj | ^ does anyone have advice on the two issues pointed out above? ^ | 12:37 |
| fultonj | d0ugal rbrady ? | 12:37 |
| fultonj | has anyone built a datastructure using yaql in one mistral task and passed it to another mistral task as input ? | 12:40 |
| fultonj | e.g http://paste.openstack.org/show/612691 | 12:40 |
| rakhmerov | fultonj: not sure I understand your last question | 12:42 |
| rakhmerov | is there any specific problem with it? | 12:42 |
| rakhmerov | you can make a datastructure and just use "publish" to pass it further to other tasks downstream | 12:42 |
| rakhmerov | and it will be accessible using $.my_datastructure | 12:42 |
| fultonj | rakhmerov: yes, publish has worked well for me for simpler variables. assuming it works well for a dict like (http://paste.openstack.org/show/612691) then I am thinking of setting it equal to a var | 12:44 |
| *** sharatss has quit IRC | 12:45 | |
| fultonj | inventory: <% $.var %> | 12:45 |
| therve | fultonj, I don't get why you don't reuse stuff publish in set_ip_lists | 12:45 |
| rakhmerov | yeah, it should work w/o problems | 12:45 |
| fultonj | ^ i would then pass the above | 12:45 |
| fultonj | however, the action tripleo.ansbile-playbook isn't working nicely with it | 12:46 |
| rakhmerov | there isn't a difference how primitive values and data structures are processed | 12:46 |
| fultonj | therve: yes, i was planning to that originally | 12:46 |
| fultonj | ultimately i think this is a formatting issue | 12:46 |
| fultonj | so do i solve it by modifying the action to deal with the json i get from the yaql OR should i be smarter about generating the datastructure? | 12:47 |
| fultonj | rakhmerov: ok, thanks for the confirmation | 12:47 |
| fultonj | the action is here https://review.openstack.org/#/c/470021/13/tripleo_common/actions/ansible.py | 12:48 |
| fultonj | my attempts to build the structure in yaql and pass it gets me errors like this | 12:49 |
| fultonj | http://paste.openstack.org/raw/612697/ | 12:49 |
| fultonj | the action expects yaml | 12:50 |
| fultonj | i can have it convert json to yaml within the action and make json a new option | 12:50 |
| fultonj | or am i overlooking the power of yaql | 12:50 |
| fultonj | ? | 12:50 |
| fultonj | wondering the direction i ought to go in | 12:50 |
| therve | fultonj, So what happens if you do eg hosts: <% mon_ips %> | 12:53 |
| fultonj | therve: i tried that and it copmlained. i will reproduce right now and share the error. | 12:53 |
| fultonj | hosts: <%$.mon_ips %> | 12:55 |
| * fultonj running | 12:55 | |
| fultonj | hosts: <% $.mon_ips %> | 12:55 |
| therve | Yeah sorry | 12:55 |
| fultonj | np | 12:55 |
| fultonj | i understand what you meant | 12:55 |
| d0ugal | FL1SK: looking. | 12:56 |
| d0ugal | fultonj: looking | 12:57 |
| d0ugal | FL1SK: sorry. | 12:57 |
| fultonj | http://paste.openstack.org/raw/612701/ | 12:57 |
| fultonj | therve: ^ | 12:57 |
| fultonj | d0ugal: thanks | 12:57 |
| fultonj | therve: i think it's not getting a good yaml list from "hosts: <% $.mon_ips %>" | 12:57 |
| fultonj | good => one that parses correctly :) | 12:58 |
| therve | fultonj, Isn't that u'mons': {u'hosts': [u'192.168.24.8']} ? | 12:58 |
| fultonj | should be | 12:58 |
| * fultonj adds task to echo the var | 12:58 | |
| d0ugal | so the issue is that it is converted into a string? | 12:58 |
| therve | I mean I wonder if the error isn't on another hosts | 12:59 |
| therve | Like when it's empty, I guess it wants [] not '' | 12:59 |
| therve | But you didn't change all hosts, did you? | 12:59 |
| fultonj | therve: just mons | 12:59 |
| fultonj | <% env().get('service_ips', {}).get('ceph_mon_ctlplane_node_ips', []) %> | 12:59 |
| fultonj | ^ it6 should be getting [] if nothing, though I know there's an IP | 13:00 |
| therve | fultonj, But did you do it for mdss? | 13:01 |
