| *** zhenzanz has joined #openstack-ironic | 00:05 | |
| *** zhenzanz_ has joined #openstack-ironic | 00:06 | |
| *** achanda has quit IRC | 00:08 | |
| *** zhenzanz has quit IRC | 00:09 | |
| *** zhenzanz_ is now known as zhenzanz | 00:09 | |
| *** naohirot has joined #openstack-ironic | 00:14 | |
| naohirot | good morning ironic! | 00:14 |
|---|---|---|
| NobodyCam | morning naohirot | 00:17 |
| Haomeng | naohirot: morning:) | 00:18 |
| *** ChuckC_ has quit IRC | 00:18 | |
| naohirot | NobodyCam: hi , thanks for review | 00:19 |
| NobodyCam | :-p | 00:19 |
| naohirot | Haomeng: good morning | 00:20 |
| Haomeng | naohirot: :) | 00:20 |
| naohirot | Haomeng: how are you today? :-) | 00:20 |
| Haomeng | naohirot: fine, and you? | 00:20 |
| naohirot | Haomeng: I'm fine too, thanks! | 00:21 |
| Haomeng | naohirot: I think you are from Japen, and your time is 7am, so early:) | 00:21 |
| naohirot | Haomeng: yes, I'm in Japan, 9am right now | 00:22 |
| Haomeng | naohirot: ok, you are early than China 1 hour:) | 00:22 |
| Haomeng | naohirot: :) | 00:22 |
| naohirot | Haomeng: Yeah, just 1 hour diff :-) | 00:23 |
| Haomeng | naohirot: yes:) | 00:23 |
| *** ChuckC_ has joined #openstack-ironic | 00:25 | |
| *** achanda has joined #openstack-ironic | 00:26 | |
| *** achanda has quit IRC | 00:37 | |
| *** davideagnello has quit IRC | 00:50 | |
| *** alexpilotti has quit IRC | 00:51 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: WIP: Exposing Hardware Capabilities https://review.openstack.org/131272 | 00:52 |
| *** NobodyCam has quit IRC | 00:56 | |
| *** achanda has joined #openstack-ironic | 00:57 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Exposing Hardware Capabilities https://review.openstack.org/131272 | 00:58 |
| *** NobodyCam has joined #openstack-ironic | 00:58 | |
| *** spandhe has quit IRC | 00:58 | |
| JayF | I modified https://review.openstack.org/#/c/131272 to be a backlog spec, if someone wants to pick it up and lay out an implementation, feel free, but I wanted to get the spec to a semi-mergable state rather than having it sit around rotting | 00:59 |
| *** hemna__ has quit IRC | 01:03 | |
| *** davideagnello has joined #openstack-ironic | 01:05 | |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Exposing Hardware Capabilities https://review.openstack.org/131272 | 01:07 |
| openstackgerrit | Jay Faulkner proposed openstack/ironic-specs: Exposing Hardware Capabilities https://review.openstack.org/131272 | 01:11 |
| JayF | IDK what happened to my vim setting that would highlight trailing whitespace in red | 01:12 |
| JayF | but losing it has been helping my "total number of patchsets" stat | 01:12 |
| *** Haomeng|2 has joined #openstack-ironic | 01:18 | |
| *** Haomeng has quit IRC | 01:19 | |
| *** achanda has quit IRC | 01:22 | |
| *** dlaube has quit IRC | 01:24 | |
| *** kfox1111 has quit IRC | 01:25 | |
| *** kfox1111 has joined #openstack-ironic | 01:30 | |
| *** achanda has joined #openstack-ironic | 01:31 | |
| *** r-daneel has quit IRC | 01:32 | |
| *** achanda has quit IRC | 01:35 | |
| *** kfox1111 has quit IRC | 01:37 | |
| JayF | https://review.openstack.org/#/c/134436/5 has been updated to make agent_ssh-src job voting on IPA only, and not make anything new vote for Ironic | 01:37 |
| JayF | tl;dr: I'd rather have teh agent vote on itself, even if Ironic can merge breaking changes for it still | 01:38 |
| *** chenglch has joined #openstack-ironic | 01:40 | |
| *** nosnos has joined #openstack-ironic | 02:14 | |
| *** rloo has quit IRC | 02:22 | |
| *** NobodyCam has quit IRC | 02:26 | |
| *** NobodyCam has joined #openstack-ironic | 02:27 | |
| *** vipul has quit IRC | 02:33 | |
| *** vipul has joined #openstack-ironic | 02:37 | |
| *** yuanying_ has quit IRC | 02:47 | |
| *** nosnos has quit IRC | 03:01 | |
| *** Marga_ has quit IRC | 03:03 | |
| *** Marga_ has joined #openstack-ironic | 03:04 | |
| *** yuanying has joined #openstack-ironic | 03:04 | |
| *** ryanpetrello has joined #openstack-ironic | 03:18 | |
| *** ramineni has joined #openstack-ironic | 03:22 | |
| *** killer_prince has quit IRC | 03:23 | |
| *** Marga_ has quit IRC | 03:24 | |
| *** harlowja is now known as harlowja_away | 03:29 | |
| *** naohirot has quit IRC | 03:30 | |
