| *** lixiaoy1 has joined #openstack-cinder | 00:06 | |
| *** brinzhang has joined #openstack-cinder | 00:12 | |
| *** zhanglong has joined #openstack-cinder | 00:41 | |
| *** bswartz has quit IRC | 00:43 | |
| *** rosmaita has left #openstack-cinder | 01:00 | |
| *** TuanDA has joined #openstack-cinder | 01:03 | |
| *** faizy98 has joined #openstack-cinder | 01:29 | |
| *** faizy_ has quit IRC | 01:32 | |
| *** jdillaman has quit IRC | 01:42 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 01:43 | |
| *** jdillaman has joined #openstack-cinder | 01:45 | |
| *** mhen has quit IRC | 01:55 | |
| *** mhen has joined #openstack-cinder | 01:57 | |
| *** raunak12 has quit IRC | 01:57 | |
| *** gnufied has quit IRC | 02:12 | |
| *** zigo has quit IRC | 02:29 | |
| *** raunak12 has joined #openstack-cinder | 02:30 | |
| *** lbragstad has joined #openstack-cinder | 02:45 | |
| *** psachin has joined #openstack-cinder | 02:53 | |
| *** Dinesh_Bhor has quit IRC | 02:57 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 03:05 | |
| *** dave-mccowan has quit IRC | 03:05 | |
| *** raunak12 has quit IRC | 03:13 | |
| *** simondodsley has joined #openstack-cinder | 03:44 | |
| *** yikun has quit IRC | 03:49 | |
| *** raunak12 has joined #openstack-cinder | 03:49 | |
| *** udesale has joined #openstack-cinder | 03:54 | |
| *** Dinesh_Bhor has quit IRC | 03:56 | |
| *** mmethot_ has joined #openstack-cinder | 04:04 | |
| *** smcginnis has quit IRC | 04:06 | |
| *** mmethot has quit IRC | 04:06 | |
| *** lixiaoy1 has quit IRC | 04:07 | |
| *** faizy_ has joined #openstack-cinder | 04:07 | |
| *** faizy98 has quit IRC | 04:11 | |
| *** lbragstad has quit IRC | 04:21 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 04:29 | |
| *** icey has quit IRC | 04:30 | |
| *** icey has joined #openstack-cinder | 04:30 | |
| *** raunak12 has quit IRC | 04:30 | |
| *** pcaruana has joined #openstack-cinder | 05:15 | |
| *** dr_gogeta86 has quit IRC | 05:32 | |
| *** lemko has joined #openstack-cinder | 05:58 | |
| *** dpawlik has joined #openstack-cinder | 06:30 | |
| *** Dinesh_Bhor has quit IRC | 06:45 | |
| *** rcernin has quit IRC | 07:03 | |
| *** sam_wan has joined #openstack-cinder | 07:12 | |
| *** Luzi has joined #openstack-cinder | 07:29 | |
| *** jiapei has joined #openstack-cinder | 07:34 | |
| *** sam_wan has quit IRC | 07:35 | |
| *** faizy98 has joined #openstack-cinder | 07:37 | |
| *** faizy_ has quit IRC | 07:41 | |
| *** sam_wan has joined #openstack-cinder | 07:41 | |
| *** jiapei has left #openstack-cinder | 07:58 | |
| *** e0ne has joined #openstack-cinder | 07:59 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 08:11 | |
| *** faizy_ has joined #openstack-cinder | 08:20 | |
| *** faizy98 has quit IRC | 08:20 | |
| *** brinzhang has quit IRC | 08:32 | |
| *** brinzhang has joined #openstack-cinder | 08:32 | |
| *** zigo has joined #openstack-cinder | 08:35 | |
| *** e0ne has quit IRC | 08:44 | |
| *** dpawlik has quit IRC | 08:46 | |
| *** e0ne has joined #openstack-cinder | 08:46 | |
| *** dpawlik has joined #openstack-cinder | 08:51 | |
| *** psachin has quit IRC | 08:55 | |
| *** sam_wan has quit IRC | 09:11 | |
| *** e0ne has quit IRC | 09:13 | |
| *** e0ne has joined #openstack-cinder | 09:16 | |
| *** faizy98 has joined #openstack-cinder | 09:35 | |
| *** faizy_ has quit IRC | 09:37 | |
| *** psachin has joined #openstack-cinder | 09:45 | |
