Monday, 2020-04-06

« Sunday, 2020-04-05 Index Tuesday, 2020-04-07 »
openstackgerritMerged openstack/cinder master: PowerMax Driver - Update single underscores  https://review.opendev.org/71657200:33
*** n-saito has joined #openstack-cinder01:04
*** ociuhandu has joined #openstack-cinder01:15
*** ociuhandu has quit IRC01:19
*** ociuhandu has joined #openstack-cinder03:13
*** mvkr has quit IRC03:15
*** psachin has joined #openstack-cinder03:19
openstackgerritwanghao proposed openstack/python-cinderclient master: Add support for Block Storage API mv 3.60  https://review.opendev.org/71071903:19
*** ociuhandu has quit IRC03:22
*** ociuhandu has joined #openstack-cinder03:24
*** mvkr has joined #openstack-cinder03:28
*** ociuhandu has quit IRC03:29
*** ykarel|away is now known as ykarel04:12
*** udesale has joined #openstack-cinder04:51
*** udesale has quit IRC04:52
*** udesale has joined #openstack-cinder04:52
*** udesale has quit IRC05:03
*** udesale has joined #openstack-cinder05:03
*** ociuhandu has joined #openstack-cinder06:28
*** dpawlik has joined #openstack-cinder06:28
*** ociuhandu has quit IRC06:33
*** zhanglong has joined #openstack-cinder07:03
*** ociuhandu has joined #openstack-cinder07:13
*** e0ne has joined #openstack-cinder07:13
*** ociuhandu has quit IRC07:18
openstackgerritRajat Dhasmana proposed openstack/cinder-tempest-plugin master: Add Snapshot data integrity test  https://review.opendev.org/70249507:21
*** tosky has joined #openstack-cinder07:58
*** spsurya_ has joined #openstack-cinder07:59
*** dpawlik has quit IRC08:29
*** dpawlik has joined #openstack-cinder08:38
*** martinkennelly has joined #openstack-cinder08:40
*** ociuhandu has joined #openstack-cinder08:41
*** ociuhandu has quit IRC08:45
*** zhanglong has quit IRC08:46
*** ykarel is now known as ykarel|lunch08:50
*** martinkennelly has quit IRC08:51
*** ajitha has joined #openstack-cinder08:58
*** rcernin has quit IRC09:03
*** ociuhandu has joined #openstack-cinder09:13
*** e0ne has quit IRC09:13
*** e0ne has joined #openstack-cinder09:14
*** vishalmanchanda has joined #openstack-cinder09:24
openstackgerritMerged openstack/cinder master: Correct ReST syntax  https://review.opendev.org/71736409:26
*** ykarel|lunch is now known as ykarel09:36
openstackgerritHelen Walsh proposed openstack/cinder master: PowerMax Driver - Version comparison correction  https://review.opendev.org/71332909:56
openstackgerritHelen Walsh proposed openstack/cinder master: PowerMax Driver - Refreshing stale metadata  https://review.opendev.org/71297710:06
*** rcernin has joined #openstack-cinder10:37
openstackgerritRaghavendra Tilay proposed openstack/cinder master: HPE 3PAR: Support duplicated FQDN in network  https://review.opendev.org/71380310:51
*** tkajinam has quit IRC10:59
*** ociuhandu has quit IRC11:05
*** ociuhandu has joined #openstack-cinder11:05
*** ociuhandu has quit IRC11:10
*** ociuhandu has joined #openstack-cinder11:16
*** ociuhandu has quit IRC11:23
*** zhanglong has joined #openstack-cinder11:25
*** rosmaita has quit IRC11:40
*** rosmaita has joined #openstack-cinder11:42
*** andrebeltrami has joined #openstack-cinder11:47
*** tkajinam has joined #openstack-cinder11:51
*** rcernin has quit IRC12:04
*** rcernin has joined #openstack-cinder12:05
*** xuanyd has joined #openstack-cinder12:10
openstackgerritRaghavendra Tilay proposed openstack/cinder master: HPE 3PAR: Support duplicated FQDN in network  https://review.opendev.org/71380312:12
walshh_Is there anyone available to take a look at https://review.opendev.org/#/c/714723/.  It has clean CI and +2.  Thank you12:14
hemna_morning12:16
