| *** Ashana has joined #openstack-powervm | 00:13 | |
| *** esberglu has quit IRC | 00:15 | |
| *** Ashana has quit IRC | 00:18 | |
| *** Ashana has joined #openstack-powervm | 00:19 | |
| *** jwcroppe_ has joined #openstack-powervm | 00:20 | |
| *** jwcroppe has quit IRC | 00:22 | |
| *** Ashana has quit IRC | 00:24 | |
| *** jwcroppe has joined #openstack-powervm | 00:28 | |
| *** jwcroppe_ has quit IRC | 00:31 | |
| *** thorst has joined #openstack-powervm | 00:31 | |
| *** thorst_ has joined #openstack-powervm | 00:39 | |
| *** Ashana has joined #openstack-powervm | 00:39 | |
| *** thorst has quit IRC | 00:41 | |
| *** thorst_ is now known as thorst | 00:41 | |
| *** Ashana has quit IRC | 00:44 | |
| *** Ashana has joined #openstack-powervm | 00:57 | |
| *** Ashana has quit IRC | 01:01 | |
| *** Ashana has joined #openstack-powervm | 01:08 | |
| *** Ashana has quit IRC | 01:12 | |
| *** Ashana has joined #openstack-powervm | 01:20 | |
| *** Ashana has quit IRC | 01:24 | |
| *** Ashana has joined #openstack-powervm | 01:32 | |
| *** Ashana has quit IRC | 01:37 | |
| *** thorst has quit IRC | 01:47 | |
| *** thorst has joined #openstack-powervm | 01:48 | |
| *** thorst has quit IRC | 01:56 | |
| *** tjakobs has joined #openstack-powervm | 02:02 | |
| *** svenkat has quit IRC | 02:40 | |
| *** wangqwsh has joined #openstack-powervm | 02:48 | |
| *** thorst has joined #openstack-powervm | 02:54 | |
| *** wangqwsh has quit IRC | 02:59 | |
| *** thorst has quit IRC | 03:02 | |
| *** thorst has joined #openstack-powervm | 03:59 | |
| *** tjakobs has quit IRC | 04:01 | |
| *** thorst has quit IRC | 04:06 | |
| *** thorst has joined #openstack-powervm | 05:04 | |
| *** thorst has quit IRC | 05:11 | |
| openstackgerrit | ymadhavi@in.ibm.com proposed openstack/nova-powervm: Change in Http Error condition check from String to Int https://review.openstack.org/349960 | 05:39 |
|---|---|---|
| *** kotra03 has joined #openstack-powervm | 05:52 | |
| *** thorst has joined #openstack-powervm | 06:09 | |
| *** thorst has quit IRC | 06:16 | |
| *** Cartoon has joined #openstack-powervm | 06:55 | |
| *** thorst has joined #openstack-powervm | 06:55 | |
| *** thorst has quit IRC | 07:03 | |
| *** Ashana has joined #openstack-powervm | 07:47 | |
| *** Ashana has quit IRC | 07:51 | |
| *** Ashana has joined #openstack-powervm | 07:53 | |
| *** Ashana has quit IRC | 07:57 | |
| *** thorst has joined #openstack-powervm | 08:01 | |
| *** thorst has quit IRC | 08:08 | |
| *** k0da has joined #openstack-powervm | 08:10 | |
| *** Ashana has joined #openstack-powervm | 08:10 | |
| *** Ashana has quit IRC | 08:14 | |
| *** Ashana has joined #openstack-powervm | 08:22 | |
| *** Ashana has quit IRC | 08:26 | |
| *** kwazar has joined #openstack-powervm | 08:56 | |
| *** Ashana has joined #openstack-powervm | 09:05 | |
| *** thorst has joined #openstack-powervm | 09:06 | |
| *** Ashana has quit IRC | 09:10 | |
| *** thorst has quit IRC | 09:13 | |
| *** Ashana has joined #openstack-powervm | 09:23 | |
| *** Ashana has quit IRC | 09:27 | |
| *** Cartoon has quit IRC | 09:38 | |
| *** Ashana has joined #openstack-powervm | 09:48 | |
| *** Ashana has quit IRC | 09:53 | |
| *** Ashana has joined #openstack-powervm | 09:54 | |
| *** Ashana has quit IRC | 09:58 | |
| *** thorst has joined #openstack-powervm | 10:11 | |
| *** Ashana has joined #openstack-powervm | 10:17 | |
| *** thorst has quit IRC | 10:18 | |
| *** Ashana has quit IRC | 10:22 | |
| *** thorst has joined #openstack-powervm | 11:16 | |