| fultonj | therve: mdss is empty, but at least ansible runs and things parse | 13:01 |
| fultonj | therve: to focus on the concatenation problem i temporarily yank out others | 13:02 |
| fultonj | so i just have osd and mon | 13:02 |
| fultonj | http://paste.openstack.org/raw/612703/ | 13:03 |
| fultonj | so... | 13:03 |
| fultonj | <% $.mon_ips %> | 13:03 |
| fultonj | is | 13:04 |
| fultonj | [ | 13:04 |
| fultonj | "192.168.24.8" | 13:04 |
| fultonj | ] | 13:04 |
| * fultonj does same to osds | 13:04 | |
| *** catintheroof has joined #openstack-mistral | 13:07 | |
| therve | fultonj, So I don't know what format ansible wants for hosts in that structure | 13:08 |
| fultonj | https://docs.ansible.com/ansible/intro_inventory.html | 13:09 |
| therve | But I'm sure we can come with it once we know :) | 13:09 |
| fultonj | therve: exactly | 13:09 |
| fultonj | i am thinking of adding a json option | 13:09 |
| fultonj | " Valid file extensions include ‘.yml’, ‘.yaml’, ‘.json" | 13:09 |
| fultonj | to the ansible action | 13:09 |
| fultonj | and building the whole thing w/ yaql http://paste.openstack.org/show/612691/ | 13:10 |
| fultonj | but didn't want to go down the road if someone thought i should just solve this 100% in the workflow | 13:10 |
| fultonj | as per line 205 | 13:12 |
| therve | fultonj, So it would seem it wants "hosts": {"192.168.24.8": ""}, no? | 13:12 |
| fultonj | https://review.openstack.org/#/c/470021/13/tripleo_common/actions/ansible.py | 13:12 |
| fultonj | paste the structure and chose the right syntax | 13:12 |
| fultonj | s/syntax/suffix | 13:12 |
| fultonj | therve: why would it want that? | 13:13 |
| fultonj | therve: i think you're right | 13:13 |
| therve | fultonj, https://github.com/ansible/ansible/blob/devel/examples/hosts.yaml | 13:13 |
| fultonj | given https://review.openstack.org/#/c/471759/1/coe/kubernetes-master.yaml | 13:13 |
| therve | That's what it looks like | 13:13 |
| fultonj | but i don't undersatnd | 13:13 |
| therve | So yeah : {} actually | 13:14 |
| *** rbrady has quit IRC | 13:15 | |
| fultonj | therve: so use yaql to build something that looks like "hosts": {"192.168.24.8": {}}, | 13:15 |
| *** rbrady has joined #openstack-mistral | 13:15 | |
| therve | fultonj, Yeah | 13:15 |
| fultonj | ok, i'll pursue that and no action modification and see how it goes | 13:16 |
| fultonj | thanks | 13:16 |
| therve | fultonj, $.mon_ips.toDict($, {}) | 13:16 |
| * d0ugal is following along | 13:16 | |
| *** rbrady has quit IRC | 13:17 | |
| * fultonj runs workbook with new yaql ^ | 13:17 | |
| *** rbrady has joined #openstack-mistral | 13:17 | |
| fultonj | therve: wins! | 13:18 |
| therve | \o/ | 13:18 |
| fultonj | http://paste.openstack.org/raw/612705/ | 13:18 |
| fultonj | problem 1 solved | 13:19 |
| * fultonj moves to problem 2 (when there are no mds) | 13:19 | |
| therve | I trust that it's the write output :) | 13:19 |
| therve | s/write/right | 13:19 |
| therve | Well I think it should work too if it's empty | 13:19 |
| *** dprince has joined #openstack-mistral | 13:20 | |
| *** catinthe_ has joined #openstack-mistral | 13:21 | |
| *** catintheroof has quit IRC | 13:24 | |
| *** catintheroof has joined #openstack-mistral | 13:25 | |
| *** catinthe_ has quit IRC | 13:25 | |
| *** catinthe_ has joined #openstack-mistral | 13:26 | |
| *** catinth__ has joined #openstack-mistral | 13:26 | |
| *** catintheroof has quit IRC | 13:26 | |
| fultonj | therve: close | 13:28 |
| fultonj | therve: i think i need a diff yaql expression | 13:28 |
| therve | fultonj, What do you have? | 13:28 |
| fultonj | to manage the case where the list is empty | 13:28 |
| fultonj | error=No method "toDict" for | 13:29 |
| fultonj | receiver mds_ips matches supplied arguments, data={}], | 13:29 |
| *** catintheroof has joined #openstack-mistral | 13:29 | |