| *** ryanpetrello has quit IRC | 03:41 | |
| *** pensu has joined #openstack-ironic | 03:58 | |
| *** naohirot has joined #openstack-ironic | 04:01 | |
| *** ryanpetrello has joined #openstack-ironic | 04:07 | |
| *** achanda has joined #openstack-ironic | 04:12 | |
| *** yuanying_ has joined #openstack-ironic | 04:14 | |
| *** nosnos has joined #openstack-ironic | 04:16 | |
| *** yuanying has quit IRC | 04:17 | |
| *** killer_prince has joined #openstack-ironic | 04:18 | |
| *** killer_prince is now known as lazy_prince | 04:18 | |
| *** ryanpetrello has quit IRC | 04:23 | |
| *** Haomeng has joined #openstack-ironic | 04:30 | |
| *** Haomeng|2 has quit IRC | 04:30 | |
| *** achanda has quit IRC | 04:30 | |
| *** Marga_ has joined #openstack-ironic | 04:35 | |
| *** Marga_ has quit IRC | 04:39 | |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic-specs: Ironic Management Interfaces to support UEFI Secure Boot https://review.openstack.org/135845 | 04:46 |
| *** pcrews has quit IRC | 04:48 | |
| *** ryanpetrello has joined #openstack-ironic | 04:49 | |
| *** yongli has joined #openstack-ironic | 04:51 | |
| *** pensu has quit IRC | 05:04 | |
| openstackgerrit | Anusha Ramineni proposed openstack/ironic: iLO Management Interface https://review.openstack.org/132746 | 05:09 |
| *** rushiagr_away is now known as rushiagr | 05:10 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic-specs: iRMC Power Driver for Ironic https://review.openstack.org/134487 | 05:14 |
| *** Haomeng|2 has joined #openstack-ironic | 05:14 | |
| *** Haomeng has quit IRC | 05:15 | |
| *** rakesh_hs has joined #openstack-ironic | 05:27 | |
| *** sirushti has joined #openstack-ironic | 05:29 | |
| openstackgerrit | Michael Davies proposed openstack/ironic-specs: Proposal to add logical names to Ironic nodes https://review.openstack.org/134439 | 05:31 |
| *** rameshg87 has joined #openstack-ironic | 05:33 | |
| *** ujuc has joined #openstack-ironic | 05:39 | |
| *** pensu has joined #openstack-ironic | 05:48 | |
| *** rushiagr is now known as rushiagr_away | 05:59 | |
| *** k4n0 has joined #openstack-ironic | 06:03 | |
| *** ryanpetrello has quit IRC | 06:05 | |
| *** Masahiro has joined #openstack-ironic | 06:19 | |
| *** Masahiro has quit IRC | 06:26 | |
| *** Masahiro has joined #openstack-ironic | 06:27 | |
| *** Masahiro has quit IRC | 06:29 | |
| *** Masahiro has joined #openstack-ironic | 06:29 | |
| *** mrda is now known as mrda-away | 06:37 | |
| *** pcrews has joined #openstack-ironic | 06:48 | |
| *** pcrews has quit IRC | 06:53 | |
| *** Masahiro has quit IRC | 07:02 | |
| *** Masahiro has joined #openstack-ironic | 07:03 | |
| *** Masahiro has quit IRC | 07:05 | |
| *** rushiagr_away is now known as rushiagr | 07:07 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic-specs: Enable iscsi_ilo driver to share boot images https://review.openstack.org/137291 | 07:19 |
| *** Haomeng|2 has quit IRC | 07:21 | |
| *** pcrews has joined #openstack-ironic | 07:23 | |
| *** Masahiro has joined #openstack-ironic | 07:25 | |
| *** ujuc has quit IRC | 07:30 | |
| *** achanda has joined #openstack-ironic | 07:31 | |
| *** pcrews has quit IRC | 07:31 | |
| *** achanda has quit IRC | 07:36 | |
| *** Masahiro has quit IRC | 07:36 | |
| *** penick has joined #openstack-ironic | 07:38 | |
| *** mitz- has joined #openstack-ironic | 07:51 | |
| *** ryanpetrello has joined #openstack-ironic | 07:51 | |
| *** mitz_ has quit IRC | 07:51 | |
| *** ryanpetrello has quit IRC | 07:56 | |
| *** athomas has joined #openstack-ironic | 08:01 | |
| *** affo has joined #openstack-ironic | 08:10 | |
| affo | hi everybody! my problem is that I cannot understand what ironic is! What is bare metal provisioning? Why would I need it? If it can take you only few words, please, explain me!:) | 08:12 |
| *** dtantsur|afk is now known as dtantsur | 08:13 | |
| dtantsur | Morning Ironic | 08:14 |
| *** lucasagomes has joined #openstack-ironic | 08:14 | |
| *** penick has quit IRC | 08:14 | |
| GheRivero | morning all | 08:15 |
| *** ifarkas has joined #openstack-ironic | 08:16 | |
| *** Masahiro has joined #openstack-ironic | 08:18 | |
| lucasagomes | GheRivero, morning | 08:19 |
| lucasagomes | affo, deployment of physical servers instead of VMs, containers etc... | 08:20 |
