| *** chhagarw has joined #openstack-powervm | 04:48 | |
| *** arunman has joined #openstack-powervm | 05:25 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 06:19 | |
| *** AlexeyAbashkin has quit IRC | 06:24 | |
| *** openstackgerrit has joined #openstack-powervm | 07:12 | |
| openstackgerrit | Arun Mani proposed openstack/nova-powervm master: Add affinity score check attribute to flavor https://review.openstack.org/555309 | 07:12 |
|---|---|---|
| *** AlexeyAbashkin has joined #openstack-powervm | 07:50 | |
| *** k0da has joined #openstack-powervm | 08:03 | |
| *** chhagarw has quit IRC | 09:02 | |
| *** chhavi__ has joined #openstack-powervm | 09:02 | |
| *** AlexeyAbashkin has quit IRC | 11:14 | |
| *** arunman has quit IRC | 11:24 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 11:29 | |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: iSCSI pre-live-migration implementation https://review.openstack.org/556495 | 11:41 |
| openstackgerrit | Chhavi Agarwal proposed openstack/nova-powervm master: [WIP]:iSCSI pre-live-migration implementation https://review.openstack.org/556495 | 12:07 |
| *** edmondsw has joined #openstack-powervm | 12:13 | |
| *** chhavi__ has quit IRC | 12:16 | |
| *** chhagarw has joined #openstack-powervm | 13:13 | |
| *** tjakobs has joined #openstack-powervm | 13:44 | |
| *** arunman has joined #openstack-powervm | 13:53 | |
| *** arunman has quit IRC | 14:30 | |
| *** arunman has joined #openstack-powervm | 14:41 | |
| *** esberglu has joined #openstack-powervm | 14:54 | |
| esberglu | edmondsw: efried: https://review.openstack.org/#/c/554688/ | 14:55 |
| esberglu | No longer seeing the not enough procs failure on IT runs | 14:55 |
| efried | noyce | 14:55 |
| esberglu | Do we want to expose that as a conf option or just set it? | 14:55 |
| efried | good question. | 14:56 |
| efried | actually | 14:56 |
| efried | I wonder if we should set it based on the cpu_allocation_ratio | 14:56 |
| efried | 'cept that has some pretty serious baggage. | 14:59 |
| esberglu | efried: And set it to what if cpu_allocation_ration is 0 (default) | 15:04 |
| esberglu | Keep it at 0.1 to stay consistent with OOT? | 15:04 |
| edmondsw | I take it the issue with just setting it is backward compat? | 15:05 |
| efried | Not really IMO | 15:05 |
| efried | It's that we'd be hardcoding an override of a default. | 15:05 |
| edmondsw | yeah, I wouldn't think a backward compat argument would hold much water | 15:05 |
| edmondsw | efried I don't have a problem hardcoding an override of a default | 15:06 |
| efried | So... yeah, I guess let's dup the conf option we have OOT so we can moot both issues. | 15:06 |
| edmondsw | when doing so makes us consistent with OOT | 15:06 |
| efried | does it? Is the OOT conf default 0.1? | 15:06 |
| edmondsw | I thought so... am I wrong? | 15:07 |
| esberglu | OOT defaults to 0.1 | 15:07 |
| edmondsw | yep, that's what I thought I remembered from last week | 15:08 |
| efried | okay, didn't know whether we were overriding it via local.conf in CI. Because of this very issue having been encountered two years ago. | 15:08 |
| edmondsw | nope https://github.com/openstack/nova-powervm/blob/master/nova_powervm/conf/powervm.py#L28 | 15:09 |
| efried | Path of least resistance is to monkey-patch it into the CI. But I guess eventually someone is going to want it to be configurable in tree. | 15:09 |
| efried | IIRC, we just took out as many conf options as we could get away with to reduce the size of the initial patches. | 15:09 |
| edmondsw | I don't want to monkey patch | 15:09 |
| efried | At this point we can trickle 'em back in hopefully fairly easily. | 15:10 |
| esberglu | Yeah we're gonna have to start exposing conf options eventually | 15:10 |
| edmondsw | I think the path of least resistance is change it 0.1 hardcoded IT, and then talk about adding conf settings in a later patch | 15:10 |
| edmondsw | when we do introduce conf settings, the default will be 0.1 for the conf setting, so hardcoding 0.1 first is fine | 15:11 |
| esberglu | I'm on board with that | 15:11 |
| efried | shrug | 15:13 |
| efried | Personally I think the hardcoding will be a harder sell to nova cores, but worth a shot. | 15:13 |
| edmondsw | sell it as consistency with the OOT driver and fixes a CI issue | 15:14 |
| esberglu | edmondsw: Is there a reason you don't want to expose it now? | 15:14 |
| edmondsw | esberglu honestly I thought it would be easier to sell hardcoding than a conf option | 15:14 |