| openstackgerrit | Ravi Kumar Kota proposed openstack/nova-powervm: WIP: PowerVM: Implement get_instance_diagnostics https://review.openstack.org/350525 | 11:21 |
| *** thorst has quit IRC | 11:25 | |
| *** svenkat has joined #openstack-powervm | 11:26 | |
| openstackgerrit | Ravi Kumar Kota proposed openstack/nova-powervm: WIP: PowerVM: Implement get_instance_diagnostics https://review.openstack.org/350525 | 11:33 |
| *** thorst has joined #openstack-powervm | 11:41 | |
| *** Ashana has joined #openstack-powervm | 11:44 | |
| *** smatzek has joined #openstack-powervm | 11:45 | |
| *** madhaviy has joined #openstack-powervm | 11:49 | |
| *** edmondsw has joined #openstack-powervm | 12:19 | |
| *** kotra03 has quit IRC | 12:31 | |
| *** seroyer has joined #openstack-powervm | 12:52 | |
| *** burgerk has joined #openstack-powervm | 13:03 | |
| Ashana | I am getting an apt-get update issue on my system | 13:05 |
| Ashana | it saying Failed to fetch http://us.archive.ubuntu.com/ubuntu/dists/xenial-updates/restricted/binary-amd64/Packages 503 Cache storage error - No such file or directory | 13:06 |
| Ashana | has anyone hit this issue before? | 13:06 |
| *** tblakeslee has joined #openstack-powervm | 13:10 | |
| *** mdrabe has joined #openstack-powervm | 13:12 | |
| *** burgerk has quit IRC | 13:18 | |
| *** esberglu has joined #openstack-powervm | 13:19 | |
| *** jwcroppe has quit IRC | 13:28 | |
| *** jwcroppe has joined #openstack-powervm | 13:42 | |
| esberglu | thorst: adreznec: Looks like the tox failure may be being caused by this change to jjb | 13:44 |
| esberglu | https://github.com/openstack-infra/jenkins-job-builder/commit/c569dc02d31e4d4bc58430f4dad290f3de96d182 | 13:44 |
| thorst | esberglu: looks like a simple refactor? | 13:47 |
| *** seroyer has quit IRC | 13:48 | |
| esberglu | The error is coming from line 303 in jenkins_jobs/builder.py | 13:50 |
| esberglu | self.parser = YamlParser(self.jjb_config, self.plugins_list) | 13:50 |
| thorst | yeah...but is that just the refactor? | 13:52 |
| thorst | did we change anything in our YAML? | 13:52 |
| adreznec | That... doesn't seem right? That commit is from last year... | 13:52 |
| adreznec | Unless it took this long to get into a verison | 13:53 |
| adreznec | *version | 13:53 |
| esberglu | But the actual error isn’t coming from the parser it’s from trying to load self.plugins_list | 13:53 |
| esberglu | Yeah it merged in recently | 13:54 |
| esberglu | adreznec: ^^ | 13:54 |
| *** burgerk has joined #openstack-powervm | 13:57 | |
| adreznec | esberglu: If you replace JJB master install with the latest release (1.6.1.) does it fix things? | 13:58 |
| esberglu | I think that that line above should be accessing self._plugins_list not self.plugins_list | 13:58 |
| esberglu | thorst: adreznec: Yeah changing that passed the tox | 14:03 |
| adreznec | Ok, so definitely a conflict with something merged since June 9th then | 14:04 |
| esberglu | adreznec. Changing the variable name not changing the env. | 14:05 |
| adreznec | Hmm ok | 14:05 |
| *** jwcroppe has quit IRC | 14:05 | |
| *** seroyer has joined #openstack-powervm | 14:06 | |
| esberglu | Actaully that might just be dodging the error | 14:06 |
| *** kotra03 has joined #openstack-powervm | 14:09 | |
| thorst | so, do we just have a bug that we need to propose a fix for? | 14:09 |
| thorst | the infra team is super cool about htat. | 14:10 |
| thorst | esberglu: want to make a bug? Or are we not sure its that? | 14:13 |
| esberglu | thorst: No I was just dodging the error | 14:15 |