| *** ifarkas has quit IRC | 08:28 | |
| *** ifarkas has joined #openstack-ironic | 08:28 | |
| *** jcoufal has joined #openstack-ironic | 08:29 | |
| affo | lucasagomes: thank you! But physical in the sense of physical or virtual? I mean, I have a physical machine and I want to deploy on it ubuntu with nova-compunte and neutron and, given the IP address, Ironic does everything for me? | 08:33 |
| *** achanda has joined #openstack-ironic | 08:33 | |
| *** mitz_ has joined #openstack-ironic | 08:34 | |
| lucasagomes | affo, physical in the sense of physical. if ur machine can be controlled remotely by some sort of management card (BMC) yes you could do that. You can build an imagine with nova-compute and deploy it via Ironic | 08:34 |
| *** andreykurilin_ has joined #openstack-ironic | 08:35 | |
| *** foexle has joined #openstack-ironic | 08:35 | |
| affo | ooook, now it is much clearer! Can it be used also to deploy vms to simulate them as physical machines or it is a silly thing? | 08:35 |
| affo | lucasagomes: ooook, now it is much clearer! Can it be used also to deploy vms to simulate them as physical machines or it is a silly thing? | 08:35 |
| *** mitz- has quit IRC | 08:36 | |
| lucasagomes | affo, we have a driver (SSH driver) that can deploy VMs but this is only used for tests in gate because we don't have access to a physical server there | 08:37 |
| lucasagomes | so we mock the BMC calls by issuing virsh commands via SSH | 08:37 |
| lucasagomes | so I would say that deploying VMs for production with ironic is silly | 08:37 |
| lucasagomes | it's a test thing | 08:37 |
| affo | lucasagomes: because I need to deploy a lot of VMs for stress-testing... and I do not know how to do it | 08:40 |
| pensu | Hi all, I have been trying to deploy a physical node using devstack ironic setup, I have set neutron according to this guide: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#configure-neutron-to-communicate-with-the-bare-metal-server | 08:42 |
| pensu | Now, when I try to boot up, my IPMI works completely fine,but I am not able to boot the server.... | 08:43 |
| pensu | I checked neutron and nova logs, there is no error.... | 08:43 |
| pensu | any idea what could I be doing wrong? | 08:43 |
| rameshg87 | pensu, is your node not pxe booting ? | 08:46 |
| pensu | rameshg87: It is....earlier I used to get the IP on the node, n/w boot and all, then one fine day I rebooted my VM and since then it's not working! | 08:48 |
| *** achanda has quit IRC | 08:48 | |
| *** romcheg has joined #openstack-ironic | 08:48 | |
| pensu | rameshg87: The issue is noone is throwing any error, nova-compute just keeps waiting and then errors out.... | 08:48 |
| rameshg87 | pensu, is it able to load the deploy kernel after pxe boot ? are you able to watch the bare metal node's console ? | 08:51 |
| rameshg87 | pensu, if it remains in wait-call back state, the deploy kernel/ramdisk never booted completely successfully on the bare metal node | 08:51 |
| pensu | rameshg87: No, that's not happening, All I can see is my node is in power off state, as I initiate boot process, it is powered on. yes, I can see the console, I get the petitboot console after power on and that's it..... | 08:53 |
| pensu | rameshg87: earlier I used to get a netboot option on the connected interface, but now, not sure why, that option is not coming.... | 08:54 |
| rameshg87 | pensu, what's petitboot console ? isn't your bare metal node attempting to boot from pxe ? | 08:54 |
| pensu | rameshg87: yeah, but there is something already on it, we use petitboot in our servers.....so when I see the console, the server boots up, gives me access to petitboot, after that nothing.... | 08:56 |
| *** andreykurilin_ has quit IRC | 08:58 | |
| *** andreykurilin_ has joined #openstack-ironic | 08:59 | |
| *** jistr has joined #openstack-ironic | 08:59 | |
| rameshg87 | pensu, are you using pxe_ipmitool driver ? if so, pxe_ipmitool driver sets the bare metal to boot from pxe. | 08:59 |
| *** zhenzanz has quit IRC | 08:59 | |
| rameshg87 | pensu, so your node should attempt to boot from pxe | 08:59 |
| rameshg87 | pensu, you might be able to see some error on conductor log if ipmi setting of boot device failed | 09:00 |
