| *** mdrabe has quit IRC | 00:27 | |
| AndyWojo | Is it better to install Novalink via impi using opal or HMC/PowerVM? | 00:30 |
|---|---|---|
| AndyWojo | Is there any performance considerations? | 00:31 |
| thorst | AndyWojo: NovaLink is PowerVM. It doesn't run on Opal | 00:32 |
| thorst | :-) | 00:32 |
| thorst | NovaLink is just another way to manage your PowerVM | 00:32 |
| *** svenkat has joined #openstack-powervm | 00:39 | |
| *** mdrabe has joined #openstack-powervm | 00:40 | |
| AndyWojo | Ohhh, okay great | 00:45 |
| AndyWojo | Well that makes sense. I was able to get more time from my executives to make this a priority, so I'm back to building this out | 00:46 |
| AndyWojo | Well in that case, connecting my 812L to my lab's HMC just seems like it'll make things easier than working through IPMI | 00:47 |
| *** mdrabe has quit IRC | 00:48 | |
| *** thorst has quit IRC | 01:30 | |
| openstackgerrit | YuYang proposed openstack/nova-powervm master: Specs:Neutron Networking Guide link https://review.openstack.org/461586 | 01:59 |
| *** svenkat has quit IRC | 02:28 | |
| *** thorst has joined #openstack-powervm | 02:30 | |
| *** thorst has quit IRC | 02:50 | |
| *** thorst has joined #openstack-powervm | 03:47 | |
| *** thorst has quit IRC | 03:51 | |
| *** arunman has joined #openstack-powervm | 04:02 | |
| *** thorst has joined #openstack-powervm | 05:48 | |
| *** thorst has quit IRC | 05:53 | |
| *** k0da has joined #openstack-powervm | 06:26 | |
| *** k0da has quit IRC | 06:41 | |
| *** chas has joined #openstack-powervm | 06:53 | |
| *** chas has quit IRC | 07:15 | |
| *** chas has joined #openstack-powervm | 07:23 | |
| *** chas has quit IRC | 07:23 | |
| *** thorst has joined #openstack-powervm | 07:49 | |
| *** thorst has quit IRC | 08:09 | |
| *** chas has joined #openstack-powervm | 08:48 | |
| *** chas has quit IRC | 08:49 | |
| *** thorst has joined #openstack-powervm | 09:06 | |
| *** thorst has quit IRC | 09:10 | |
| *** k0da has joined #openstack-powervm | 09:15 | |
| *** chas has joined #openstack-powervm | 09:35 | |
| *** chas has quit IRC | 09:47 | |
| *** edmondsw has joined #openstack-powervm | 09:55 | |
| *** edmondsw has quit IRC | 10:00 | |
| *** thorst has joined #openstack-powervm | 10:07 | |
| *** thorst has quit IRC | 10:12 | |
| *** chas has joined #openstack-powervm | 10:24 | |
| *** k0da has quit IRC | 10:30 | |
| *** chas has quit IRC | 10:35 | |
| *** arunman has quit IRC | 10:36 | |
| *** arunman has joined #openstack-powervm | 10:40 | |
| *** k0da has joined #openstack-powervm | 11:27 | |
| *** thorst has joined #openstack-powervm | 11:36 | |
| *** edmondsw has joined #openstack-powervm | 11:43 | |
| *** edmondsw has quit IRC | 11:47 | |
| *** smatzek has joined #openstack-powervm | 11:49 | |
| *** svenkat has joined #openstack-powervm | 11:51 | |
| *** jpasqualetto has joined #openstack-powervm | 12:12 | |
| *** edmondsw has joined #openstack-powervm | 12:12 | |
| *** efried has quit IRC | 12:15 | |
| *** jay1_ has joined #openstack-powervm | 12:25 | |
| *** efried has joined #openstack-powervm | 12:28 | |
| *** jpasqualetto_ has joined #openstack-powervm | 12:44 | |
| *** jpasqualetto has quit IRC | 12:47 | |
| *** chas has joined #openstack-powervm | 12:48 | |
| *** chas has quit IRC | 12:49 | |
| efried | #startmeeting Weekly OpenStack driver meeting | 12:59 |
| openstack | Meeting started Tue May 2 12:59:30 2017 UTC and is due to finish in 60 minutes. The chair is efried. Information about MeetBot at http://wiki.debian.org/MeetBot. | 12:59 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 12:59 |
| openstack | The meeting name has been set to 'weekly_openstack_driver_meeting' | 12:59 |
| efried | #topic cinder/iSCSI testing efforts | 12:59 |
| efried | jay1_ How's it going? | 13:00 |
| *** nbante has joined #openstack-powervm | 13:00 | |
| thorst | not sure if Jay sees it. I'll PM him. But there are issues we think with the v7k | 13:01 |
| efried | I PMed him | 13:01 |
| jay1_ | Hey Eric, still getting connectivity issues | 13:01 |
| thorst | so he's going to ask a team where we know it works to see if we can use that temporarily | 13:01 |
| thorst | and maybe see what the difference is in the configs between the two | 13:01 |
| efried | I talked to tjakobs yesterday (he was on vacation Friday) and he said he would be around today to help if we needed. | 13:01 |
| efried | He had this working in his env, right? | 13:02 |
| edmondsw | o/ | 13:02 |
| jay1_ | thorst: got the other SVC, checking the difference.. | 13:02 |
| thorst | efried: I don't think he used a v7k | 13:02 |
| thorst | I think he built a LVM iSCSI host | 13:02 |
| efried | thorst Ugh, here's where my total ignorance comes into play. | 13:02 |
| thorst | :-) | 13:03 |
| thorst | there isn't just one storage in the world :-p | 13:03 |
