| *** edmondsw has joined #openstack-powervm | 00:13 | |
| *** edmondsw has quit IRC | 00:18 | |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm stable/queens: Stringify instance UUID https://review.openstack.org/567434 | 01:03 |
|---|---|---|
| *** edmondsw has joined #openstack-powervm | 02:01 | |
| *** edmondsw has quit IRC | 02:06 | |
| *** edmondsw has joined #openstack-powervm | 03:50 | |
| *** edmondsw has quit IRC | 03:54 | |
| *** edmondsw has joined #openstack-powervm | 05:38 | |
| *** edmondsw has quit IRC | 05:42 | |
| *** mujahidali has joined #openstack-powervm | 05:58 | |
| *** mujahidali has quit IRC | 06:34 | |
| *** edmondsw has joined #openstack-powervm | 07:26 | |
| *** edmondsw has quit IRC | 07:30 | |
| *** chhagarw has joined #openstack-powervm | 08:47 | |
| *** edmondsw has joined #openstack-powervm | 09:14 | |
| *** edmondsw has quit IRC | 09:19 | |
| *** edmondsw has joined #openstack-powervm | 11:02 | |
| *** edmondsw has quit IRC | 11:07 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: iSCSI Live Migration Support https://review.openstack.org/567575 | 12:12 |
| *** AlexeyAbashkin has joined #openstack-powervm | 12:28 | |
| *** AlexeyAbashkin has quit IRC | 12:37 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 12:47 | |
| *** AlexeyAbashkin has quit IRC | 12:48 | |
| *** edmondsw_ has joined #openstack-powervm | 12:51 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: iSCSI Live Migration Support https://review.openstack.org/567575 | 13:20 |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: iSCSI Live Migration Support https://review.openstack.org/567575 | 14:02 |
| *** esberglu has joined #openstack-powervm | 14:09 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: iSCSI Live Migration Support https://review.openstack.org/567575 | 14:18 |
| *** tjakobs has joined #openstack-powervm | 14:20 | |
| esberglu | edmondsw_: You around to review https://review.openstack.org/#/c/567434/? Was hoping to merge the nova-powervm ones quick to avoid patching in CI | 14:28 |
| edmondsw_ | ack | 14:28 |
| *** edmondsw_ is now known as edmondsw | 14:28 | |
| edmondsw | esberglu +2/+A | 14:30 |
| esberglu | edmondsw: tx | 14:30 |
| edmondsw | esberglu efried wondering if the corresponding nova change is correct: https://review.openstack.org/#/c/567599/1/nova/virt/powervm/vm.py | 14:32 |
| edmondsw | we're not just adding str() there but also copying the is_uuid_like thing from OOT | 14:33 |
| edmondsw | should we be? | 14:33 |
| edmondsw | if so, we probably need to explain why in the commit message | 14:34 |
| esberglu | edmondsw: I actually don't think we need it | 14:40 |
| edmondsw | I suspect not, or we'd have a bug | 14:40 |
| edmondsw | might be interesting to look at history OOT and see where that was added | 14:40 |
| esberglu | edmondsw: Will fix | 14:42 |
| efried | We either cleaned that up when we ported in tree, or the paths where it could be a string rather than an instance are not yet ported. | 14:42 |
| esberglu | edmondsw: https://github.com/openstack/nova-powervm/commit/1ad83fe4c if curious | 14:42 |
| openstackgerrit | Merged openstack/nova-powervm stable/queens: Stringify instance UUID https://review.openstack.org/567434 | 14:42 |
| efried | ah, mdrabe did it. | 14:43 |
| efried | let's blame him for all the bugs. | 14:43 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm stable/queens: DNM: ci check https://review.openstack.org/544689 | 14:43 |
| mdrabe | That happens already anwyay | 14:43 |
| edmondsw | I remember that... it was a good change. But not ported IT yet? | 14:55 |
| edmondsw | yeah, we don't have nvram and swift IT, so I guess not | 14:55 |
| edmondsw | esberglu you'll respin https://review.openstack.org/#/c/567599 then? | 14:56 |
| esberglu | edmondsw: Just pushed it | 14:56 |
| edmondsw | tx | 14:56 |
| edmondsw | esberglu can you recheck https://review.openstack.org/#/c/567418 once the queens CI is working? | 14:57 |
| edmondsw | I think I rechecked too early | 14:57 |
| esberglu | edmondsw: Cool if I just wait for the current run to fail? I can go abort it if you're in a hurry | 14:58 |
| esberglu | Also, it might work, depends on the timing | 14:58 |
| edmondsw | I can wait | 14:59 |
| esberglu | Oh wait nvm, you checked it earlier than I thought, it will fail | 14:59 |
| esberglu | I'll kick another one off after it does | 14:59 |
| *** chhagarw has quit IRC | 14:59 | |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm stable/pike: Stringify instance UUID https://review.openstack.org/567616 | 15:07 |
| esberglu | edmondsw: efried: ^ pike version | 15:08 |
| efried | duh, I can't +2 the nova changes because they're in stable. | 15:14 |
| esberglu | edmondsw: efried: Either of you have any idea what is required to get active RMC connections for tempest instances | 16:31 |
| esberglu | To be clear I'm talking about the instances created by tempest inside of the AIO devstack | 16:32 |
| esberglu | We can't attach volumes unless we can get active RMC | 16:32 |
| efried | esberglu: Sorry, no. I only have the vaguest of notions what RMC even is. | 16:32 |
| esberglu | same | 16:32 |