| *** imacdonn has quit IRC | 09:53 | |
| *** imacdonn has joined #openstack-cinder | 09:53 | |
| *** Luzi has quit IRC | 09:55 | |
| *** Dinesh_Bhor has quit IRC | 09:57 | |
| *** zhanglong has quit IRC | 09:58 | |
| *** brinzhang has quit IRC | 10:02 | |
| *** TuanDA has quit IRC | 10:13 | |
| *** hoangcx has quit IRC | 10:17 | |
| *** psachin has quit IRC | 10:33 | |
| *** e0ne has quit IRC | 10:34 | |
| *** Dinesh_Bhor has joined #openstack-cinder | 10:35 | |
| *** udesale has quit IRC | 10:36 | |
| *** psachin has joined #openstack-cinder | 10:49 | |
| *** Dinesh_Bhor has quit IRC | 11:10 | |
| *** faizy_ has joined #openstack-cinder | 11:20 | |
| *** faizy98 has quit IRC | 11:24 | |
| *** e0ne has joined #openstack-cinder | 11:27 | |
| *** psachin has quit IRC | 11:28 | |
| *** psachin has joined #openstack-cinder | 11:33 | |
| *** dave-mccowan has joined #openstack-cinder | 11:38 | |
| *** rosmaita has joined #openstack-cinder | 11:38 | |
| *** abishop has joined #openstack-cinder | 11:52 | |
| *** jaypipes is now known as leakypipes | 11:58 | |
| *** e0ne has quit IRC | 12:16 | |
| *** e0ne has joined #openstack-cinder | 12:17 | |
| *** gnufied has joined #openstack-cinder | 12:20 | |
| *** hoonetorg has joined #openstack-cinder | 12:24 | |
| *** psachin has quit IRC | 12:45 | |
| *** mriedem has joined #openstack-cinder | 12:58 | |
| *** mmethot_ has quit IRC | 13:00 | |
| *** mmethot has joined #openstack-cinder | 13:03 | |
| *** eharney has joined #openstack-cinder | 13:10 | |
| *** markvoelker has quit IRC | 13:15 | |
| *** mriedem is now known as hansmoleman | 13:22 | |
| *** munimeha1 has joined #openstack-cinder | 13:22 | |
| *** lbragstad has joined #openstack-cinder | 13:23 | |
| *** erlon has joined #openstack-cinder | 13:28 | |
| *** munimeha1 has quit IRC | 13:38 | |
| *** dansmith is now known as SteelyDan | 13:41 | |
| *** munimeha1 has joined #openstack-cinder | 13:57 | |
| hansmoleman | is there any way to create a volume in read-only mode immediately? rather than have to hit the volume action API for that? | 13:57 |
|---|---|---|
| hansmoleman | see https://storyboard.openstack.org/#!/story/2002158 | 13:57 |
| hansmoleman | iow, merge os-update_readonly_flag and the 'readonly' parameter into the volume create API | 13:58 |
| hansmoleman | that seems like a no-brainer | 13:58 |
| hansmoleman | new microversion?! | 13:59 |
| *** smcginnis has joined #openstack-cinder | 14:09 | |
| *** ChanServ sets mode: +o smcginnis | 14:09 | |
| *** eharney has quit IRC | 14:09 | |
| *** bswartz has joined #openstack-cinder | 14:23 | |
| *** lbragstad is now known as elbragstad | 14:25 | |
| zul | smcginnis: what py3.7 issues with dell emc? | 14:25 |
| zul | (im just curious) | 14:26 |
| smcginnis | zul: I get errors when I run unit tests under py 3.7. I forget exactly where now. I started to look into it, but it wasn't immediately obvious so I haven't gotten back to it. | 14:26 |
| smcginnis | * running now | 14:27 |
| zul | smcginnis: cool thanks | 14:27 |
| smcginnis | StopIteration failures in 3 tests/ | 14:28 |
| smcginnis | One in vmax, two in vnx. | 14:29 |
| smcginnis | tox -e py37 -- dell_emc.vnx.test_adapter.TestCommonAdapter.test_do should hit the two. | 14:29 |
| *** munimeha1 has quit IRC | 14:32 | |
| *** faizy_ has quit IRC | 14:35 | |
| *** markvoelker has joined #openstack-cinder | 14:41 | |