| thorst | what we've been told to test is storwize | 13:03 |
| efried | Yeah, I (kinda) get that; what I don't have a grasp on is what piece of the puzzle we're getting stuck on. | 13:03 |
| thorst | which I know PVC has done with iSCSI on KVM | 13:03 |
| thorst | yeah, right now it looks like the storage side | 13:03 |
| efried | If I'm debugging vif plugging, I don't care if I'm SSP or localdisk, kind of thing. | 13:03 |
| thorst | the negotiation between the hosts (storage and compute) is failing | 13:04 |
| efried | Okay. | 13:04 |
| thorst | looks like its on the storage side | 13:04 |
| efried | gfm was helpful the other day. Maybe we can ask him again. | 13:04 |
| efried | and yfeng, and darosale. | 13:05 |
| efried | Basically, let's get a lot of people involved and see who's useful. | 13:05 |
| efried | Cause it won't be me. | 13:05 |
| thorst | +2 | 13:05 |
| jay1_ | Yeah.. ideally when I registered the ISCSI host it is still showing the ports as offline, host and svc are not able discover each other. | 13:06 |
| thorst | jay1_: Yeah, that could be that the SVC you used (or v7k) only has one wire on it. | 13:06 |
| thorst | which is why I'm hoping you can just use what the KVM team uses | 13:06 |
| thorst | if it works on KVM it should work for PowerVM | 13:07 |
| edmondsw | jay1_ I assume the word "ideally" was a mistake there? | 13:07 |
| jay1_ | typo :p | 13:07 |
| edmondsw | :) np | 13:07 |
| edmondsw | making sure I'm following... | 13:07 |
| *** madhaviy has joined #openstack-powervm | 13:07 | |
| edmondsw | (sort of) | 13:08 |
| *** chhavi has joined #openstack-powervm | 13:08 | |
| chhavi | hi all | 13:08 |
| chhavi | efried,thorst: iscsiadm discovery is failing on the setup | 13:08 |
| efried | Hi chhavi. We were going to try to gather some SMEs to debug this. | 13:09 |
| jay1_ | thorst, efried: chhavi was also looking into this today. | 13:09 |
| efried | Be nice to get it done first thing, so nobody in India has to stay up super late. | 13:09 |
| thorst | ok. I think the next actions are clear | 13:09 |
| thorst | try with the SVC we know is set up properly. | 13:10 |
| efried | Else figure a way to hand it off to the US so we can get it ready for jay1_ to take over in the (India) morning. | 13:10 |
| chhavi | yes, first this needs to be debug from the SVC side why the iscsiadm discovery for the iscsi target iqn is failing | 13:10 |
| chhavi | another point efried, i think we should fix that as well in nova-powervm is if get_iscsi_initiator is not found, we should not pass the connector with the wwpn | 13:11 |
| efried | That sounds like a thing tjakobs could look into? | 13:12 |
| thorst | chhavi: why not? | 13:12 |
| chhavi | if the connection-type is iscsi, there is no use of passing wwpn, because of that SVC is using that and providing different error logs | 13:12 |
| *** mdrabe has joined #openstack-powervm | 13:12 | |
| thorst | the host can have both FC and iSCSI connectivity simulatenously | 13:12 |
| thorst | it's up to the driver to decide what to return | 13:12 |
| thorst | maybe the storwize code has a bug there | 13:12 |
| thorst | I'd rather fix the storwize code... | 13:12 |
| chhavi | thorst: from the nova side, if user had requested for the the iscsi attachment, and we are passing the connector, why one should send the wwpn | 13:13 |
| chhavi | creating the connector is nova task | 13:14 |
| thorst | chhavi: but when get_volume_connector is called from the nova manager you have no idea what the BDM's are. | 13:14 |
| thorst | so you don't know what they're asking for - iSCSI or FC | 13:14 |
| thorst | so you can't try to outsmart the driver. | 13:15 |
| jay1_ | Just verified on the other SVC which is being used with PKVM, there is no chap secret code set on it. It is blank. | 13:15 |
| edmondsw | ugh... that doesn't sound good | 13:16 |
| thorst | OK - we may need to work with chhavi and gfm to continue | 13:16 |
| edmondsw | which SVC is that? | 13:16 |
| chhavi | when i tested with iSCSI earlier also, i never set the CHAP secret code, and i got the iscsiadm as well. | 13:16 |
| edmondsw | we should be testing with appropriate security setup, to match a real customer environment. | 13:17 |
| chhavi | shall i try to remove the chap secret and retry, if the iscsiadm discovery works. | 13:17 |
| chhavi | edmondsw: before we test with CHAP enabled to rule out | 13:18 |
| edmondsw | chhavi sure, I'm not saying we have to *only* test with security setup, but we should at least *also* test with security setup | 13:18 |
| edmondsw | if not only | 13:18 |
| chhavi | yeah, i know | 13:18 |
| edmondsw | i.e. testing without security is optional... testing with security is not | 13:18 |
| thorst | so, I agree...but lets walk before we run here? | 13:19 |
| chhavi | edmondsw: one more thing do you know how to configure CHAP for username/password | 13:19 |