| *** catinthe_ has quit IRC | 13:30 | |
| therve | Hum | 13:30 |
| therve | fultonj, Can you paste the template? | 13:30 |
| fultonj | therve: yes | 13:30 |
| fultonj | http://paste.openstack.org/raw/612707/ | 13:31 |
| therve | fultonj, You're missing $. no? | 13:32 |
| fultonj | therve: yeah sorry about that | 13:32 |
| * fultonj re-runs | 13:33 | |
| *** catinth__ has quit IRC | 13:33 | |
| therve | That's ok :) | 13:33 |
| therve | Actually that makes think about a task, which is to unit test yaql in tht | 13:33 |
| therve | Wonder how hard it would be | 13:33 |
| fultonj | therve: wins again! | 13:34 |
| fultonj | http://paste.openstack.org/raw/612708/ | 13:35 |
| fultonj | therve: thanks you just solved my problems and got me unstuck | 13:35 |
| therve | fultonj, You're very welcome! | 13:35 |
| fultonj | with less code, not more code too | 13:35 |
| therve | Thanks for pioneering those interfaces! They are not super dry yet | 13:36 |
| *** bobh has joined #openstack-mistral | 13:43 | |
| *** bobh has quit IRC | 13:46 | |
| *** bobh has joined #openstack-mistral | 13:49 | |
| *** zhurong has joined #openstack-mistral | 13:51 | |
| *** jistr is now known as jistr|mtg | 13:59 | |
| *** flaper87 has joined #openstack-mistral | 14:02 | |
| flaper87 | fultonj: did you see my patches? | 14:02 |
| fultonj | flaper87: yes (473587 + 473586) | 14:03 |
| fultonj | flaper87: i will try them out in my env and comment | 14:04 |
| fultonj | flaper87: and thank you | 14:04 |
| flaper87 | fultonj: my pleasure! :) | 14:04 |
| flaper87 | thank you for testing them | 14:04 |
| *** zhurong has quit IRC | 14:22 | |
| openstackgerrit | Mike Fedosin proposed openstack/mistral master: Use recommended function to setup auth middleware in devstack https://review.openstack.org/473796 | 14:51 |
| *** jistr|mtg is now known as jistr | 14:59 | |
| *** dtantsur is now known as dtantsur|bbl | 15:02 | |
| *** gkadam has quit IRC | 15:04 | |
| *** jkilpatr_ has joined #openstack-mistral | 15:40 | |
| *** jkilpatr has quit IRC | 15:43 | |
| *** dtantsur|bbl is now known as dtantsur | 15:49 | |
| *** jpich has quit IRC | 16:06 | |
| *** rbrady is now known as rbrady-afk | 16:08 | |
| *** jkilpatr has joined #openstack-mistral | 16:36 | |
| *** jkilpatr_ has quit IRC | 16:37 | |
| *** dtantsur is now known as dtantsur|afk | 16:38 | |
| *** bobh has quit IRC | 16:45 | |
| *** jtomasek has quit IRC | 17:13 | |
| *** bobh has joined #openstack-mistral | 17:14 | |
| *** shardy has quit IRC | 17:47 | |
| *** amoralej is now known as amoralej|off | 18:16 | |
| *** catinthe_ has joined #openstack-mistral | 19:07 | |
| *** jaosorior is now known as jaosorior_away | 19:08 | |
| *** catintheroof has quit IRC | 19:11 | |
| *** catintheroof has joined #openstack-mistral | 19:22 | |
| *** catinthe_ has quit IRC | 19:25 | |
| *** dprince has quit IRC | 19:50 | |
| *** jkilpatr has quit IRC | 20:18 | |
| openstackgerrit | Merged openstack/mistral master: Increase the Environment variable column length https://review.openstack.org/473813 | 20:21 |
| fultonj | flaper87: i found two issues so far and i'm looking at a 3rd | 20:27 |
| fultonj | https://review.openstack.org/#/c/473586/2/tripleo_common/actions/ansible.py | 20:27 |
| fultonj | ^ the two | 20:27 |
| fultonj | http://paste.openstack.org/raw/612749/ | 20:32 |
| *** jkilpatr has joined #openstack-mistral | 20:59 | |
| *** catinthe_ has joined #openstack-mistral | 21:06 | |
| *** catintheroof has quit IRC | 21:09 | |
| *** catinthe_ has quit IRC | 21:28 | |
| *** bobh has quit IRC | 22:05 | |
| *** jkilpatr has quit IRC | 22:24 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/mistral master: Updated from global requirements https://review.openstack.org/474655 | 22:43 |
| *** jkilpatr has joined #openstack-mistral | 23:43 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!