| *** eharney has joined #openstack-cinder | 14:44 | |
| *** dustins has joined #openstack-cinder | 14:56 | |
| *** e0ne has quit IRC | 14:58 | |
| *** jmlowe has quit IRC | 15:03 | |
| *** keper7 has joined #openstack-cinder | 15:04 | |
| keper7 | hi, I'm seeing some really weird behaviour where cinder is mounting the wrong iscsi volume. Provider is Pure Storage, cinder is Pike from Kolla-Ansible. cinder get-pools shows 4 different pools, one per node, for the pure device. Cinder manageable-list shows different output for each pool. Fdisk on one of my hosts has a bunch of volumes it shouldn't. Has anyone ever seen anything like this? | 15:06 |
| *** e0ne has joined #openstack-cinder | 15:06 | |
| smcginnis | patrickeast, simondodsley: Around? ^ | 15:07 |
| simondodsley | yes | 15:07 |
| simondodsley | let me look | 15:07 |
| keper7 | Hey Simon! | 15:08 |
| keper7 | Guess I could have emailed you, you helped me earlier over email | 15:08 |
| keper7 | this is the same cloud that had the weird duplicate iqn initiators before they were corrected | 15:09 |
| simondodsley | keper7: this could be an artifact of those duplicate iqns - lets take this to email | 15:09 |
| keper7 | ok | 15:09 |
| *** dpawlik has quit IRC | 15:17 | |
| *** abishop has quit IRC | 15:17 | |
| *** jmlowe has joined #openstack-cinder | 15:18 | |
| *** e0ne has quit IRC | 15:37 | |
| *** e0ne has joined #openstack-cinder | 15:40 | |
| *** e0ne has quit IRC | 15:40 | |
| *** abishop has joined #openstack-cinder | 15:44 | |
| *** abishop has quit IRC | 15:45 | |
| *** dpawlik has joined #openstack-cinder | 15:47 | |
| *** abishop has joined #openstack-cinder | 15:48 | |
| *** markvoelker has quit IRC | 15:51 | |
| *** dpawlik has quit IRC | 15:51 | |
| *** markvoelker has joined #openstack-cinder | 15:52 | |
| *** erlon has quit IRC | 16:02 | |
| *** hansmoleman is now known as mriedem_away | 16:18 | |
| *** Bhujay has joined #openstack-cinder | 17:07 | |
| simondodsley | so simple question. If we have multiple backends configured all with the same volume_backend_name do they have to all use the same volume_driver or is that irrelevant? Wanting to achieve load balancing over multiple backends from one volume_type. | 17:22 |
| smcginnis | simondodsley: Different storage types can all be set to the same volume_backend_name. | 17:24 |
| simondodsley | smcginnis: thought so but wanted to check with you gurus. Thanks. | 17:25 |
| smcginnis | I think that should do what you are looking for, but you can also look at the various scheduler and weigher settings to control how it does that load balancing within them. | 17:25 |
| *** faizy98 has joined #openstack-cinder | 17:25 | |
| simondodsley | leaving the scheduling to the customer who asked about multiple vendors using the same volume_backend_name.... | 17:26 |
| smcginnis | Good call | 17:27 |
| *** e0ne has joined #openstack-cinder | 17:28 | |
| *** Bhujay has quit IRC | 17:50 | |
| *** dustins_ has joined #openstack-cinder | 18:02 | |
| *** dustins has quit IRC | 18:04 | |
| *** eharney has quit IRC | 18:34 | |
| *** e0ne has quit IRC | 18:37 | |
| *** eharney has joined #openstack-cinder | 18:45 | |
| *** kmalloc is now known as needscoffee | 18:48 | |
| *** pooja-jadhav has joined #openstack-cinder | 19:03 | |
| *** pooja_jadhav has quit IRC | 19:05 | |
| *** pooja_jadhav has joined #openstack-cinder | 19:10 | |