| thorst | lets get the negotiation going and then make it more complex | 13:19 |
| thorst | we're fighting enough problems already here. | 13:19 |
| thorst | once we solve those, we can add the complexity | 13:19 |
| edmondsw | thorst right, my horror is more that the pkvm testing was doing this... not powervm | 13:20 |
| thorst | sure | 13:20 |
| edmondsw | chhavi sorry, no, I don't know how to set that up | 13:20 |
| thorst | I just don't want jay1_ or chhavi going down the rabbit hole and delaying this stuff. That is a parallel separate thread | 13:20 |
| chhavi | i removed the chap secret for now to see | 13:20 |
| edmondsw | thorst yes, I'll ping gfm about it... if someone will tell me which SVC this was | 13:20 |
| efried | Agree we should try to get it working at all first. Then get it working right. | 13:21 |
| edmondsw | we all agree :) | 13:21 |
| thorst | k | 13:21 |
| * efried tries to figure out how to use meetbot's #agreed tag... | 13:21 | |
| edmondsw | so... jay1_ or chhavi, which SVC did you get from the pkvm testers that didn't have CHAP setup? | 13:22 |
| efried | #agreed Get it working first. Then get it working right. | 13:22 |
| efried | (guess we'll see how that shows up in the minutes) | 13:22 |
| efried | (docs are not helpful) | 13:22 |
| jay1_ | edmondsw: you want the Ip of it ? | 13:22 |
| edmondsw | sure... something I can use to identify it when I talk to gfm | 13:23 |
| jay1_ | sure.. let me ping that to you. | 13:23 |
| edmondsw | tx | 13:23 |
| chhavi | ping me as well | 13:23 |
| jay1_ | sure folks sending this to you all. | 13:24 |
| *** mdrabe_ has joined #openstack-powervm | 13:24 | |
| *** mdrabe has quit IRC | 13:24 | |
| esberglu_ | o/ | 13:24 |
| efried | We done with that topic? | 13:26 |
| thorst | yep | 13:27 |
| efried | #topic In-tree driver | 13:28 |
| efried | No movement since last week. Hesitant to poke mriedem/sdague. Don't want to be *too* squeaky. Thoughts? | 13:28 |
| thorst | I'd hold off for a few more days | 13:29 |
| thorst | after that powervm binge :-) | 13:29 |
| thorst | (which was awesome) | 13:29 |
| thorst | or, maybe just ask them if you should hold off for a few days | 13:29 |
| efried | "a few more days" will take us to forum. | 13:29 |
| thorst | I mean, doesn't hurt to ask | 13:29 |
| edmondsw | you already have Sean's +2 on https://review.openstack.org/#/c/391288/ and https://review.openstack.org/#/c/409402/ | 13:31 |
| efried | Yes. | 13:31 |
| efried | Sigh | 13:31 |
| efried | #action efried to poke mriedem about https://review.openstack.org/#/c/391288/ and https://review.openstack.org/#/c/409402/ | 13:31 |
| edmondsw | I think the other 4 reviews all still need work, right? | 13:32 |
| efried | Yes. | 13:32 |
| efried | The SSP one doesn't need work. Just a stupid xenserver recheck, and re-look from cores. | 13:33 |
| efried | Wouldn't mind in-team +1s there. | 13:33 |
| *** smatzek has quit IRC | 13:33 | |
| efried | Previous patch set was +2ed by sdague; latest patch set was just a (manual) rebase. | 13:33 |
| edmondsw | efried I added myself | 13:34 |
| efried | Oh, were you not on there? Sorry. | 13:34 |
| edmondsw | nope... I'll try to look later today | 13:34 |
| efried | thx | 13:34 |
| edmondsw | do we need to say anything about the other 3, or next topic? | 13:35 |
| efried | I'm going to wait til after the forum to get re-cranking on those, I think. | 13:35 |
| edmondsw | +1 | 13:35 |
| efried | Unless by some miracle all three of the ready ones get merged, like, today. | 13:35 |
| efried | Anything else in-tree? | 13:35 |
| thorst | how's the non-powervm driver stuff going? | 13:36 |
| edmondsw | thorst non-powervm? | 13:36 |
| efried | #topic OOT driver | 13:36 |
| efried | Oh, non-powervm. | 13:36 |
| thorst | I'll follow up later. Its not powervm related...but the service BP | 13:36 |
| thorst | we can chat outside the meeting | 13:36 |
| efried | thorst Yeah, let's do that. | 13:37 |
| edmondsw | oh, the catalog stuff | 13:37 |
| efried | Yaknow, since this is the PowerVM driver meeting. | 13:37 |
| thorst | ;-) | 13:37 |
| edmondsw | :) | 13:37 |
| efried | So OOT driver. | 13:37 |
| efried | I reviewed esberglu_'s patches yesterday. Good start, a few comments. | 13:37 |
| efried | https://review.openstack.org/#/c/461147/ https://review.openstack.org/#/c/460331/ | 13:37 |
| efried | edmondsw Might as well add you to those. | 13:38 |
| thorst | I'll review today | 13:38 |
| thorst | and agree with edmondsw being on review | 13:38 |
| edmondsw | +1 | 13:38 |
| efried | done | 13:38 |
| esberglu_ | efried: Yep gonna fix those up. Gonna continue to work on the backports from in-tree this week | 13:38 |
| edmondsw | are any of these for carrying our in-tree changes forward into the OOT driver? | 13:38 |
| efried | edmondsw Yes, both of them. | 13:38 |
