*** amalagon has quit IRC | 01:05 | |
*** eglynn-regus has quit IRC | 04:11 | |
*** eglynn-regus has joined #openstack-monasca | 04:12 | |
*** stackedsax has quit IRC | 05:09 | |
*** stackedsax has joined #openstack-monasca | 05:12 | |
*** sanjayu has joined #openstack-monasca | 05:53 | |
*** krtaylor has joined #openstack-monasca | 05:57 | |
*** eglynn-regus has quit IRC | 06:20 | |
*** eglynn-regus has joined #openstack-monasca | 06:52 | |
*** dorry has quit IRC | 08:23 | |
*** sanjayu has quit IRC | 08:44 | |
*** sanjayu has joined #openstack-monasca | 08:46 | |
*** sanjayu has quit IRC | 10:36 | |
*** sanjayu has joined #openstack-monasca | 10:42 | |
*** ddieterl_ has joined #openstack-monasca | 12:14 | |
*** ddieterly has quit IRC | 12:15 | |
*** MVenesio has joined #openstack-monasca | 12:38 | |
*** sanjayu has quit IRC | 14:32 | |
*** dschroeder has joined #openstack-monasca | 16:07 | |
*** openstack has joined #openstack-monasca | 17:01 | |
*** amalagon has joined #openstack-monasca | 17:13 | |
*** garyh has joined #openstack-monasca | 17:40 | |
*** ddieterly has joined #openstack-monasca | 18:02 | |
*** ddieterl_ has quit IRC | 18:05 | |
*** MVenesio has quit IRC | 18:17 | |
*** MVenesio has joined #openstack-monasca | 18:19 | |
*** MVenesio has quit IRC | 18:33 | |
*** shaneduan has joined #openstack-monasca | 19:13 | |
*** ddieterly has quit IRC | 21:32 | |
*** ddieterly has joined #openstack-monasca | 21:34 | |
*** LeanReox has joined #openstack-monasca | 21:48 | |
LeanReox | hi guys, anyone has the latest thresh-config.yml to use influxdb as backend and not mysql ? | 21:49 |
---|---|---|
LeanReox | sandywalsh: how you doing Sandy, just one question, looking for configuring influxdb as monasca-thresh backend, is this possible ? or we have to have a mysql for thresholding purposes? | 22:37 |
openstackgerrit | Tim Kuhlman proposed stackforge/monasca-notification: Fixes for statsd integration https://review.openstack.org/143555 | 22:49 |
sandywalsh | LeanReox: good question ... haven't messed with the thres engine. I'm not sure if influx support has been added there yet. Lemme look at the repo | 22:49 |
sandywalsh | LeanReox: my guess is that the thres engine can use any backend for storing the alarm definitions. It gets the metrics from the kafka queues and sends them into Storm directly (so no influx requirement) | 22:52 |
sandywalsh | but I could be wrong about that. | 22:52 |
sandywalsh | there is no mention of influx in the thres repo, so I think it's just using whatever Java-based db is supported. | 22:53 |
LeanReox | sandywalsh: it seems that its trying to do something with mysql | 22:53 |
LeanReox | sandywalsh: its configured on the thresh-config.yaml | 22:53 |
sandywalsh | LeanReox: just the alarm definitions I suspect | 22:53 |
sandywalsh | or just the default DAO / persistance definition | 22:53 |
LeanReox | sandywalsh: i bet for the second, because i can create alarms definitions without any problem and the monasca-thresh is not running | 22:54 |
sandywalsh | LeanReox: so, you mean the alarm definitions are getting stored via the API layer? | 22:55 |
sandywalsh | and not in the thres service itself? | 22:55 |
sandywalsh | (could very well be) | 22:55 |
LeanReox | sandywalsh: yes, cause the thresh is giving an stacktrace trying to initiaty the hybernate layer to mysql, and doesnt even start, so yes | 22:56 |
sandywalsh | makes sense (though I don't like cluttered api layers :) | 22:56 |
LeanReox | sandywalsh: haha, ill try pointing the mysql to the nova one, and see how it goes | 22:57 |
LeanReox | sandywalsh: and what actually stores there | 22:57 |
sandywalsh | good test | 22:58 |
LeanReox | should be great that we can store everything to influx or vertica, avoiding mixed db layers | 22:58 |
sandywalsh | definitely, though influx is kinda special-purpose. | 23:00 |
sandywalsh | like Chris Rock says "You can drive a car with your feet ... doesn't make it a good idea." | 23:00 |
LeanReox | sandywalsh: got a point there :) | 23:02 |
sandywalsh | not a biggie though ... monasca is pretty smart about alarm definitions. Not heavy weight. | 23:03 |
*** ddieterly has quit IRC | 23:10 | |
*** ddieterly has joined #openstack-monasca | 23:11 | |
openstackgerrit | Tim Kuhlman proposed stackforge/monasca-notification: Fixes for statsd integration https://review.openstack.org/143555 | 23:12 |
*** ddieterly has quit IRC | 23:16 | |
LeanReox | sandywalsh: now its working, and i figured out why wasnt working at first ... if you configure a numeric pass for mysql on the notification.yaml ... it gets casted as integer, should be casted as string by default | 23:17 |
sandywalsh | ah, good to know. You should file an issue for that one. | 23:18 |
LeanReox | sandywalsh: should be casted here self.mysql = MySQLdb.connect(host=mysql_host, user=mysql_user, passwd=mysql_passwd, db=dbname, ssl=mysql_ssl) | 23:18 |
LeanReox | at alarm_processor.py | 23:18 |
LeanReox | ill issue a bug on this one | 23:18 |
sandywalsh | cool | 23:19 |
*** LeanReox has quit IRC | 23:23 | |
openstackgerrit | Tim Kuhlman proposed stackforge/monasca-agent: Various fixes and code cleanup related to monasca statsd https://review.openstack.org/143561 | 23:26 |
*** ddieterly has joined #openstack-monasca | 23:36 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!