| rameshg87 | pensu, otherwise it might be something to do with the bare metal node | 09:01 |
| *** naohirot has quit IRC | 09:01 | |
| *** affo has quit IRC | 09:02 | |
| pensu | rameshg87: yup, I am using pxe_ipmitool only, and there is no error with conductor logs..... | 09:02 |
| *** derekh has joined #openstack-ironic | 09:12 | |
| *** Masahiro has quit IRC | 09:13 | |
| romcheg | Hi Ironic! | 09:16 |
| lucasagomes | romcheg, morning | 09:20 |
| *** Masahiro has joined #openstack-ironic | 09:23 | |
| *** dlpartain has joined #openstack-ironic | 09:27 | |
| romcheg | lucasagomes: Morning | 09:30 |
| romcheg | could you please take a look on the policy patch https://review.openstack.org/#/c/126265/ ? | 09:30 |
| romcheg | it's been almost a week w/o any feedback there :) | 09:31 |
| *** dlpartain has quit IRC | 09:31 | |
| lucasagomes | romcheg, I'm back in paris for a meeting | 09:46 |
| lucasagomes | I will take a look when I get some time | 09:46 |
| lucasagomes | :( | 09:46 |
| *** lazy_prince has quit IRC | 09:48 | |
| *** andreykurilin_ has quit IRC | 09:52 | |
| *** jcoufal_ has joined #openstack-ironic | 09:57 | |
| *** jcoufal has quit IRC | 10:00 | |
| *** sambetts has joined #openstack-ironic | 10:02 | |
| *** subscope has quit IRC | 10:08 | |
| pensu | Hi all, while booting a node through Ironic I am getting this error: | 10:19 |
| pensu | network ready | 10:20 |
| pensu | iscsi_target_iqn is not defined | 10:20 |
| pensu | Troubleshooting required, press t to launch shell........... | 10:20 |
| pensu | Could not find disk to use. | 10:20 |
| pensu | Troubleshooting required, press t to launch shell........... | 10:20 |
| pensu | start iSCSI target on | 10:20 |
| pensu | waiting for tgtd socket...not found | 10:20 |
| pensu | waiting for tgtd socket...found | 10:20 |
| pensu | tftp: timeout | 10:20 |
| pensu | tftp: invalid option -- 'c' | 10:20 |
| pensu | BusyBox v1.19.4 (2014-07-23 21:28:14 UTC) multi-call binary. | 10:20 |
| pensu | Usage: tftp [OPTIONS] HOST [PORT] | 10:20 |
| pensu | Transfer a file from/to tftp server | 10:20 |
| pensu | -l FILELocal FILE | 10:20 |
| pensu | -r FILERemote FILE | 10:20 |
| pensu | -gGet file | 10:20 |
| *** ndipanov_gone is now known as ndipanov | 10:20 | |
| *** Masahiro has quit IRC | 10:25 | |
| *** andreykurilin_ has joined #openstack-ironic | 10:27 | |
| *** dlpartain has joined #openstack-ironic | 10:30 | |
| *** subscope has joined #openstack-ironic | 10:34 | |
| openstackgerrit | Imre Farkas proposed openstack/ironic-specs: New driver interface for RAID configuration https://review.openstack.org/135899 | 10:44 |
| *** killer_prince has joined #openstack-ironic | 10:56 | |
| *** killer_prince is now known as lazy_prince | 10:56 | |
| *** MattMan has quit IRC | 11:00 | |
| *** MattMan has joined #openstack-ironic | 11:00 | |
| *** ramineni has quit IRC | 11:03 | |
| *** BertieFulton has joined #openstack-ironic | 11:09 | |
| *** chenglch has quit IRC | 11:10 | |
| openstackgerrit | Yuriy Zveryanskyy proposed openstack/ironic: Change methods from classmethod to staticmethod https://review.openstack.org/137335 | 11:11 |
| *** rameshg87 has quit IRC | 11:21 | |
| *** Masahiro has joined #openstack-ironic | 11:26 | |
| *** Masahiro has quit IRC | 11:30 | |
| GheRivero | lucasagomes: ping | 11:34 |
| openstackgerrit | Vladyslav Drok proposed openstack/python-ironicclient: Fix log_curl_request API version duplication https://review.openstack.org/136773 | 11:39 |
| *** pensu has quit IRC | 11:41 | |
| *** naohirot has joined #openstack-ironic | 11:43 | |
| *** viktors|afk is now known as viktors | 11:52 | |
| *** rakesh_hs2 has joined #openstack-ironic | 12:12 | |
| *** rakesh_hs has quit IRC | 12:12 | |
| lucasagomes | GheRivero, pong | 12:31 |
| GheRivero | lucasagomes: about your olso.config patch. I was looking about putting the list_opts per file (instead of opts.py) but apparently that way, we will have to populate setup.cfg with allthe files. AM I right? | 12:36 |
| lucasagomes | GheRivero, yeah, that's why I was like hmm | 12:37 |
| lucasagomes | GheRivero, I don't know a better way to do as well.. in both cases it seems that you need to maintain a list of configs somewhere | 12:37 |
| lucasagomes | :/ | 12:37 |