| edmondsw | awesome | 13:38 |
| efried | And the pending ones esberglu_ mentions above. | 13:38 |
| efried | Any other OOT work in the offing? Bugs? Blueprint work? Do we want to talk about the SR-IOV metrics? | 13:39 |
| thorst | I know that there is one patch that we need to review for svenkat today | 13:39 |
| efried | thorst link? | 13:40 |
| *** a1fisher has joined #openstack-powervm | 13:40 | |
| efried | Not on my list (at least not if svenkat is the author) | 13:40 |
| thorst | I don't see it, he just brought it up in a call but I don't see it anywhere in the queue | 13:40 |
| thorst | ahh, I see it now | 13:40 |
| thorst | https://review.openstack.org/#/c/460672/ | 13:40 |
| thorst | ymadhavi proposed it on their behalf | 13:41 |
| thorst | you +2'd | 13:41 |
| thorst | so I'm on point here. | 13:41 |
| svenkat | yes.. i added my +1 to it | 13:41 |
| thorst | nevermind! | 13:41 |
| efried | Added edmondsw | 13:41 |
| efried | for form's sake | 13:41 |
| esberglu_ | I don't know if you saw in slack yesterday but the SRIOV utilization data is currently available from phyp | 13:41 |
| efried | esberglu_ Needs to be added to REST, right? | 13:42 |
| efried | changh on the hook for that? | 13:42 |
| edmondsw | I see 2 changes with the same subject and associated bug: the one above and https://review.openstack.org/461653 | 13:43 |
| edmondsw | why is that? | 13:43 |
| esberglu_ | efried: I was talking to him about it. I can finish up that convo today, got distracted by CI stuff yesterday | 13:43 |
| efried | edmondsw Is the latter the ocata cherry-pick? | 13:44 |
| efried | yeah | 13:44 |
| edmondsw | efried ah, yeah... so we're not waiting for it to merge first | 13:44 |
| efried | edmondsw Waiting for what to merge? | 13:45 |
| efried | oh, waiting for master to merge before cherry-picking? | 13:45 |
| edmondsw | the master change, before backporting | 13:45 |
| edmondsw | not a big deal | 13:45 |
| efried | Nah, this one's pretty trivial, we're pretty confident it's not gonna change since the +2. | 13:45 |
| edmondsw | sure | 13:45 |
| edmondsw | especially since thorst just +W'd master :) | 13:46 |
| efried | #action thorst +W https://review.openstack.org/461653 | 13:46 |
| efried | Anything else OOT? | 13:46 |
| efried | #topic CI | 13:47 |
| thorst | nada from me | 13:47 |
| efried | esberglu_ Go. | 13:47 |
| esberglu_ | CI died last night | 13:47 |
| esberglu_ | Well not died but the runs are all failing | 13:47 |
| esberglu_ | Looks to be cell related (which I still don't really understand that wel) | 13:47 |
| *** openstackgerrit has quit IRC | 13:48 | |
| esberglu_ | HostMappingNotFound: Host 'powervm-ci-powervm-devstacked-4112' is not mapped to any cell | 13:48 |
| esberglu_ | I think we should be able to just do that nova-manage discover_hosts thing hopefully | 13:48 |
| esberglu_ | Gonna start debugging that right after the meeting | 13:48 |
| *** dwayne has joined #openstack-powervm | 13:49 | |
| efried | k | 13:49 |
| efried | Couple other things to keep an eye on: | 13:49 |
| esberglu_ | Other that that there are some tests that are failing on and off. I think after the summit I'm gonna devote a week/sprint | 13:49 |
| efried | Live migration tests started failing last night. We probably have those disabled. | 13:49 |
| esberglu_ | to debugging those intermittent failures and trying to work on the temp. disabled tests in the skip lists | 13:49 |
| efried | http://lists.openstack.org/pipermail/openstack-dev/2017-May/116160.html | 13:50 |
| efried | Also take a look at https://review.openstack.org/#/c/461716/ to see if it affects us. | 13:50 |
| esberglu_ | efried: Yeah I tried a manual run with that the other day. It failed stacking, but looked like a problem connecting to git.o.o | 13:51 |
| efried | esberglu_ Okay, I guess keep it on yer watch list. | 13:51 |
| esberglu_ | Should be able to try again in the background today | 13:52 |
| efried | Anything else CI? | 13:52 |
| esberglu_ | Nope | 13:52 |
| *** openstackgerrit has joined #openstack-powervm | 13:53 | |
| openstackgerrit | Merged openstack/nova-powervm master: Deallocate network on reschedule https://review.openstack.org/460672 | 13:53 |
| efried | #topic Open discussion | 13:53 |
| efried | Anything? Anyone? | 13:53 |
| thorst | esberglu_: keep fighting the good fight! Great job with the CI :-) | 13:53 |
| esberglu_ | Haha thanks | 13:54 |
| edmondsw | efried, does that OOT network deallocate change need to make it into one of our in-tree driver patches? | 13:54 |
| efried | Okay, if nothing else... | 13:54 |
| edmondsw | or are we not far enough along there to need to worry about that? | 13:54 |
| efried | edmondsw Not until we have networks in tree ;-) | 13:54 |
| edmondsw | right | 13:54 |
| edmondsw | I think one of your outstanding patches is for that, no? | 13:55 |