*** jv__ has quit IRC12:17
*** jv has joined #openstack-cinder12:17
*** tkajinam has quit IRC12:18
xuanydwhoami-rajat: hello, are you free now12:23
whoami-rajatxuanyd, hey12:23
xuanydwhoami-rajat: the path has updated, can you review again? https://review.opendev.org/#/c/705876/12:24
whoami-rajatxuanyd, i looked at it and the code looks fine but i haven't got time to test it yet12:25
whoami-rajatxuanyd, i will try to do it as soon as i get some free time12:26
xuanydwhoami-rajat: Ok, thanks very much12:27
whoami-rajatxuanyd, np. thanks for making it better12:28
*** udesale_ has joined #openstack-cinder12:28
*** enriquetaso has joined #openstack-cinder12:29
*** udesale has quit IRC12:31
*** Liang__ has joined #openstack-cinder12:32
*** ociuhandu has joined #openstack-cinder12:42
*** rcernin has quit IRC12:52
*** rcernin has joined #openstack-cinder12:53
*** AJaeger has joined #openstack-cinder12:57
openstackgerritBrian Rosmaita proposed openstack/cinder master: Revert "Remove ProphetStor Flexvisor Driver"  https://review.opendev.org/71738512:57
AJaegerrosmaita: gmann commented on https://review.opendev.org/#/c/717451/ that these py27 cleanups are ok to merge now, please review again.12:58
rosmaitaAJaeger: ok, but i think the nonclient libraries already went out without them12:58
AJaegerI have the following changes up for Py27 cleanups, reviews welcome, please: https://review.opendev.org/#/c/717487/ https://review.opendev.org/#/c/717451/ https://review.opendev.org/#/c/717447/ https://review.opendev.org/71761012:58
AJaegerrosmaita: still won't hurt to merge them ;)12:59
rosmaitasure, but i think it would be better if we were consistent about this across the release12:59
rosmaitabut i need to read what gmann said on that review13:00
AJaegerrosmaita: sure, please do13:00
*** Luzi has joined #openstack-cinder13:01
rosmaitai guess i am just a worrier, but it seems non-optimal to do this week of M-3 and feature freeze13:01
AJaegerrosmaita: some problems we only noticed recently...13:04
rosmaitaAJaeger: ok, that's different, if not having it there is causing problems13:04
*** eharney has joined #openstack-cinder13:07
*** ykarel is now known as ykarel|afk13:11
*** ociuhandu has quit IRC13:11
*** ociuhandu has joined #openstack-cinder13:12
*** ociuhandu has quit IRC13:17
*** xuanyd has quit IRC13:40
openstackgerritMerged openstack/python-brick-cinderclient-ext master: Cleanup py27 support  https://review.opendev.org/71745113:47
AJaegerrosmaita: uncapped dependencies with universal wheels cause problems13:53
*** zhanglong has quit IRC13:54
*** lennyb has quit IRC14:04
*** ykarel|afk is now known as ykarel14:06
*** Luzi has quit IRC14:15
openstackgerritSofia Enriquez proposed openstack/cinder-tempest-plugin master: Create new Ceph job with c-bak service enabled  https://review.opendev.org/71737914:18
openstackgerritBrian Rosmaita proposed openstack/python-cinderclient master: Add support for Block Storage API mv 3.60  https://review.opendev.org/71071914:19
walshh_Hi all, tox -e docs  is failing this morning on the reviews. Does anyone know what is going on?  Thank you14:29
walshh_ https://zuul.opendev.org/t/openstack/build/0415a4e8cfa64d75a0a1a472b5681ae0/log/job-output.txt14:29
AJaegerwalshh_: sphinx 3.0.0 was released - is that breaking it?14:29
*** gary_perkins has quit IRC14:30
walshh_all was well Friday, so it is possible I guess14:30
smcginnisAJaeger: That's probably it. There was an issue in the glance repo that I had to fix this morning due to behavior changes.14:30
smcginnisDocs source that hadn't been touched for a very long time.14:31
*** ociuhandu has joined #openstack-cinder14:31
*** gary_perkins has joined #openstack-cinder14:32
smcginnisFull traceback: http://paste.openstack.org/show/791673/14:34