| *** tjakobs has joined #openstack-powervm | 14:17 | |
| *** kriskend_ has joined #openstack-powervm | 14:21 | |
| *** kriskend has joined #openstack-powervm | 14:21 | |
| thorst | esberglu: https://storyboard.openstack.org/#!/project/723 | 14:22 |
| thorst | adreznec: we found the root issue. Jenkins now needs to be running for the tox | 14:34 |
| adreznec | srsly... | 14:35 |
| adreznec | that seems like a terrible requirement | 14:35 |
| thorst | for rizzles | 14:36 |
| adreznec | we should ask if that was intentional.. | 14:36 |
| *** burgerk has quit IRC | 14:38 | |
| *** efried has joined #openstack-powervm | 14:40 | |
| *** burgerk has joined #openstack-powervm | 14:43 | |
| *** mdrabe has quit IRC | 14:53 | |
| *** mdrabe has joined #openstack-powervm | 15:01 | |
| *** lmtaylor1 has joined #openstack-powervm | 15:01 | |
| thorst | Ashana: vhost0 | 15:12 |
| efried | thorst, https://review.openstack.org/#/c/350154/ / https://review.openstack.org/#/c/350268/ tested and appear copacetic. Going to W+1 the former. Still need +2 on the latter. | 15:22 |
| thorst | efried: will do in ~10 | 15:24 |
| openstackgerrit | Merged openstack/nova-powervm: Scrub storage-less VSCSI mappings in destroy https://review.openstack.org/350154 | 15:29 |
| openstackgerrit | Merged openstack/nova-powervm: Change in Http Error condition check from String to Int https://review.openstack.org/349960 | 15:39 |
| *** lmtaylor1 has quit IRC | 15:49 | |
| *** lmtaylor has joined #openstack-powervm | 15:49 | |
| *** mdrabe_ has joined #openstack-powervm | 15:51 | |
| *** mdrabe has quit IRC | 15:55 | |
| efried | thorst, svenkat and I would like to discuss SR-IOV enablement in nova | 15:55 |
| efried | Would like to explore the possibility of getting a change in core nova to enable "virtual" SR-IOV attachment. | 15:56 |
| efried | There was another hypervisor that needed to do something similar. They actually deferred their blueprint to ocata - possible partially for that reason. | 15:57 |
| svenkat | ok… | 15:58 |
| efried | Would we need an actual blueprint in nova to make that happen at this point, or could we slide it in as a "bug"? | 15:58 |
| svenkat | efried : do you know exactly what needs to be done? what chnages to pci white list | 15:58 |
| svenkat | so when you create a vnic type direct port and boot a vm with it, create_pci_requests_for_sriov_ports in nova/network/neutronv2/api.py adds pci request to network... | 15:59 |
| efried | Not exactly. You will have to create an entry that doesn't trigger any of the libvirt-centric device special file-based validation code. | 16:00 |
| efried | Means you need to avoid setting certain fields, but I don't remember exactly which ones those are. It wasn't hard to figure out by poking at the code, though. | 16:00 |
| svenkat | ok.. | 16:00 |
| efried | And the other bits should be able to be dummied out. | 16:00 |
| svenkat | ok… | 16:00 |
| efried | I think the code will simply require there to be *something* there. I'm not sure it's going to matter what it is. | 16:01 |
| svenkat | ok… | 16:01 |
| svenkat | i will look at compute init code that reads pci whitelist again. | 16:01 |
| efried | But if we could propose a simple change to nova that would allow the whitelist to be empty without erroring, that would be best. | 16:01 |
| svenkat | sure. | 16:02 |
| thorst | we could say that, but we need to find another hypervisor that can benefit from that | 16:02 |
| thorst | cause of obvious reasons | 16:02 |
| efried | Yeah, I would have to find that other blueprint. | 16:02 |