| chhavi | thorst: does neo host network is configured to receive all packets | 13:55 |
| efried | Yeah, good idea to slap a comment on https://review.openstack.org/#/c/422512/ | 13:55 |
| edmondsw | will do | 13:55 |
| chhavi | i am just wondering if the neo network can be an issue for iscsi discovery | 13:55 |
| chhavi | as per the error it says the neo is not able to receive the packets | 13:55 |
| efried | edmondsw Done. | 13:55 |
| efried | #endmeeting | 13:56 |
| openstack | Meeting ended Tue May 2 13:56:10 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 13:56 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/weekly_openstack_driver_meeting/2017/weekly_openstack_driver_meeting.2017-05-02-12.59.html | 13:56 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/weekly_openstack_driver_meeting/2017/weekly_openstack_driver_meeting.2017-05-02-12.59.txt | 13:56 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/weekly_openstack_driver_meeting/2017/weekly_openstack_driver_meeting.2017-05-02-12.59.log.html | 13:56 |
| edmondsw | efried we both made the same comment :) | 13:56 |
| *** smatzek has joined #openstack-powervm | 13:56 | |
| thorst | chhavi: send me the whole env details | 13:56 |
| thorst | I'll take a look when I break out of the meeting. | 13:56 |
| thorst | if you can ssh to the host, it can receive packets. | 13:56 |
| esberglu_ | thorst: Is that ovs meeting still in the works? | 13:56 |
| thorst | it doesn't have multiple NICs | 13:56 |
| thorst | esberglu_: o crap...yeah. | 13:57 |
| thorst | I'll schedule for Friday right now. | 13:57 |
| edmondsw | efried thorst want to talk about that service catalog work now? | 13:58 |
| efried | thorst Avoid EOD Friday. Ima be on the hook for school runs :( | 13:59 |
| edmondsw | efried I saw you had new patches last night, haven't checked them yet. | 13:59 |
| efried | edmondsw thorst Well, the ksa patch is merging! | 13:59 |
| edmondsw | awesome! | 13:59 |
| efried | Totally | 13:59 |
| *** thorst is now known as thorst_afk | 13:59 | |
| efried | I think what happens now is, that needs to appear in a release of ksa, then I bump the g-r for ksa, then my nova patch starts verifying. | 14:00 |
| *** esberglu_ is now known as esberglu | 14:00 | |
| edmondsw | efried so have you talked to the keystone guys about releasing ksa? | 14:01 |
| efried | edmondsw Not since about a week ago. I'll fup now. | 14:02 |
| *** madhaviy_ has joined #openstack-powervm | 14:03 | |
| *** madhaviy has quit IRC | 14:04 | |
| *** madhaviy_ is now known as madhaviy | 14:04 | |
| *** jpasqualetto_ has quit IRC | 14:18 | |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm master: In-Tree Backport: TaskFlow https://review.openstack.org/460331 | 14:21 |
| *** jpasqualetto_ has joined #openstack-powervm | 14:37 | |
| *** tjakobs_ has joined #openstack-powervm | 14:38 | |
| *** arunman has quit IRC | 14:38 | |
| *** arunman has joined #openstack-powervm | 14:40 | |
| efried | chhavi Are you still working on the iSCSI thing? | 14:46 |
| *** esberglu has quit IRC | 14:46 | |
| chhavi | discovery worked, after giving proper target ip | 14:47 |
| chhavi | verifying the attach | 14:47 |
| efried | Great! | 14:48 |
| efried | tjakobs_ ^^ chhavi is working on it with jay1_ | 14:48 |
| tjakobs_ | ok cool, chhavi can you let me know if you run into anymore issues. And i can try to help | 14:51 |
| chhavi | sure, i will just run once and verify if the nova-powervm is now returning the proper connector with initiatorname | 14:52 |
| *** esberglu has joined #openstack-powervm | 14:52 | |
| chhavi | thorst,tjakobs_: "nova_powervm.virt.powervm.volume.get_iscsi_initiator" is still returning the null initiator | 14:58 |
| chhavi | do we have a pvmctl to do iscsi target discovery | 14:59 |
| chhavi | tjakobs_ u can have a look further will be back after dinner | 15:00 |
| tjakobs_ | I don't think there is any pvmctl for iscsi | 15:00 |
| efried | tjakobs_ Do you have the system details? | 15:09 |
| tjakobs_ | i don't | 15:09 |
| efried | If you can tell me the pypowervm commands we use to do that discovery, I can try to punch it in with ipython. | 15:09 |
| efried | jay1_ Is this still on neo34? | 15:09 |
| tjakobs_ | efried: pypowervm command is here: https://github.com/powervm/pypowervm/blob/develop/pypowervm/tasks/hdisk/_iscsi.py#L83 | 15:10 |
| chhavi | yes efried its still on neo34 | 15:10 |
| chhavi | i don't see the ISCSIDiscovery job invoked in the pypowervm logs | 15:11 |
| efried | tjakobs_ And the VIOS in this case will be... the nvl or the real VIOS? | 15:11 |
| chhavi | https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/volume/__init__.py#L103 | 15:12 |
| tjakobs_ | found this in Doc, don't remember exact reason however... "The uuid of the VIOS (VIOS must be a Novalink VIOS type)." | 15:12 |