| * lucasagomes will ask mark | 12:38 | |
| GheRivero | I revisited other projects and other oslo.* libraries and everyone is doing it that way (everything in opts.py) | 12:38 |
| GheRivero | don;t like either to have to maintain a list, but I prefer it in opts.py than in setup.cfg | 12:38 |
| lucasagomes | yeah :/ | 12:39 |
| GheRivero | :/ | 12:39 |
| lucasagomes | idk if I like it or not | 12:39 |
| lucasagomes | I mean, I don't like but I feel the necessity of syncing with oslo | 12:39 |
| GheRivero | Having to maintain a list manually is like... agh | 12:41 |
| lucasagomes | indeed | 12:41 |
| lucasagomes | GheRivero, another way I thought about doing it would be to having all the configurations to be created in one file in ironic | 12:41 |
| lucasagomes | instead of creating the configuration options and groups on various of files | 12:41 |
| lucasagomes | it would be centralized | 12:41 |
| lucasagomes | so that, we could have only one entry on the setup.cfg and 1 function in the same file that would list all the config options | 12:42 |
| lucasagomes | It sounds okish to me, but requires a more broad refactor | 12:42 |
| GheRivero | I've seen that in other project. can't remember which one | 12:42 |
| lucasagomes | right, but u think this approach would be better? | 12:43 |
| GheRivero | better don't know. Simple, sure | 12:43 |
| lucasagomes | right simple is good :) | 12:43 |
| GheRivero | It will have less errors | 12:44 |
| lucasagomes | yeah | 12:44 |
| GheRivero | are you in Paris the whole week? | 12:45 |
| lucasagomes | yeah, I will be back on firday | 12:45 |
| lucasagomes | friday* | 12:45 |
| lucasagomes | GheRivero, for that oslo.config work, u want to take over that patch? | 12:45 |
| *** rushiagr is now known as rushiagr_away | 12:46 | |
| GheRivero | I have some free time this week so yeah, I can do something | 12:46 |
| lucasagomes | thanks | 12:46 |
| *** jcoufal_ has quit IRC | 12:47 | |
| *** jcoufal has joined #openstack-ironic | 12:48 | |
| *** Haomeng has joined #openstack-ironic | 12:50 | |
| lucasagomes | GheRivero, seems that confs.py is the way to go | 12:59 |
| lucasagomes | GheRivero, also take a look at https://github.com/markmc/nova/blob/config-generator/nova/config.py | 12:59 |
| *** Masahiro has joined #openstack-ironic | 13:14 | |
| GheRivero | lucasagomes: thanks for the info | 13:16 |
| lucasagomes | np problem | 13:16 |
| *** Masahiro has quit IRC | 13:19 | |
| *** nosnos has quit IRC | 13:20 | |
| BertieFulton | lucasagomes: Are you there and may I discuss something about the state machine with you? | 13:21 |
| lucasagomes | BertieFulton, hi, I'm here but i'm in a meeting so I won't be able to pay full attention to it | 13:21 |
| *** alexpilotti has joined #openstack-ironic | 13:22 | |
| BertieFulton | lucasagomes: no worries - I'm happy to wait on replies if you don't mind being bugged :) | 13:22 |
| lucasagomes | BertieFulton, but anyway, whats up ? | 13:22 |
| BertieFulton | so we're in full agreement here with you on your comment on states and actions (I've been discussing it this morning) | 13:23 |
| BertieFulton | on your proposed change - would you see ZAPPING as an action that could happen in different states then? | 13:23 |
| BertieFulton | e.g. one use case we might have is that a DEPLOYED node would require a firmware change for a particular workload | 13:24 |
| lucasagomes | BertieFulton, this can be a problem. I think we are missing the context of what is ZAPPING | 13:25 |
| lucasagomes | ZAPPING as far as I understand is about cleaning the node | 13:25 |
| lucasagomes | removing the data from the previous tenant (or data that already exist in the node if it's recyclable) | 13:25 |
| lucasagomes | so this is also powerful to not mix many things in one state | 13:25 |
| lucasagomes | we need to have very well defined states, therefor I don't see ZAPPING coming from different states | 13:26 |
| BertieFulton | +1 | 13:26 |
| lucasagomes | because each state may have a well defined task | 13:26 |
| BertieFulton | my confusion lies in that it is described as a catch-all for long running tasks | 13:26 |
| lucasagomes | let's say u want to update/flash BIOS firmware | 13:26 |
| lucasagomes | so this is diff than zapping we could have another state for that | 13:26 |