smcginnisAJaeger: Looks like it's in the autodoc module. Have you seen that? ^14:34
AJaegersmcginnis: autodoc is old, you should use apidoc nowadays14:36
AJaegersmcginnis: see what I did in https://review.opendev.org/#/c/717615/14:37
AJaegersmcginnis: sorry in meetings right now, so can't help further14:37
smcginnisOdd, I don't think we specify that.14:38
AJaegerif you need help with apidoc, stephenfin can help as well14:39
AJaegerargh, cinderlib is also failing14:40
smcginnisSeveral repos: https://zuul.opendev.org/t/openstack/builds?job_name=openstack-tox-docs&result=FAILURE14:41
*** ajitha has quit IRC14:41
smcginnisAJaeger: We already use apidoc. So I don't know where autodoc is coming from.14:42
smcginnisI'll ping stephenfin.14:43
*** psachin has quit IRC14:43
walshh_https://github.com/sphinx-doc/sphinx/issues/7422  this was logged 8 minutes ago14:44
smcginnisThanks walshh_!14:44
AJaegerSo, quick fix: block 3.0.0?14:45
smcginnisAJaeger: That might be best until we can figure out what's happening.14:46
smcginnisAJaeger: From Stephen "sphinxcontrib.apidoc is using autodoc under the hood"14:46
AJaegerah!14:48
openstackgerritMerged openstack/python-cinderclient master: Cleanup py27 support  https://review.opendev.org/71744714:50
*** Liang__ has quit IRC14:51
rosmaitawhoami-rajat: that was a good suggestion on https://review.opendev.org/#/c/710719/ to get the usage info into the docs14:51
*** cmorpheus is now known as cmurphy14:55
*** Xuchu has joined #openstack-cinder15:11
*** ykarel is now known as ykarel|away15:28
*** vishalmanchanda has quit IRC15:33
*** ociuhandu has quit IRC15:36
*** ociuhandu has joined #openstack-cinder15:37
*** ociuhandu has quit IRC15:42
openstackgerritBrian Rosmaita proposed openstack/cinder master: RBD: add support for revert-to-snapshot  https://review.opendev.org/71033416:04
rosmaitajbernard: need you to look this over and make sure you agree with the commit message: ^^16:05
rosmaitasmcginnis: if you have some time, https://review.opendev.org/#/c/697636/ is ready for your re-review16:14
smcginnisack16:14
rosmaitasmcginnis: also, lmk if there's anything i can do to help with the docs problem16:15
smcginnisrosmaita: I proposed https://review.opendev.org/#/c/717772/ for now. If that release is actually bad, then we will have to blacklist until it is fixed. Otherwise if it ends up something we need to tweak in the Cinder repo, we can do that (and any other repos that need it too) and then revert that cap.16:17
rosmaitasmcginnis: thanks for following up on that16:18
smcginnisnp16:19
*** udesale_ has quit IRC16:30
*** ircuser-1 has joined #openstack-cinder16:37
hemna_does anyone remember why the scheduler has allocated_capacity_gb and provisioned_capacity_gb ?16:54
hemna_how/why are those 2 numbers different ?16:57
hemna_https://github.com/openstack/cinder/blob/master/cinder/scheduler/host_manager.py#L370-L37716:57
*** ociuhandu has joined #openstack-cinder16:57
*** e0ne has quit IRC16:59
hemna_looks like the volume manager calculates allocated_cap_gb itself from volumes that are 'in-use' and 'available'17:08
hemna_at init_host time.17:08
hemna_and then keeps track of it via provisioning requests for the lifetime of the volume manager's instance in memory.17:08
*** ociuhandu has quit IRC17:08
hemna_but if a volume was available and failed to attach between restarts, then cinder loses track of it17:09
hemna_or any other process that puts the volume in a state other than in-use or available17:09
hemna_https://github.com/openstack/cinder/blob/master/cinder/volume/manager.py#L53017:09
hemna_some drivers report provisioned_capacity_gb and others report allocated_capacity_gb17:10