| efried | What's another platform that doesn't have the hypervisor owning all the hardware? | 16:02 |
| *** k0da has quit IRC | 16:03 | |
| thorst | well, isn't it more, what's another platform that doesn't have linux as the hypervisor/ | 16:03 |
| efried | sure, that would work. | 16:03 |
| thorst | VMware & hyper-v | 16:03 |
| efried | I read an actual blueprint from an actual hypervisor that wanted to implement SR-IOV something-or-other, but they had to punt to ocata. | 16:03 |
| efried | It may have been VMWare. | 16:03 |
| efried | Not sure how I'm going to find that sucker again... | 16:03 |
| thorst | garyk may have it? | 16:04 |
| thorst | I think he's vmware | 16:04 |
| adreznec | Yeah | 16:06 |
| adreznec | It was vmware | 16:06 |
| adreznec | efried: I think it was https://review.openstack.org/#/c/317126/ | 16:07 |
| efried | adreznec, yup, that was the one, thanks. | 16:08 |
| efried | maybe | 16:09 |
| adreznec | It was the one I remembered | 16:09 |
| adreznec | There may be more | 16:09 |
| adreznec | Something under https://review.openstack.org/#/q/project:openstack/nova-specs+sr-iov | 16:10 |
| thorst | esberglu: the CI under cloud console is very slow to respond... | 16:10 |
| *** efried has quit IRC | 16:11 | |
| *** jwcroppe has joined #openstack-powervm | 17:04 | |
| *** openstackgerrit_ has joined #openstack-powervm | 17:06 | |
| *** openstackgerrit_ has quit IRC | 17:08 | |
| thorst | well...unexpected win. Had a compute node die. Restacked just the NovaLink. Everything is fine...no VMs lost, all managable. | 17:08 |
| thorst | makes complete sense...just so used to AIO | 17:08 |
| *** madhaviy has quit IRC | 17:40 | |
| *** apearson has joined #openstack-powervm | 17:48 | |
| openstackgerrit | Sridhar Venkat proposed openstack/networking-powervm: WIP: Mechanism driver for powervm SR-IOV interface https://review.openstack.org/343423 | 18:38 |
| *** apearson has quit IRC | 18:43 | |
| *** kotra03 has quit IRC | 19:15 | |
| *** k0da has joined #openstack-powervm | 19:22 | |
| *** efried has joined #openstack-powervm | 19:29 | |
| *** tblakeslee is now known as tblakes | 19:29 | |
| thorst | efried: any priority reviews I need to do that're holding you up? | 19:50 |
| efried | thorst, think I'm good for the moment. | 19:50 |
| thorst | phew | 19:50 |
| efried | You got something I should focus on next? | 19:50 |
| thorst | efried: want to take over a change set? | 19:51 |
| efried | Depends on the change set ;-) | 19:51 |
| thorst | I'm about to ship a doozy | 19:54 |
| thorst | WIP | 19:54 |
| thorst | you'll love it | 19:54 |
| openstackgerrit | Drew Thorstensen proposed openstack/nova-powervm: [WIP] Plug updated VIFs https://review.openstack.org/350768 | 20:02 |
| thorst | efried: ^^ | 20:02 |
| *** kriskend_ has quit IRC | 20:05 | |
| *** kriskend has quit IRC | 20:05 | |
| thorst | efried: take a peak. See if you want to pick up | 20:07 |
| thorst | I'd give you an e-high five | 20:07 |
| thorst | I have to try it out...one of the benefits of having a system failure is seeing the reboot scenarios | 20:08 |
| efried | thorst, Fish outta water there, probably take you less time to finish it yourself than it would to hold my hand while I figure it out, but I'll take it if you want. | 20:08 |
| thorst | efried: well right now I'm a bear lost in a swamp | 20:08 |
| thorst | (favorite line from a TV show) | 20:08 |
| efried | Better than a pope sh*tting in the woods. | 20:09 |
| efried | Or a little worm on a big f'ing hook (favorite line from a movie, 20 points if you can identify it) | 20:09 |