| BertieFulton | :) | 13:26 |
| lucasagomes | I will call it normalised | 13:26 |
| lucasagomes | just as an example | 13:27 |
| lucasagomes | so I would say it should go like | 13:27 |
| lucasagomes | INIT -> [NORMALIZING] -> NORMALIZED -> [ ZAPPPING ] -> AVAILABLE | 13:27 |
| lucasagomes | something like that | 13:27 |
| lucasagomes | so that way it's very predicable, if something fails when the state is NORMALIZED and transition to AVAILABLE, u know that it failed at the ZAPPING action | 13:28 |
| lucasagomes | it's easy to debug, and easy to understand the state machine | 13:28 |
| BertieFulton | agreed | 13:28 |
| lucasagomes | BertieFulton, sounds correct? | 13:28 |
| lucasagomes | ok | 13:29 |
| BertieFulton | most definitely | 13:29 |
| BertieFulton | So if we have, for example, a firmware update to do when DEPLOYED | 13:30 |
| lucasagomes | BertieFulton, what was the output of ur discussion this morning? | 13:30 |
| BertieFulton | I thought that actions (I was calling them tasks) and states were being mixed | 13:30 |
| BertieFulton | and that there were long running tasks that we would want to perform that would not fit into the ZAPPING state | 13:31 |
| BertieFulton | as described by INIT-> ZAPPING -> AVAILABLE | 13:31 |
| lucasagomes | yeah | 13:31 |
| BertieFulton | and that ZAPPING wasn't really a state | 13:31 |
| lucasagomes | yeah that confuses me too | 13:31 |
| BertieFulton | your comment on the review made my mood much better :) | 13:32 |
| *** alexpilotti has quit IRC | 13:32 | |
| lucasagomes | good to hear :) | 13:32 |
| *** erwan_taf has joined #openstack-ironic | 13:33 | |
| BertieFulton | so perhaps the thing to do is for me to get together with our guys and come up with a proposal - the only issue is that we would like to see the outcome of your suggestion first as we would rather work in that paradigm | 13:33 |
| erwan_taf | jumping in | 13:33 |
| BertieFulton | or maybe I should try to discuss later today when devananda, NobodyCam etc are online? | 13:34 |
| lucasagomes | BertieFulton, sure I'm planning to propose something but I have no time this week (I'm travelling for work) | 13:40 |
| lucasagomes | BertieFulton, both are important, it would be great to catch up with you and try to come up with something | 13:40 |
| lucasagomes | but also it would be good to talk to devananda NobodyCam etc... about the inital idea | 13:40 |
| *** ryanpetrello has joined #openstack-ironic | 13:42 | |
| *** rushiagr_away is now known as rushiagr | 13:42 | |
| *** rakesh_hs2 has quit IRC | 13:43 | |
| BertieFulton | lucasagomes: hopefully I'll get a chance to gauge opinion tonight (I have a 3 week old so I can guarantee nothing) and I'll contact you after that | 13:46 |
| BertieFulton | lucasagomes: Big thanks for your help | 13:46 |
| *** kfox1111 has joined #openstack-ironic | 13:53 | |
| lucasagomes | BertieFulton, no problem :) | 13:53 |
| lucasagomes | BertieFulton, yes please contact me, I'm more than happy to help with the state machine (it's critical to me to get it implemented asap) | 13:55 |
| BertieFulton | lucasagomes: Will do and thanks again | 13:57 |
| lucasagomes | yvw, thanks too | 13:58 |
| *** mjturek has joined #openstack-ironic | 14:04 | |
| openstackgerrit | Syed Ismail Faizan Barmawer proposed openstack/ironic-specs: Partial image support for agent driver https://review.openstack.org/137363 | 14:08 |
| *** Haomeng has quit IRC | 14:13 | |
| *** tchaypo has quit IRC | 14:15 | |
| *** Haomeng has joined #openstack-ironic | 14:16 | |
| *** tchaypo has joined #openstack-ironic | 14:17 | |
| NobodyCam | good morning Ironic | 14:24 |
| lucasagomes | NobodyCam, morning | 14:25 |
| NobodyCam | :) | 14:29 |
| NobodyCam | I see you had a great chat with BertieFulton :) | 14:29 |
| jroll | morning everybody :) | 14:31 |
| NobodyCam | good morning jroll :) | 14:31 |
| *** jgrimm is now known as zz_jgrimm | 14:32 | |
| BertieFulton | Morning NobodyCam | 14:32 |
| *** rloo has joined #openstack-ironic | 14:32 | |
| NobodyCam | hey hey BertieFulton your still here... Good morning :) | 14:32 |
| *** romcheg has quit IRC | 14:33 | |
| *** romcheg has joined #openstack-ironic | 14:33 | |
| NobodyCam | anyone with turkey day plans? | 14:35 |