hemna_smh17:10
hemna_rbd sets provisioned_capacity if and only if the rbd pool is not exclusively managed by cinder17:14
hemna_otherwise the value comes from cinder manager17:14
hemna_which has a potential bug17:14
*** AJaeger has left #openstack-cinder17:15
hemna_it would be better if drivers were consistent in what they reported17:15
hemna_I think drivers should report provisioned_capacity_gb17:15
hemna_since the volume manager uses allocated_capaciity_gb and calculates that value17:16
*** throne82 has joined #openstack-cinder17:16
hemna_and some drivers don't report either of them.17:18
hemna_smh17:18
hemna_@!17:20
_pewp_hemna_ (♦亝д 亝)ﾉ17:20
*** kaisers has quit IRC17:21
openstackgerritMerged openstack/python-cinderclient master: Add support for Block Storage API mv 3.60  https://review.opendev.org/71071917:26
*** kaisers has joined #openstack-cinder17:27
throne82Hello folks, I found a bug (https://bugs.launchpad.net/cinder/+bug/1870367) while testing extend attached volume with NFS. The problem is that qemu-img fails to get "write" lock on the image, so it can't extend the volume. The problem only occurs while the volume is attached on a nova instance.17:28
openstackLaunchpad bug 1870367 in Cinder "Fail to extend attached volume using generic NFS driver" [Undecided,New]17:28
throne82This feature affects all the vendors that are based on the generic NFS implementation.17:28
throne82This feature is marked as supported by the suppport matrix so it should be updated, and this bug fixed.17:28
throne82could someone please help me to investigate/fix this issue?17:28
*** kaisers has quit IRC17:34
*** kaisers has joined #openstack-cinder17:38
*** ajitha has joined #openstack-cinder17:48
openstackgerritFernando Ferraz proposed openstack/cinder master: NetApp SolidFire: Add active/active replication  https://review.opendev.org/71279917:49
*** dpawlik has quit IRC17:51
*** lseki_ has joined #openstack-cinder18:07
*** lseki has quit IRC18:08
*** lseki_ is now known as lseki18:08
*** ociuhandu has joined #openstack-cinder18:32
jbernardrosmaita: i agree (re commit message)18:38
rosmaitajbernard: cool, ty18:39
jbernardrosmaita: the alternative is terrible18:39
jbernardrosmaita: np, thanks for updating it18:39
*** Lucas_Gray has joined #openstack-cinder18:40
*** Lucas_Gray has quit IRC18:42
*** Lucas_Gray has joined #openstack-cinder18:44
*** ociuhandu has quit IRC18:52
*** ociuhandu has joined #openstack-cinder18:53
*** kukacz has quit IRC18:55
*** kukacz has joined #openstack-cinder18:57
*** Lucas_Gray has quit IRC18:59
*** ociuhandu has quit IRC18:59
*** Lucas_Gray has joined #openstack-cinder19:00
openstackgerritFernando Ferraz proposed openstack/cinder master: NetApp SolidFire: Add active/active replication  https://review.opendev.org/71279919:06
openstackgerritFernando Ferraz proposed openstack/cinder master: NetApp SolidFire: Add active/active replication  https://review.opendev.org/71279919:06
*** keper7 has joined #openstack-cinder19:41
*** Lucas_Gray has quit IRC19:41
keper7Hi everyone. Has anyone seen this before? I'm trying to create a Ceph-backed Cinder type (on Rocky) with multi-attach support using `--property multiattach="<is> True"`. When I try to create a volume though cinder-scheduler.log is throwing an error saying "No weighed backends available" and the volume goes straight to error state.19:44
*** Lucas_Gray has joined #openstack-cinder19:45
smcginniskeper7: The logs should state somewhere before that point why available backends get filtered out.19:55
smcginnisBut specifying multiattach on volume create is deprecated. At least I'm pretty sure it was by Rocky. That should be an extra spec in the volume type you select now.19:55
