*** ysandeep|out is now known as ysandeep | 05:08 | |
*** ysandeep is now known as ysandeep|afk | 06:34 | |
*** ysandeep|afk is now known as ysandeep|trng | 07:03 | |
pas-ha[m] | gregwork_: what exactly is running slow? heat is mainly making the same REST api calls to CInder as any other client would do, and then a lot of times it will also poll for resource it CRUD'ed reach some status. So if something feels slow here, I would first check Cinder itself, e.g. tried to create the same sized volume on the same backend as your template does and see how long it takes... | 07:52 |
---|---|---|
gregwork_ | well the issue is im standing up 50 instances with this stack so its more about the speed under pressure | 07:54 |
gregwork_ | ive been playing with cinder settings to adjust and so far it hasnt come back but it did happen a few times so I figured i would ask | 07:55 |
*** ysandeep|trng is now known as ysandeep|afk | 10:19 | |
pas-ha[m] | I don't think there's anything meaningful to tune in Heat in this regard. If your cinder/its backend can not properly handle that massive parallel volume create operations, and you get errors creating volumes, at most you could try organizing your stack template to create them serially or in smaller batches - but that of course will be "slower" too. | 11:14 |
*** ysandeep|afk is now known as ysandeep | 12:21 | |
*** ysandeep is now known as ysandeep|away | 16:41 | |
opendevreview | Federico Ressi proposed openstack/heat master: Use the new devstack-tobiko-heat jobs to gate heat changes https://review.opendev.org/c/openstack/heat/+/814424 | 20:00 |
opendevreview | Merged openstack/heat master: Allow for creating server_group with max_nova_api_microversion https://review.opendev.org/c/openstack/heat/+/814388 | 21:21 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!