| *** Haomeng|2 has joined #openstack-ironic | 14:35 | |
| BertieFulton | # NobodyCam: I am indeed, sorry I missed the summit and meeting you | 14:36 |
| *** Haomeng has quit IRC | 14:37 | |
| *** r-daneel has joined #openstack-ironic | 14:37 | |
| jroll | NobodyCam: my plans are all the food, all the beer | 14:40 |
| NobodyCam | not at all | 14:40 |
| jroll | :) | 14:40 |
| *** erwan_taf has quit IRC | 14:40 | |
| NobodyCam | jroll: LOL | 14:40 |
| NobodyCam | :) | 14:40 |
| NobodyCam | BertieFulton: we will again have a chance to meet :) | 14:41 |
| naohirot | NobodyCam: jroll: BertieFulton: Happy thanks giving :-) | 14:45 |
| jroll | naohirot: :) | 14:45 |
| NobodyCam | morning naohirot :) | 14:46 |
| NobodyCam | and thank you | 14:46 |
| naohirot | jroll: I have a present for you, actually a basic question | 14:46 |
| jroll | hehe | 14:46 |
| jroll | what's up? | 14:46 |
| naohirot | jroll: http://docs.openstack.org/developer/ironic/deploy/install-guide.html#image-requirements | 14:47 |
| naohirot | jroll: the second bullet "bin/disk-image-create -u ubuntu -o my-image" | 14:47 |
| naohirot | jroll: this creates User OS image | 14:47 |
| jroll | mhmmm | 14:48 |
| * jroll makes a note to do an IPA docs day real soon | 14:48 | |
| naohirot | jroll: in case of bare metal server, each server needs their own driver. | 14:49 |
| *** jjulien_ has joined #openstack-ironic | 14:49 | |
| naohirot | jroll: In case of Fujitsu PRIMERGY, NIC is Emulex CNA. | 14:49 |
| jroll | ah | 14:50 |
| jroll | naohirot: so you want to know how to add drivers to the user image? | 14:50 |
| naohirot | jroll: can disk-image-create create User OS for PRIMERGY, I believe not. | 14:50 |
| naohirot | jroll: that's right | 14:51 |
| jroll | it could :) | 14:51 |
| jroll | so, I am not a diskimage-builder expert, you may want to confirm with the #tripleo folks, but... | 14:51 |
| jroll | as I understand it, you can combine multiple elements | 14:51 |
| jroll | so you could make an element that installs the right drivers | 14:51 |
| jroll | and add that into the disk-image-create command | 14:52 |
| kfox1111 | Question. With the pxe driver, it looks like you need to set the node to always pxe boot. But the agent driver tries to image again if it reboots. | 14:52 |
| kfox1111 | does it need to be set to boot local by default? | 14:52 |
| jroll | naohirot: (I have no idea how to do that) | 14:52 |
| jroll | kfox1111: the agent driver should set "always boot from disk" itself | 14:52 |
| kfox1111 | jroll: Ok. but that doesn't seem to work with ironic on my hardware. | 14:53 |
| naohirot | jroll: I see, I'll check tripleo forks. | 14:53 |
| jroll | kfox1111: unless the persistent flag doesn't work on your bmc or whatever | 14:53 |
| kfox1111 | yeah. seems to have no affect. | 14:53 |
| lucasagomes | NobodyCam, yes :) please take a look at the state machine spec | 14:53 |
| lucasagomes | jroll, morning | 14:53 |
| kfox1111 | ok. I'll just set it back in the bios. thanks. | 14:53 |
| jroll | kfox1111: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/agent.py#L446 | 14:53 |
| naohirot | lucasagomes: hi | 14:53 |
| jroll | but yeah, bios could help you out | 14:53 |
| naohirot | kfox1111: hi | 14:53 |
| jroll | morning lucasagomes :) | 14:53 |
| lucasagomes | naohirot, hi there | 14:54 |
| kfox1111 | naohirot: hi. | 14:54 |
| *** erwan_taf has joined #openstack-ironic | 14:54 | |
| naohirot | lucasagomes: no thanks giving in EU, right? | 14:54 |
| lucasagomes | naohirot, heh nop :) | 14:54 |
| jroll | EU people get a day off from the US folks, that's enough reason to celebrate | 14:54 |
| NobodyCam | lol | 14:55 |
| lucasagomes | :D | 14:56 |
| NobodyCam | oh lucasagomes just a quick question: are you going to put out a "vote for name" email to the whole list? | 14:57 |
| rloo | hi naohirot, just one little thing (from me) in your spec https://review.openstack.org/#/c/134487/ | 14:58 |
| NobodyCam | morning rloo :) | 14:58 |
| naohirot | rloo: good morning | 14:58 |
| lucasagomes | NobodyCam, I will... I'm sorry this week is being mad here | 14:58 |
| lucasagomes | so I didn't have enough time, but I will do it | 14:58 |
| rloo | morning NobodyCam, jroll, kfox1111, lucasagomes and everyone else :-) | 14:58 |