| edmondsw | if the cores want a conf setting, I'm fine with that, as long as the default can be 0.1 | 15:14 |
| edmondsw | I think they'll have a bigger issue about the default changing than about whether we have a conf setting or not | 15:15 |
| edmondsw | and the default needs to change | 15:15 |
| edmondsw | we need consistency with OOT | 15:15 |
| edmondsw | efried? | 15:15 |
| efried | So then the really path of least resistance would be to make the conf option default to None or 0.5 and then override it in our CI. | 15:16 |
| efried | But I think we can sell consistency with OOT. | 15:16 |
| edmondsw | I totally didn't follow that | 15:16 |
| edmondsw | default to None or 0.5 doesn't seem like an option... | 15:17 |
| edmondsw | oh, you mean if we can't sell consistency... yeah, I think we have to | 15:18 |
| edmondsw | efried you're needed on slack | 15:18 |
| efried | oh, look, I wasn't signed in for some reason. | 15:18 |
| esberglu | edmondsw: efried: So we're going hardcode for now on the basis of consistency with a fall back plan of exposing it as a conf option? | 15:20 |
| esberglu | I just need to update the commit message then | 15:20 |
| edmondsw | esberglu I think so | 15:20 |
| edmondsw | I added something about introducing conf options on the todo etherpad | 15:22 |
| esberglu | edmondsw: Okay. The localdisk IT change introduces some conf options already | 15:22 |
| esberglu | disk_driver and volume_group_name | 15:22 |
| edmondsw | sure | 15:23 |
| edmondsw | I think it's ok to start introducing conf options as needed | 15:23 |
| edmondsw | and if efried wants to lead with that here, I'm fine with it | 15:24 |
| edmondsw | so long as the default is 0.1 | 15:24 |
| efried | I think that's preferable to hardcoding, yes. | 15:24 |
| edmondsw | I yield to whatever option he thinks will be best | 15:24 |
| esberglu | If we expose conf options here it will also reduce the localdisk footprint | 15:24 |
| esberglu | Okay conf option it is, patch incoming | 15:25 |
| *** k0da has quit IRC | 15:25 | |
| efried | dig | 15:26 |
| *** arunman has quit IRC | 15:36 | |
| *** arunman has joined #openstack-powervm | 15:42 | |
| *** arunman has quit IRC | 15:47 | |
| *** arunman has joined #openstack-powervm | 15:49 | |
| *** arunman has quit IRC | 15:54 | |
| *** arunman has joined #openstack-powervm | 15:56 | |
| *** arunman has quit IRC | 16:01 | |
| *** arunman has joined #openstack-powervm | 16:24 | |
| openstackgerrit | Arun Mani proposed openstack/nova-powervm master: Add affinity score check attribute to flavor https://review.openstack.org/555309 | 16:37 |
| *** arunman has quit IRC | 16:45 | |
| *** chhagarw has quit IRC | 16:53 | |
| *** arunman has joined #openstack-powervm | 17:06 | |
| openstackgerrit | Arun Mani proposed openstack/nova-powervm stable/queens: Add affinity score check attribute to flavor https://review.openstack.org/556593 | 17:11 |
| edmondsw | efried I think ^ is good now, ready for you to look | 17:11 |
| *** AlexeyAbashkin has quit IRC | 17:15 | |
| efried | done | 17:18 |
| *** arunman has quit IRC | 17:43 | |
| *** arunman has joined #openstack-powervm | 17:47 | |
| *** arunman has quit IRC | 17:52 | |
| *** k0da has joined #openstack-powervm | 18:05 | |
| *** k0da has quit IRC | 18:26 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 18:42 | |
| esberglu | efried: edmondsw: Localdisk is ready for you guys to start reviewing | 18:43 |
| edmondsw | ack | 18:43 |
| efried | ack | 18:43 |
| edmondsw | esberglu just left you a comment on the proc_unit_factor change | 18:43 |
| esberglu | edmondsw: ack | 18:44 |
| esberglu | edmondsw: efried: I did have one q on localdisk. | 18:47 |
| esberglu | OOT we have https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/disk/driver.py#L349 | 18:47 |
| *** k0da has joined #openstack-powervm | 18:47 | |
| esberglu | Which will retry _create_disk_from_image() a few times | 18:47 |
| esberglu | For SSP IT, we didn't include that retry logic at all | 18:48 |
| esberglu | I also didn't include it for localdisk | 18:48 |
| esberglu | But didn't have any context on why that retry logic was in there in the 1st place | 18:49 |
| edmondsw | esberglu good question... I have no idea why that's there though | 18:50 |
| esberglu | edmondsw: I'm inclined to leave it as is (no retries) unless I hit something live testing | 18:51 |
| edmondsw | yeah | 18:51 |
| esberglu | edmondsw: efried: We also need to decide how to handle conf options IT | 18:52 |
