*** sapd1_x has quit IRC | 00:24 | |
*** ricolin has joined #openstack-self-healing | 00:43 | |
*** dasp has joined #openstack-self-healing | 02:11 | |
*** dasp has quit IRC | 04:08 | |
*** dasp has joined #openstack-self-healing | 04:30 | |
*** witek has joined #openstack-self-healing | 06:45 | |
*** bogdando has joined #openstack-self-healing | 07:45 | |
*** ricolin has quit IRC | 08:24 | |
*** sapd1_x has joined #openstack-self-healing | 08:29 | |
*** bogdando has quit IRC | 08:41 | |
*** sapd1_x has quit IRC | 08:56 | |
*** bogdando has joined #openstack-self-healing | 09:15 | |
*** sapd1_x has joined #openstack-self-healing | 09:19 | |
*** tojuvone has quit IRC | 09:46 | |
*** tojuvone has joined #openstack-self-healing | 09:46 | |
*** sapd1_x has quit IRC | 11:09 | |
*** sapd1_x has joined #openstack-self-healing | 13:27 | |
*** sapd1_x has quit IRC | 13:56 | |
*** witek has quit IRC | 15:56 | |
*** bogdando has quit IRC | 16:20 | |
*** ekcs has joined #openstack-self-healing | 17:53 | |
joadavis | The only (small) downside I've seen to alert based is that you split the logic - some bit goes in the monitoring to decide if it is an alarm, and some bit goes in the root cause or clustering tool to decide what to do about the alarm. Its not really a problem, but you do need to be aware of both parts. | 20:55 |
---|---|---|
aspiers | joadavis: that's a good point! can you ensure it's mentioned in the relevant spec(s)? | 20:58 |
joadavis | I can take a look for watcher and vitrage - anywhere else? | 21:08 |
*** joadavis has quit IRC | 21:47 | |
*** joadavis has joined #openstack-self-healing | 22:05 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!