| *** jmlowe has quit IRC | 19:12 | |
| *** pooja-jadhav has quit IRC | 19:13 | |
| *** rosmaita has left #openstack-cinder | 19:32 | |
| *** rosmaita has joined #openstack-cinder | 19:39 | |
| *** needscoffee is now known as kmalloc | 20:20 | |
| *** dustins_ has quit IRC | 20:34 | |
| *** ianychoi has joined #openstack-cinder | 20:35 | |
| *** eharney has quit IRC | 20:40 | |
| *** mriedem_away is now known as mriedem | 20:51 | |
| *** dave-mccowan has quit IRC | 20:57 | |
| *** lemko has quit IRC | 20:58 | |
| mriedem | cinder doesn't have a db archive command? | 21:02 |
| mriedem | oh purge i see | 21:02 |
| smcginnis | Nothing like a good purge. | 21:03 |
| mriedem | smcginnis: any thoughts on my read-only flag on volume create question from this morning | 21:03 |
| mriedem | or maybe no one creates readonly volumes | 21:03 |
| smcginnis | IIRC, it was broken for awhile. So probably no one cares. | 21:05 |
| smcginnis | But I had IRC issues this morning - I don't think I saw your question. | 21:05 |
| mriedem | re https://storyboard.openstack.org/#!/story/2002158 | 21:06 |
| mriedem | if we could create a volume with readonly=False then we could avoid the create and update dance | 21:06 |
| smcginnis | You mean readonly=True? | 21:07 |
| mriedem | sorry yeah | 21:08 |
| smcginnis | Looks like the openstackclient code for create does the create first, then tries to set the readonly flag. | 21:08 |
| mriedem | yeah, but doesn't wait for the volume to be available, | 21:08 |
| smcginnis | Is it not waiting long enough and it's still in "creating"? | 21:08 |
| mriedem | i'll push a change for that at some point | 21:08 |
| smcginnis | Ah | 21:08 |
| mriedem | unrelated, i think i've also found that encryption types get orphaned in the cinder db | 21:08 |
| smcginnis | I think there was some talk of that and not being able to authoritatively decide if they can be deleted. Not sure. | 21:09 |
| mriedem | ok, | 21:09 |
| smcginnis | But as far as readonly, seems like we should allow that on "creating" states too. | 21:09 |
| mriedem | cinder encryption-type-list doesn't show it, but it's in the db in the encryption table | 21:09 |
| smcginnis | We can change that probably. I don't think we would need to mv that since it could be considered a bug fix that was already assumed to work (obviously by the openstackclient code). | 21:10 |
| mriedem | umm, | 21:10 |
| smcginnis | mriedem: Is it 'deleted=true'? | 21:10 |
| mriedem | that's a request schema change, surely a microversion | 21:10 |
| mriedem | no it's deleted=0 | 21:10 |
| mriedem | http://paste.openstack.org/show/731998/ | 21:10 |
| mriedem | oh i guess you mean allow setting readonly on creating status | 21:11 |
| mriedem | that's still a behavior change | 21:11 |
| mriedem | anyway, i'll fix osc for how the api works today | 21:11 |
| mriedem | and open a bug for this other thing | 21:11 |
| smcginnis | Would it really be a schema change? | 21:11 |
| mriedem | i meant to change POST /volumes to take a readonly param | 21:12 |
| smcginnis | Not adding the flag to the create call itself, but not to error out when it's attempted to be set while it's in creating. | 21:12 |
| mriedem | yeah that's not a schema change, it's a behavior change | 21:12 |
| mriedem | just forget i asked :) | 21:12 |
| smcginnis | :) | 21:12 |
| smcginnis | Oh, it's a contrib API. So we can do whatever the heck we want to it. :P | 21:17 |