| esberglu | So that we don't get duplicate opt errors OOT | 18:52 |
| efried | hum, I had this answer in my head once. | 18:53 |
| edmondsw | I remember that coming up the other day | 18:53 |
| esberglu | edmondsw: I don't think we ever came to a decision, we just discussed it IIRC | 18:53 |
| *** AlexeyAbashkin has quit IRC | 18:54 | |
| edmondsw | would we have to put something IT first, to check for nova settings with the same name and not register in that case? | 18:54 |
| efried | The retry thing I think is bogus. We were experiencing intermittent failures of the upload and had no idea why, so we put the retry in to work around it. We've since completely reworked the upload mechanisms from soup to nuts and are no longer having those failures (or we're retrying some other way, or something). | 18:56 |
| efried | Actually I think we killed it with pypowervm.util.retry_io_command. | 18:56 |
| edmondsw | so it should come out IT then | 18:56 |
| edmondsw | esberglu add a TODO for that? | 18:57 |
| efried | Sounds like it's already out IT. Isn't that what we said? | 18:57 |
| esberglu | OOT you mean | 18:57 |
| efried | yeah | 18:57 |
| edmondsw | right, OOT, sorry | 18:57 |
| *** AlexeyAbashkin has joined #openstack-powervm | 19:14 | |
| *** AlexeyAbashkin has quit IRC | 19:18 | |
| efried | esberglu: See mriedem's question in -nova? | 20:20 |
| *** edmondsw has quit IRC | 20:21 | |
| *** edmondsw has joined #openstack-powervm | 20:47 | |
| *** edmondsw has quit IRC | 21:05 | |
| *** edmondsw has joined #openstack-powervm | 21:08 | |
| *** edmondsw has quit IRC | 21:26 | |
| *** tjakobs has quit IRC | 22:00 | |
| *** esberglu has quit IRC | 22:04 | |
| *** esberglu has joined #openstack-powervm | 22:04 | |
| *** esberglu has quit IRC | 22:09 | |
| *** AlexeyAbashkin has joined #openstack-powervm | 22:12 | |
| *** AlexeyAbashkin has quit IRC | 22:16 | |
| tonyb | esberglu, edmondsw, efried: are any of y'all around? | 22:55 |
| efried | tonyb: I'm here | 22:55 |
| efried | the others are gone | 22:55 |
| tonyb | efried: 1 is all I need ;P | 22:56 |
| efried | Not if it's the wrong one... | 22:56 |
| tonyb | efried: ;p | 22:56 |
| efried | Like, if you're going to ask me about RHEL packaging, yer scrood | 22:56 |
| tonyb | efried: https://review.openstack.org/#/c/554977/ ... do you know if esberglu picked 3.3.? for a technical reason or was it just "it's the latest" | 22:56 |
| tonyb | efried: clearly esberglu would be best to answer that but there are good reasons to pick 3.2.17 and I don't wnat to break things WRT 554977 | 22:58 |
| efried | tonyb: Are any of the things explicitly mentioned in the commit message missing in 3.2.17? | 22:58 |
| efried | arm64/ppc64? | 22:58 |
| tonyb | let me check I know ppc64le is supported in 3.2 I don't knwo about ppc64(be) | 22:59 |
| tonyb | efried: I have to admit I just assumed that was a typo | 22:59 |
| efried | tonyb: I can be *nearly* certain he just picked the latest. But yeah, you'll have to ask him to be sure. | 23:00 |
| efried | assumed what was a typo? | 23:00 |
| tonyb | efried: So from an architecture POV 3.2 and 3.3 support the same list (based on a 5sec read of the release notes) | 23:01 |
| tonyb | and neither support ppc64(be) it's le all the way ;P | 23:01 |
| tonyb | efried: "arm64 and ppc64" ppc64 vs ppc64le | 23:01 |
| tonyb | efried: from esberglu's commit message | 23:02 |
| efried | Oh. I wasn't aware of any big-endian that we cared about, so I read ppc64 as ppc64le anyway. | 23:02 |
| tonyb | huh 3.217 ... skip a few 3.3.2 strange releases from etcd ;P | 23:02 |
| efried | Okay. So you'd like to -1 this patch with a comment to the effect that 3.2.17 would be preferred for <reasons>? Based on the info you've provided above, that seems like a reasonable thing to do. | 23:03 |
| tonyb | efried: Right yeah that was basically what I did | 23:03 |
| efried | As your alter ego ianw? | 23:03 |
| tonyb | efried: Yeah infra are happy to fix it if we can settle on a version | 23:03 |
| efried | cool cool. | 23:04 |
| efried | tonyb: I dropped a link to this discussion in there FYI. | 23:05 |
| efried | well, if gerrit decides to get back to me. | 23:05 |
| efried | there we go. | 23:05 |
| tonyb | efried: Cool. I pined esberglu when I poked infra ianw responded quicker than expected ;P | 23:06 |
| *** AlexeyAbashkin has joined #openstack-powervm | 23:12 | |
| *** AlexeyAbashkin has quit IRC | 23:16 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!