| thorst | lol | 20:09 |
| efried | (-20 for using google) | 20:10 |
| thorst | heh | 20:10 |
| thorst | can't identify | 20:10 |
| efried | The Crow. | 20:10 |
| efried | And I don't know if it's my *favorite* line, but it's a good one. | 20:10 |
| thorst | I'm going to try a live test of the change now...just to validate my assumptions... | 20:16 |
| thorst | I can also have madhavi take a peak if you've got another thing to chase...it is for the OVS / SDN stuff | 20:17 |
| efried | thorst, you need to add new_vif=False to the new plug call, yes? | 20:22 |
| thorst | yeah, I missed a couple things | 20:22 |
| efried | ...and finish the comment on line 124? | 20:22 |
| efried | thorst, so you need me for what, UT? | 20:24 |
| openstackgerrit | Drew Thorstensen proposed openstack/nova-powervm: [WIP] Plug updated VIFs https://review.openstack.org/350768 | 20:24 |
| thorst | uhhh...polish. UT. | 20:24 |
| thorst | after I get a live test of the bugger. | 20:24 |
| openstackgerrit | Tyler Blakeslee proposed openstack/nova-powervm: WIP https://review.openstack.org/350773 | 20:24 |
| efried | thorst, where does this "treat" thing happen? | 20:26 |
| thorst | efried: an example... vif.py at line 494 | 20:26 |
| efried | oic | 20:27 |
| efried | thorst, so line 107 needs new_vif=new_vif ? | 20:28 |
| thorst | efried: yep | 20:28 |
| thorst | just saw that | 20:29 |
| efried | k | 20:29 |
| efried | Well, once you have the code, you can huck it over the fence and I can UT it. | 20:29 |
| thorst | thx...this system has a few 'issues' | 20:29 |
| thorst | every time I get to the LPM test...something happens :-/ | 20:29 |
| *** smatzek has quit IRC | 20:49 | |
| *** lmtaylor has left #openstack-powervm | 20:55 | |
| *** jwcroppe has quit IRC | 20:58 | |
| efried | thorst, svenkat https://review.openstack.org/#/c/343423/7/networking_powervm/plugins/ibm/agent/powervm/sriov_agent.py@66 | 21:03 |
| efried | Got a moment to discuss? | 21:03 |
| svenkat | efried sure… | 21:04 |
| thorst | name to name | 21:04 |
| efried | svenkat, could you make use of the pport loc codes? Presumably these mappings are accessible to the plug method? | 21:05 |
| svenkat | but this data has to go to mechaism driver… and there they have to be matched with neutron network object data.. ( one ofthemis physcal network. thats why i am puling port labels and includeing them in bridge mapping) | 21:06 |
| svenkat | neutron network object does not have location codes… | 21:06 |
| *** seroyer has quit IRC | 21:06 | |
| efried | The keys would still be the net names. | 21:07 |
| svenkat | ok. i can do that | 21:07 |
| efried | The values would be the physloc codes, as a list. | 21:07 |
| thorst | why not just 'default:default' | 21:07 |
| svenkat | but what is the prupose of location code in bridge mapping | 21:07 |
| efried | That would save you the entire chunk from line 500-511 in https://review.openstack.org/#/c/343419/5/nova_powervm/virt/powervm/vif.py | 21:07 |
| efried | ...assuming you could get at that data from plug. | 21:08 |
| svenkat | oh ok… | 21:08 |
| efried | thorst ^^ | 21:08 |
| svenkat | i can see if mech driver can push that data into vif | 21:08 |
| svenkat | like vlan id | 21:08 |
| svenkat | under vif details, i can put whatever | 21:08 |
| thorst | yep | 21:08 |
| thorst | dig it...however, remember | 21:08 |
| svenkat | so i thiik it is possible to put loc codes in vif... | 21:09 |
| thorst | the mechanism driver runs on the neutron server. | 21:09 |