| edmondsw | efried congrats on what I assume is your first keystoneauth merge! | 15:12 |
| efried | edmondsw It is indeed. Thanks! | 15:13 |
| efried | tjakobs_ chhavi Looking at that code above, is it possible we try the discovery the first time, it returns nothing, then we set _ISCSI_LOOKUP_COMPLETE to True, so it doesn't even try it the next time? | 15:14 |
| efried | I.e. does the actual deployment process do something that would make those initiators available? After the initial driver load? | 15:14 |
| chhavi | yes i think since this is global, even on failure this is getting true | 15:14 |
| jay1_ | efried: Yes it is still neo34 | 15:14 |
| efried | tjakobs_ Well, the nvl here isn't a VIOS type. So does that put a kibosh on the whole process? | 15:15 |
| chhavi | i am setting it false and retrying | 15:16 |
| efried | In [9]: ret = hdisk.discover_iscsi_initiator(adap, v.uuid) | 15:16 |
| efried | In [10]: ret | 15:16 |
| efried | Out[10]: 'iqn.1993-08.org.debian:01:3718826ceea8\n' | 15:16 |
| efried | That's using 'vios1' | 15:16 |
| chhavi | yes, | 15:16 |
| tjakobs_ | that looks like the right output to me | 15:16 |
| efried | ...which is a real VIOS | 15:16 |
| chhavi | condition of global variable shud be set to true if the initiator is not null | 15:17 |
| efried | chhavi Not sure that's the problem here. | 15:18 |
| efried | In this case, the nvl is the management partition, but it's not a VIOS type. | 15:18 |
| efried | So the line https://github.com/openstack/nova-powervm/blob/master/nova_powervm/virt/powervm/volume/__init__.py#L103 will be False, and we won't attempt the discovery. | 15:18 |
| efried | Can anyone explain why we can't do iscsi through a traditional VIOS? | 15:19 |
| efried | Do we need to reinstall in SDE mode? Or do we need to fix the code to allow iSCSI through real VIOS? | 15:19 |
| efried | thorst_afk ^^ | 15:19 |
| chhavi | efried: this env is in SDE mode | 15:21 |
| chhavi | output above returned for iqn is correct as per the iscsi-initiatorname | 15:21 |
| chhavi | why we need to pass the mgmt partition | 15:21 |
| *** mdrabe_ has quit IRC | 15:22 | |
| thorst_afk | chhavi: iScsi through the PowerVM driver is only working currently through the NL partition (like OVS support) | 15:23 |
| efried | chhavi If it's in SDE mode, why is the nvl not a VIOS type? | 15:23 |
| thorst_afk | efried: not through a traditional VIOS | 15:23 |
| efried | thorst_afk Right, we can see that - but why? | 15:23 |
| thorst_afk | efried: It was installed in SDN mode, so that it had a VIOS | 15:23 |
| thorst_afk | for the SSP support | 15:23 |
| thorst_afk | there is a command to post config change the NVL to VIOS type | 15:24 |
| efried | Can't we have traditional VIOS + nvl-as-vios? | 15:24 |
| thorst_afk | to enable co-exist of the two | 15:24 |
| thorst_afk | the installer just doesn't plop it down that way | 15:24 |
| efried | seroyer question? | 15:24 |
| thorst_afk | though maybe it should | 15:24 |
| thorst_afk | I'd ask apearson first | 15:24 |
| efried | thorst_afk chhavi jay1_ tjakobs_ Okay, flipped nvl to VIOS. | 15:26 |
| efried | I assume we have to restart services to make this work. | 15:26 |
| thorst_afk | yep | 15:26 |
| efried | esberglu Heads-up https://review.openstack.org/#/c/461803/ failed PowerVM CI hard, and they're merging it. | 15:36 |
| efried | Can't see how it could affect us, but... | 15:36 |
| esberglu | efried: Everything is failing hard | 15:36 |
| efried | okay. | 15:36 |
| *** mdrabe_ has joined #openstack-powervm | 15:36 | |
| esberglu | efried: I'm actually gonna disable runs for the time being until I get this figured out | 15:37 |
| efried | esberglu Doesn't look like our driver is being hit at all. Is this the cells setup deal? | 15:37 |
| esberglu | Yeah | 15:37 |
| efried | k | 15:37 |
| efried | chhavi jay1_ Did you retry since I flipped the nvl to VIOS mode? | 15:38 |
| jay1_ | efried: Shall I try now ? | 15:38 |
| efried | jay1_ Need to restart services. | 15:38 |
| efried | but yes | 15:38 |
| efried | Something is still wrong. | 15:40 |
| efried | Now when I try discovery, giving the nvl UUID, I get: | 15:40 |
| efried | The 'ISCSIDiscovery' operation failed. HTTP error 404 for method PUT on path /rest/api/uom/VirtualIOServer/45DB94F8-D3CA-4F53-B5CB-8C26A31A9022/do/ISCSIDiscovery: Not Found -- REST000B The URL presented to the Management Console REST Web Services is not valid. | 15:40 |
| efried | Mebbe we need to restart pvm-rest? | 15:41 |
| thorst_afk | efried: was the REST server restarted? | 15:41 |
| efried | nope. I'll do that now... | 15:41 |
| jay1_ | efried: once done, pls let me know will try attach | 15:42 |
| efried | When I said 'services', I was talking about openstack services. thorst_afk Those need to be restarted too, presumably? | 15:42 |
| thorst_afk | efried: yep | 15:42 |
