tristanC | pabelanger: is there an option to tune the upload frequency? Otherwise we are going to upload image daily to vexxhost... isn't there a cost for each image creation? | 02:09 |
---|---|---|
pabelanger | tristanC: right, no tune setting | 02:09 |
pabelanger | we'll publish new images each day | 02:09 |
pabelanger | but only keep 2 online at once | 02:10 |
*** atarakt has joined #softwarefactory | 02:10 | |
tristanC | pabelanger: what's the point of keep 2 online? is there a doc to rollback or something? | 02:10 |
pabelanger | tristanC: yah, rollback to last known good image | 02:11 |
pabelanger | think it is in nodepool docs | 02:11 |
pabelanger | I've often wanted it to be 3 images online upstream | 02:11 |
pabelanger | we'd have somebody delete the wrong good iamge before | 02:12 |
pabelanger | then you are down until new image is rebuilt, upstream that used to be 2 hours | 02:12 |
pabelanger | but about 30mins now | 02:12 |
tristanC | pabelanger: can't find such rollback documentation. what do you need to do to rollback? | 02:13 |
pabelanger | tristanC: there is some docs around pause | 02:14 |
pabelanger | for nodepool | 02:14 |
pabelanger | then you'd run dib-image-delete | 02:14 |
pabelanger | pause will stop an upload / build, if need to work on broken image | 02:14 |
tristanC | oh ok... didn't knew that :-) | 02:15 |
pabelanger | https://docs.openstack.org/infra/system-config/nodepool.html#bad-images | 02:15 |
pabelanger | guess we need to move that into nodepool | 02:15 |
pabelanger | yah, it has saved us upstream a bunch of times | 02:15 |
tristanC | pabelanger: anyway, i think we should check if there is a cost to create glance image daily | 02:15 |
*** atarakt has left #softwarefactory | 02:16 | |
pabelanger | tristanC: there is, but it is cheap: https://vexxhost.com/public-cloud/object-storage/ | 02:16 |
pabelanger | $0.10/GB/mo for 1TB | 02:17 |
tristanC | pabelanger: it seems like image creation is free, only space used is charged. | 04:35 |
*** nchakrab has joined #softwarefactory | 05:27 | |
sfbender | Tristan de Cacqueray created software-factory/os-loganalyze-distgit master: Bump version and add user feedback patch https://softwarefactory-project.io/r/13158 | 05:41 |
sfbender | Merged logreduce master: common: small fixes for automated process https://softwarefactory-project.io/r/12943 | 05:45 |
*** nchakrab has quit IRC | 06:22 | |
*** nchakrab has joined #softwarefactory | 06:31 | |
*** nchakrab has quit IRC | 07:29 | |
*** nchakrab has joined #softwarefactory | 07:36 | |
*** nchakrab_ has joined #softwarefactory | 08:29 | |
*** nchakrab has quit IRC | 08:31 | |
*** sfbender has quit IRC | 09:02 | |
*** nchakrab_ has quit IRC | 10:42 | |
*** apevec has joined #softwarefactory | 11:31 | |
*** apevec has joined #softwarefactory | 11:31 | |
apevec | sbadia, can you make mhu or fbo ops please? | 11:32 |
apevec | also https://www.redhat.com/mailman/listinfo/softwarefactory-dev needs updates: Softwarefactory-dev list run by cschwede at redhat.com, fboucher at redhat.com, sbadia at redhat.com, tdecacqu at redhat.com | 11:33 |
*** sshnaidm|ruck has joined #softwarefactory | 11:35 | |
*** chkumar|rover has joined #softwarefactory | 11:48 | |
fbo | apevec: now Softwarefactory-dev list run by fboucher at redhat.com, tdecacqu at redhat.com, mhuin at redhat.com, nhicher at redhat.com, apevec at redhat.com | 11:56 |
apevec | sounds good | 12:06 |
*** nchakrab has joined #softwarefactory | 12:12 | |
*** ChanServ sets mode: +o mhu | 12:34 | |
mhu | apevec, ChanServ can be used to switch ops | 12:35 |
*** nchakrab has quit IRC | 14:54 | |
*** nchakrab has joined #softwarefactory | 15:30 | |
*** nchakrab has quit IRC | 15:32 | |
*** nchakrab has joined #softwarefactory | 16:13 | |
*** nchakrab has quit IRC | 16:15 | |
*** sshnaidm|ruck is now known as sshnaidm|bbl | 16:37 | |
*** apevec has quit IRC | 17:41 | |
*** sshnaidm|bbl is now known as sshnaidm|ruck | 20:11 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!