| svenkat | yes.. | 21:09 |
| thorst | not necessarily on the compute node | 21:09 |
| thorst | so we'd have to have the mechanism driver get it from the agent | 21:09 |
| thorst | which I think can be done | 21:09 |
| svenkat | yes. bridge mapings flow from agent to server and mech driver to vif is between neutron server to nova compute… | 21:09 |
| efried | Pretty sure that's how it *is* done. | 21:09 |
| efried | yeah, what he said. | 21:09 |
| svenkat | yes. just like we push vlan id to vif from mech driver | 21:09 |
| svenkat | good point efried. i wilil give it a shot | 21:10 |
| efried | Cool beans. | 21:10 |
| thorst | seems like that gets rid of all the calls in the vif to neutron | 21:10 |
| svenkat | i thought we removed all that already… no more neutron calls in vif | 21:10 |
| efried | what, vif.get_physical_network()? | 21:10 |
| svenkat | that is also flowing from mech driver to vif | 21:11 |
| svenkat | to get physical network, vif need not have to go to neutron | 21:11 |
| svenkat | with loc code in vif suggestion, we will save using pypowervm to get port labels. instead wewill get them from vif directly | 21:12 |
| svenkat | more on get_physical_network it i smade possible by the way port is created | 21:12 |
| svenkat | neutron port-create --debug --vnic-type direct e23ef566-2942-47fc-aaf6-35b905904767 -- --binding:profile type=dict physical_network=default | 21:12 |
| svenkat | binding profile physical network dictionary item in port makes mech driver to push it to vif | 21:12 |
| svenkat | if you create a port without binding profile, then it does not work | 21:13 |
| *** seroyer has joined #openstack-powervm | 21:14 | |
| svenkat | i have to run … i will be back tonight. | 21:14 |
| *** svenkat has quit IRC | 21:15 | |
| openstackgerrit | Tyler Blakeslee proposed openstack/nova-powervm: WIP https://review.openstack.org/350773 | 21:17 |
| *** seroyer has quit IRC | 21:26 | |
| *** jwcroppe has joined #openstack-powervm | 21:28 | |
| *** seroyer has joined #openstack-powervm | 21:34 | |
| *** thorst has quit IRC | 21:40 | |
| *** seroyer has quit IRC | 21:51 | |
| *** dwayne has quit IRC | 21:52 | |
| *** burgerk_ has joined #openstack-powervm | 21:53 | |
| *** burgerk has quit IRC | 21:56 | |
| *** jwcroppe_ has joined #openstack-powervm | 21:56 | |
| *** jwcroppe has quit IRC | 21:58 | |
| *** burgerk_ has quit IRC | 21:58 | |
| *** esberglu has quit IRC | 22:00 | |
| *** thorst has joined #openstack-powervm | 22:03 | |
| *** mdrabe_ has quit IRC | 22:07 | |
| *** thorst has quit IRC | 22:07 | |
| *** svenkat has joined #openstack-powervm | 22:11 | |
| *** jwcroppe has joined #openstack-powervm | 22:14 | |
| *** jwcroppe_ has quit IRC | 22:17 | |
| *** svenkat has quit IRC | 22:18 | |
| *** Ashana has quit IRC | 22:20 | |
| *** Ashana has joined #openstack-powervm | 22:21 | |
| *** Ashana has quit IRC | 22:25 | |
| *** Ashana has joined #openstack-powervm | 22:27 | |
| *** Ashana has quit IRC | 22:31 | |
| *** k0da has quit IRC | 22:32 | |
| *** Ashana has joined #openstack-powervm | 22:33 | |
| openstackgerrit | Tyler Blakeslee proposed openstack/nova-powervm: WIP https://review.openstack.org/350773 | 22:37 |
| *** Ashana has quit IRC | 22:37 | |
| *** Ashana has joined #openstack-powervm | 22:39 | |
| *** Ashana has quit IRC | 22:43 | |
| *** efried has quit IRC | 22:43 | |
| *** tjakobs has quit IRC | 22:54 | |
| *** tblakes has quit IRC | 23:12 | |
| *** tblakes has joined #openstack-powervm | 23:40 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!