| thorst_afk | efried: just really nova-compute | 15:42 |
| efried | jay1_ Wait until pvm-rest is done restarting. | 15:42 |
| efried | Then restart n-cpu | 15:42 |
| jay1_ | ok.. | 15:43 |
| efried | bingo. | 15:43 |
| efried | jay1_ Restart n-cpu and retry attach. | 15:43 |
| jay1_ | sure | 15:44 |
| efried | I'm getting 'iqn.1993-08.org.debian:01:3718826ceea8\n' from discover_iscsi_attach with the nvl UUID. | 15:44 |
| *** k0da has quit IRC | 15:45 | |
| thorst_afk | that seems reasonable | 15:45 |
| efried | It's the same thing I was getting from the VIOS before. | 15:45 |
| efried | And in fact I still get the same thing from the VIOS. Is that normal? | 15:45 |
| thorst_afk | right...I assumed the problem was the vscsi mappings (since you weren't a VIOS) | 15:45 |
| jay1_ | efried: attached a volume "New_VOL1" to the VM: "New_ISCSI_VM" | 15:46 |
| efried | jay1_ Successfully? | 15:46 |
| jay1_ | no errors on the UI though | 15:46 |
| thorst_afk | pvmctl scsi list | 15:46 |
| thorst_afk | what's the output | 15:47 |
| jay1_ | let me verify more.. | 15:47 |
| efried | pvmctl scsi list only shows the boot disk and cfg drive | 15:47 |
| thorst_afk | VM was powered off for the attach? | 15:48 |
| thorst_afk | I know that volume/nic attachment are odd. They 'submit the job' successfully but then it can fail behind the scenes | 15:48 |
| efried | nope. | 15:48 |
| efried | jay1_ Shut down the LPAR, and try again. | 15:48 |
| jay1_ | right now it is not successful, let me try shoutdown and attach | 15:49 |
| thorst_afk | it should work if powered on, but only if RMC was good | 15:49 |
| thorst_afk | so you can just avoid that with shutting it down | 15:49 |
| *** chas has joined #openstack-powervm | 15:49 | |
| jay1_ | RMC is active only but VM details shows blank volumes_attached: | 15:49 |
| jay1_ | let me try power off VM now | 15:50 |
| thorst_afk | good deal | 15:50 |
| *** chas has quit IRC | 15:54 | |
| jay1_ | Just tried attaching New_VOL2 to New_ISCSI_VM, but not reflecting that on the UI also. | 15:59 |
| efried | yup, also not seeing it in scsi list. | 16:00 |
| efried | Wait, does chhavi still have a pdb going in there? | 16:01 |
| efried | doesn't look like it. | 16:01 |
| efried | We should tame down the logging a bit. | 16:01 |
| efried | jay1_ Did you restart the compute service before you tried this? | 16:02 |
| jay1_ | yes but not after shutoff VM | 16:02 |
| efried | I wouldn't expect that should matter. | 16:03 |
| jay1_ | Yeah.. | 16:03 |
| thorst_afk | so no luck? | 16:03 |
| thorst_afk | want me to hop in and poke around? | 16:03 |
| jay1_ | thorst_afk: yes volumes are not getting attached to shutoff VM also. | 16:07 |
| thorst_afk | k. | 16:08 |
| thorst_afk | I'm going to pop in now | 16:08 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm master: In Tree Backport: Exception handling #1 https://review.openstack.org/461147 | 16:10 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm master: In Tree Backport: Exception handling #1 https://review.openstack.org/461147 | 16:11 |
| *** jay1_ has quit IRC | 16:17 | |
| *** madhaviy has quit IRC | 16:20 | |
| thorst_afk | {'initiator': 'iqn.1993-08.org.debian:01:3718826ceea8\n', 'wwpns': ['21000024ff64f56c'], 'host': 'neo34'} | 16:23 |
| thorst_afk | it seems like a bug that there is a \n on the initiator | 16:23 |
| thorst_afk | tjakobs_: is that what you would expect? | 16:23 |
| thorst_afk | FYI - its failing in cinder. | 16:31 |
| thorst_afk | I'm investigating. | 16:31 |
| thorst_afk | so this is totally a bug in the SVC driver. | 16:33 |
| *** jay1_ has joined #openstack-powervm | 16:34 | |
| thorst_afk | I'll see about proposing a patch up | 16:34 |
| chhavi | thorst: its failing while matching the iscsi hostname | 16:35 |
| thorst_afk | right. | 16:35 |
| thorst_afk | its a bug in their driver. | 16:36 |
| thorst_afk | I'm going to propose a fix | 16:36 |
| thorst_afk | this isn't a nova_powervm bug... | 16:36 |
| thorst_afk | https://bugs.launchpad.net/cinder/+bug/1343142 | 16:36 |
| openstack | Launchpad bug 1343142 in Cinder "IBM storwize_svc can not get the right host" [Medium,Fix released] - Assigned to Li Min Liu (liminliu) | 16:36 |
| thorst_afk | looks like it was a bug back in juno | 16:36 |
| thorst_afk | and must've been reintroduced in one of the many re-writes | 16:36 |
| chhavi | thorst: i don't think the issue mentioned in the above fix, is the problem in our environment | 16:48 |
| chhavi | thorst_afk: | 16:48 |
| thorst_afk | chhavi: trying it now. | 16:52 |
| thorst_afk | it no longer detects the wrong host | 16:52 |
| chhavi | issue was with /n na or something else | 16:52 |
| thorst_afk | its two fold. | 16:53 |
| thorst_afk | that's an issue in nova-powervm | 16:53 |
| thorst_afk | for sure | 16:53 |
| thorst_afk | but also there is one in cinder. | 16:53 |