| mriedem | https://bugs.launchpad.net/cinder/+bug/1797640 | 21:18 |
| openstack | Launchpad bug 1797640 in Cinder "encryption records are not deleted" [Undecided,Triaged] | 21:18 |
| mriedem | not sure what the severity would be on ^ | 21:18 |
| mriedem | or impact, maybe just crufty db over time | 21:18 |
| mriedem | better title would be "encryption records are not deleted when related volume type is deleted" | 21:18 |
| smcginnis | It's hard to tell how many people are using that. | 21:19 |
| smcginnis | Or how often they create and delete them. | 21:19 |
| mriedem | yup | 21:19 |
| mriedem | i think what you can do with encrypted volumes is pretty limited in nova at least | 21:19 |
| mriedem | only the libvirt driver supports it, and live migration didn't support those types of volumes until i think queens and even then a few restrictions (luks and very new qemu) | 21:20 |
| mriedem | lyarwood would know more than me | 21:20 |
| smcginnis | I seem to remember hearing of a few other restrictions still present lately too. | 21:20 |
| mriedem | oh i'm sure, we don't really test it, | 21:20 |
| smcginnis | Yeah, I usually defer all encryption things to eharney. :) | 21:20 |
| mriedem | tempest has 1 scenario test where it just creates an encrypted volume, attaches/detaches, that's it | 21:21 |
| smcginnis | No read/write of data? | 21:22 |
| mriedem | don't think so | 21:22 |
| mriedem | https://github.com/openstack/tempest/blob/master/tempest/scenario/test_encrypted_cinder_volumes.py | 21:22 |
| smcginnis | Doesn't sound like it's actually testing anything specific to encryption then. | 21:22 |
| mriedem | it just tests all the plumbing between nova and cinder | 21:23 |
| mriedem | but nothing within the guest | 21:23 |
| mriedem | barbican might have some more exotic tempest tests, idk | 21:23 |
| mriedem | https://github.com/openstack/barbican-tempest-plugin/blob/master/barbican_tempest_plugin/tests/scenario/test_volume_encryption.py | 21:23 |
| mriedem | looks like that's better, | 21:24 |
| mriedem | it writes a timestamp to the volume and reads off it | 21:24 |
| smcginnis | Seems like the minimum that should be done. | 21:24 |
| mriedem | agree | 21:25 |
| smcginnis | Umm, gerrit bot. Sleeping on the job? | 21:38 |
| smcginnis | https://review.openstack.org/610177 | 21:39 |
| smcginnis | mriedem: ^ | 21:39 |
| smcginnis | And with that, I'm outta here. | 21:40 |
| *** jistr has quit IRC | 21:55 | |
| *** jistr has joined #openstack-cinder | 21:56 | |
| *** jistr has quit IRC | 22:05 | |
| *** mriedem is now known as mriedem_away | 22:05 | |
| *** jistr has joined #openstack-cinder | 22:06 | |
| *** pcaruana has quit IRC | 22:20 | |
| *** rosmaita has left #openstack-cinder | 22:54 | |
| *** elbragstad has quit IRC | 22:56 | |
| *** elbragstad has joined #openstack-cinder | 23:15 | |
| *** elbragstad has quit IRC | 23:15 | |
| *** jmlowe has joined #openstack-cinder | 23:18 | |
| DHE | I am preparing to deploy Ceph as the Cinder backend. I have it working in a lab but have one concern. How does Cinder load an image out of glance onto an RBD device? (Glance images stored traditionally inside glance itself and not inside cinder/ceph in any way) | 23:26 |
| DHE | as in, does cinder map the rbd as /dev/rbd0 and apply it? | 23:27 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!