keper7Thanks, yes I'm trying to do this with the volume type. I'm specifying --type with `openstack volume create` instead of `--multiattach`19:56
keper7No weighed backend found for volume with properties: {...} could the problem be in those properties?19:56
keper7Or maybe this CapabilitiesFilter(start:1, end:0) bit?19:57
keper7Filtering removed all hosts for the request with volume ID '37c1034d-7c00-4fe2-b1be-82d9fd187375'. Filter results: AvailabilityZoneFilter: (start: 1, end: 1), CapacityFilter: (start: 1, end: 1), CapabilitiesFilter: (start: 1, end: 0)19:57
*** Lucas_Gray has quit IRC20:03
openstackgerritFernando Ferraz proposed openstack/cinder master: NetApp SolidFire: Fix failback failing after service restart  https://review.opendev.org/70824020:07
openstackgerritFernando Ferraz proposed openstack/cinder master: NetApp SolidFire: Add active/active replication  https://review.opendev.org/71279920:07
smcginniskeper7: Yeah, looks like it's the CapbilitiesFilter that's excluding it.20:08
smcginniskeper7: I think there should be something in the logs earlier stating what criteria are not being met that it filters out the backend.20:08
smcginnisMight need to set the logging to debug level, but I can't remember.20:09
keper7Thanks, I don't see anything but I think I have the logging set to the default Kolla-Ansible sets it to. I'll up it to debug and try again20:09
smcginnisI was thinking we really should log that, but in big busy systems that would probably cause a lot of unnecessary log entries.20:14
smcginnisWith debug level on, it should be a lot more clear what it isn't happy about.20:14
keper7haven't done this before - looks like I just set debug=true and verbose=true in the cinder.conf files right?20:15
smcginniskeper7: Yep. There is also an API for that too, but I don't remember if we expose that through the cinder client.20:15
*** ajitha has quit IRC20:16
keper7great, gonna run the kolla-ansible reconfigure now then. Will report back once I have new data. Thanks so much!20:17
smcginniskeper7: No problem - good luck!20:17
*** Lucas_Gray has joined #openstack-cinder20:19
*** Lucas_Gray has quit IRC20:26
*** e0ne has joined #openstack-cinder20:50
keper7Progress - so there's an error saying Volume type extra spec requirement "multiattach=<is> True" does not match reported capability "False" _satisfies_extra_specs <path...>20:52
keper7Poked around the referenced file, got me thinking. I ran `cinder get-capabilities` and I don't see `multiattach` in the list. Just compression, qos, replication_enabled, and thin_provisioning.20:53
keper7Should multiattach be in there too?20:53
smcginniskeper7: Ah, ceph didn't declare support for multiattach until stein - https://review.opendev.org/#/c/595827/20:54
smcginnisSorry - I missed that part from your first comment.20:55
keper7Haha! Woo! Yay that's so much easier to fix, I was about to upgrade this cluster anyways. THANK YOU20:55
smcginnisOh, perfect.20:55
*** Lucas_Gray has joined #openstack-cinder21:00
*** ociuhandu has joined #openstack-cinder21:02
*** ociuhandu has quit IRC21:06
*** enriquetaso has quit IRC21:08
*** e0ne has quit IRC21:16
*** Lucas_Gray has quit IRC21:35
*** enriquetaso has joined #openstack-cinder21:36
*** enriquetaso has quit IRC22:01
*** rcernin has joined #openstack-cinder22:30
*** tkajinam has joined #openstack-cinder22:42
*** Liang__ has joined #openstack-cinder23:07
*** tosky has quit IRC23:28
*** Liang__ has quit IRC23:39
*** brinzhang has joined #openstack-cinder23:57
« Sunday, 2020-04-05 Index Tuesday, 2020-04-07 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!