| *** nbante has quit IRC | 16:54 | |
| *** arunman has quit IRC | 16:55 | |
| openstackgerrit | Shyama proposed openstack/nova-powervm master: Allow vios to be filtered for rebuild https://review.openstack.org/460723 | 16:55 |
| chhavi | could be in chap secret | 16:57 |
| chhavi | logging off let us know, we can pick tomorrow | 16:59 |
| *** chhavi has quit IRC | 17:05 | |
| thorst_afk | efried: when you get a chance: https://review.openstack.org/#/c/460723/2 | 17:08 |
| thorst_afk | jay1_: for what its worth, I now have it actually calling into the nova-powervm code | 17:11 |
| jay1_ | thorst_afk: Didn't get it. | 17:33 |
| jay1_ | is it working now ? | 17:33 |
| jay1_ | ah okay.. got it now. | 17:35 |
| esberglu | efried: thorst_afk: Still figuring out CI. Not a cell issue like I initially thought. | 17:45 |
| esberglu | https://github.com/openstack-dev/devstack/commit/bb4431126b80be4805f1072442676797e6019576 | 17:45 |
| esberglu | I think that's what's hitting us | 17:45 |
| esberglu | Should be able to get around it by setting TEMPEST_AUTH_VERSION in our local.conf | 17:47 |
| esberglu | I don't have any clue about the identity stuff. Looks like it is defaulting to use v3 but we use v2 | 17:48 |
| esberglu | Maybe we should be making the move to v3? | 17:48 |
| *** jay1_ has quit IRC | 17:51 | |
| openstackgerrit | Drew Thorstensen (thorst) proposed openstack/nova-powervm master: Fix newline character on iscsi initiator https://review.openstack.org/461863 | 18:01 |
| jpasqualetto_ | Hey everyone.. I've asked a question a few days ago, pointing to this errors: https://pastebin.com/0FisErDm | 18:53 |
| jpasqualetto_ | thorst_afk, told me "it looks like you only have 5 slots associated with it you should probably bump that up to like 1000 (which will require a reboot)" | 18:53 |
| jpasqualetto_ | well, a team member changed this configuration.. but still not working... how can I check this? | 18:54 |
| thorst_afk | efried: do you know how to dump out the slots assigned to a partition? | 19:19 |
| efried | `pvmctl scsi list` will show you current adapters. | 19:20 |
| efried | ...including slot numbers. | 19:20 |
| efried | But that's probably not what you're looking for, is it? | 19:20 |
| efried | There are VIOS commands, but I would have to punt to someone like seroyer for that. | 19:21 |
| jpasqualetto_ | efried, this command show me my current adapters... but, what about a limit? It seems that my problem is some kind of limitation. Is there a way to see it? | 19:29 |
| efried | jpasqualetto_ Yeah, should be. Gimme a sec... | 19:30 |
| efried | jpasqualetto_ pvmctl vm list -i id=<LPARID> -d max_virtual_slots | 19:31 |
| efried | unless it's a VIOS you care about - then use pvmctl vios list | 19:31 |
| efried | If you wanna see all the LPARs, you can do something like pvmctl vm list -d name max_virtual_slots | 19:32 |
| jpasqualetto_ | efried, ok, thans. What if I want to change this value? This should work: pvmctl vm update -i id=3 -s max_virtual_slots=1000 ? | 19:39 |
| efried | jpasqualetto_ Yes, that should work, but rumor has it you would have to reboot for it to take effect. | 19:40 |
| efried | thorst_afk Can you confirm ^^ ? | 19:40 |
| jpasqualetto_ | efried, I can reboot, no problem.. test environment :-) | 19:40 |
| thorst_afk | yeah, you need to reboot for that to take effect | 19:40 |
| *** chas has joined #openstack-powervm | 19:50 | |
| *** chas has quit IRC | 19:54 | |
| *** k0da has joined #openstack-powervm | 20:03 | |
| *** smatzek has quit IRC | 20:16 | |
| *** tjakobs__ has joined #openstack-powervm | 20:28 | |
| edmondsw | esberglu we should definitely use keystone v3 | 20:29 |
| edmondsw | v2 is going away | 20:29 |
| edmondsw | if you ever have any keystone questions, ping me | 20:29 |
| *** tjakobs_ has quit IRC | 20:31 | |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm master: DNM: ci check https://review.openstack.org/328315 | 20:53 |
| openstackgerrit | Eric Berglund proposed openstack/nova-powervm master: DNM: CI Check2 https://review.openstack.org/328317 | 20:54 |
| *** thorst_afk has quit IRC | 20:59 | |
| *** tjakobs__ has quit IRC | 21:00 | |
| *** dwayne has quit IRC | 21:02 | |
| *** tjakobs has joined #openstack-powervm | 21:05 | |
| *** thorst_afk has joined #openstack-powervm | 21:25 | |
| *** svenkat has quit IRC | 21:27 | |
| *** thorst_afk has quit IRC | 21:29 | |
| *** dwayne has joined #openstack-powervm | 21:45 | |
| *** mdrabe_ has quit IRC | 21:49 | |
| edmondsw | esberglu tx for http://morpheus.pok.stglabs.ibm.com/#/c/5216/ | 22:04 |
| *** jpasqualetto_ has quit IRC | 22:34 | |
| *** k0da has quit IRC | 22:38 | |
| *** thorst_afk has joined #openstack-powervm | 23:21 | |
| *** thorst_afk has quit IRC | 23:46 | |
| *** tjakobs has quit IRC | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!