| efried | seroyer knows things about RMC. | 16:33 |
| esberglu | efried: edmondsw: Also have a question about this https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/driver.py#L1106 | 16:35 |
| efried | tjakobs: ^ | 16:35 |
| efried | esberglu: what about it? | 16:36 |
| esberglu | get_volume_connector is used for vSCSI, but the get_iscsi_initiator() call fails in that case | 16:36 |
| esberglu | Shouldn't we only be running that if iSCSI? | 16:36 |
| efried | That seems reasonable to me. Let's get tjakobs to have a look, shall we? | 16:37 |
| esberglu | efried: tjakobs: In the in-tree implementation we just set connector['initiator'] = None | 16:37 |
| efried | esberglu: So wait, vscsi oot is broken since https://github.com/openstack/nova-powervm/commit/390347afc6a12af7ba35e672fee9ca2bff20684d ? | 16:38 |
| esberglu | And I didn't have any issues with that part when testing in-tree | 16:38 |
| esberglu | efried: Very well could be, unless I'm missing something | 16:38 |
| tjakobs | I personally don't have any problem only running this for iSCSI. I'm guessing this isn't "supposed" to fail, just log errors if it doesn't work (shouldn't break anything) | 16:40 |
| esberglu | http://paste.openstack.org/show/720780/ | 16:41 |
| esberglu | tjakobs: ^ What I'm seeing | 16:41 |
| tjakobs | esberglu: I'm guessing you don't have a new enough VIOS that has the iscisadm commands built in. (We should tolerate this though). | 16:43 |
| tjakobs | pretty sure this was done as part of Chavvi's work to support iSCSI on traditional VIOS | 16:43 |
| *** tjakobs has quit IRC | 16:54 | |
| *** tjakobs has joined #openstack-powervm | 16:54 | |
| *** edmondsw has quit IRC | 17:02 | |
| *** edmondsw has joined #openstack-powervm | 17:03 | |
| edmondsw | esberglu at the point where we call get_volume_connector we don't know what kind of volumes we'll be dealing with | 17:25 |
| edmondsw | someone could attach iscsi, rbd, vscsi fc, etc. all to a single LPAR if they want | 17:25 |
| edmondsw | but that would be later... we don't have a particular volume in mind here | 17:25 |
| edmondsw | right, efried tjakobs? | 17:25 |
| edmondsw | so I believe https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/driver.py#L1106 is correct | 17:26 |
| edmondsw | and we just need to make sure it works by handling error cases better | 17:26 |
| edmondsw | did a bug get opened? | 17:26 |
| openstackgerrit | Matthew Edmonds proposed openstack/nova-powervm master: Handle errors getting iSCSI initiators https://review.openstack.org/567656 | 17:43 |
| edmondsw | esberglu ^ | 17:43 |
| edmondsw | please open a bug and I'll update ^ to reference it | 17:43 |
| esberglu | edmondsw: Tx, ^ worked for me | 17:46 |
| esberglu | Opening bug now | 17:47 |
| edmondsw | efried I expect this is the way it is because nova basically assumes volumes will be iSCSI. But outside of our control unless we want to change nova and the driver interface | 17:49 |
| esberglu | edmondsw: https://bugs.launchpad.net/nova-powervm/+bug/1770466 | 17:56 |
| openstack | Launchpad bug 1770466 in nova-powervm "JobRequestFailed discovering iSCSI initiator" [Undecided,New] | 17:56 |
| openstackgerrit | Matthew Edmonds proposed openstack/nova-powervm master: Handle errors getting iSCSI initiators https://review.openstack.org/567656 | 17:57 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm stable/pike: Do not merge: Checking status of zuul on pike branch https://review.openstack.org/567666 | 18:13 |
| efried | esberglu: We may need to nail down our basepython in tox.ini for ^ | 19:05 |
| openstackgerrit | Eric Fried proposed openstack/nova-powervm stable/pike: Do not merge: Checking status of zuul on pike branch https://review.openstack.org/567666 | 19:08 |
| efried | esberglu: Trying ^ | 19:08 |
| esberglu | efried: Yep, tx | 19:09 |
| openstackgerrit | Merged openstack/nova-powervm master: Handle errors getting iSCSI initiators https://review.openstack.org/567656 | 19:13 |
| esberglu | efried: Failed with your tox.ini change too | 19:43 |
| efried | esberglu: okay. The same way? I actually wonder if we need to go to py3 there. But that would be weird. | 19:44 |
| esberglu | efried: Same way | 19:44 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm stable/pike: Do not merge: Checking status of zuul on pike branch https://review.openstack.org/567666 | 20:07 |
| efried | esberglu: Keep an eye on https://review.openstack.org/#/c/567669/ -- it may or may not break that same instance UUID check again. | 21:00 |
| efried | oh, looks like pvm ci succeeded, so I guess it's not a problem. | 21:01 |
| esberglu | yep | 21:01 |
| efried | esberglu: I added a sme to https://review.openstack.org/#/c/567666/ | 21:07 |
| efried | esberglu: He's GMT, so it'll be tomorrow (at least) before he looks. | 21:07 |
| *** edmondsw has quit IRC | 21:14 | |
| *** edmondsw has joined #openstack-powervm | 21:15 | |
| *** edmondsw has quit IRC | 21:19 | |
| *** tjakobs has quit IRC | 22:06 | |
| *** esberglu has quit IRC | 22:19 | |
| *** esberglu has joined #openstack-powervm | 22:33 | |
| *** esberglu has quit IRC | 22:33 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!