| lucasagomes | rloo, morning :) | 14:59 |
| lucasagomes | how r u rloo ? | 14:59 |
| NobodyCam | nothing to be sorry about :) | 14:59 |
| jroll | hiya rloo :) | 14:59 |
| rloo | hey lucasagomes, I'm 'ok'. I've been sick since last Tues, it gets better then worse. Am wondering if I'm getting better or worse now :-( | 15:00 |
| NobodyCam | rloo: :( | 15:00 |
| * NobodyCam hopes rloo feels better | 15:01 | |
| naohirot | rloo: I checked your comment, you are right, I created [irmc] section, so irmc_ is no longer necessary, I'll fix it, thanks! | 15:01 |
| rloo | NobodyCam: yeah. I can't even blame it on JayF (Paris) ! | 15:01 |
| rloo | thx naohirot | 15:01 |
| NobodyCam | :-p | 15:01 |
| dtantsur | morning rloo, NobodyCam, jroll and everyone else :) | 15:01 |
| dtantsur | rloo, get better! | 15:01 |
| NobodyCam | morning dtantsur :) | 15:01 |
| rloo | lucasagomes: qq. Is there a spec for separating boot from deploy? | 15:01 |
| jroll | heya dtantsur :) | 15:01 |
| rloo | hi dtantsur. thx! | 15:01 |
| naohirot | dtantsur: hi | 15:02 |
| dtantsur | naohirot, g'evening (right?) | 15:02 |
| NobodyCam | brb | 15:02 |
| naohirot | dtantsur: just passed mid night here :-) | 15:03 |
| *** Masahiro has joined #openstack-ironic | 15:03 | |
| rloo | naohirot: i was just reading your reply to tan. It might be useful to add that to the spec (why you picked SCCI) | 15:04 |
| kfox1111 | nice. the local boot thing fixed my sl7 image. thanks again. :) | 15:05 |
| kfox1111 | ok. I think this should be stable enough to consider replacing cobbler with ironic. :) | 15:05 |
| kfox1111 | great work all. :) | 15:06 |
| naohirot | rloo: Okay, I'll do that. which section should I put it? How about at the end of "Proposed change"? | 15:06 |
| jroll | \o/ | 15:07 |
| jroll | kfox1111: that's awesome :D | 15:07 |
| rloo | naohirot: yes, in Proposed change is good. You could put it with the 'SCCI uses HTTP/... line. | 15:07 |
| rloo | naohirot: or add it as a separate para to the end of that section. | 15:08 |
| lucasagomes | rloo, no there's not | 15:08 |
| *** Masahiro has quit IRC | 15:08 | |
| lucasagomes | rloo, there's a blueprint and some patches (PoC) | 15:08 |
| lucasagomes | but no spec | 15:08 |
| rloo | lucasagomes: ah. ok, cuz in monday's meeting we were all wondering. and we didn't discuss cuz you weren't there. | 15:08 |
| rloo | yeah, another convert. welcome kfox1111 :-) | 15:09 |
| lucasagomes | rloo, :( I see sorry | 15:10 |
| NobodyCam | kfox1111: Awesome news :) | 15:10 |
| *** jcoufal has quit IRC | 15:10 | |
| kfox1111 | i gota clean up the docs a bit but hopefully I can post it somewhere. I got rdo/packstack to deploy almost everything in one shot, then tweaked it to do ironic instead of kvm. Simplify's deployment greatly. | 15:10 |
| lucasagomes | next monday I will be on | 15:10 |
| rloo | lucasagomes: no worries. maybe for next mon, put some links in the agenda beforehand if you want to discuss. | 15:11 |
| lucasagomes | ack | 15:11 |
| jroll | NobodyCam: rloo: and kfox is using the agent driver :o | 15:11 |
| rloo | lucasagomes: i saw in some spec (stateless pxe driver) you mentioned that it would be hard to separate boot/deploy, so I was a bit confused | 15:12 |
| NobodyCam | w00t | 15:12 |
| NobodyCam | kfox1111: I've love to review some doc patches :) | 15:12 |
| kfox1111 | yeah. the pxe driver wouldn't work for us. the agent does. | 15:12 |
| kfox1111 | NobodyCam: ok. I'll see what I can do. | 15:12 |
| rloo | jroll: maybe we need to start a count of how many are using agent vs pxe. didn't someone say they wanted to change the default to be agent? | 15:13 |
| NobodyCam | :) | 15:13 |
| jroll | rloo: that may have been me :D | 15:13 |
| jroll | rloo: though devananda also supports that movement | 15:13 |
| * rloo thinks we can't consider the agent as default until there is more docn :-) | 15:13 | |
| jroll | rloo: it should be trivial to make the IPA ramdisk work with the PXE driver, btw :) | 15:13 |
| jroll | ++ | 15:13 |
| jroll | agree | 15:14 |
| rloo | i also think if we want to change the default, we might want to do it in kilo (sooner) | 15:14 |
| jroll | yeah | 15:14 |
| jroll | need feature parity | 15:14 |
| jroll | and then I'm all for it | 15:14 |
| naohirot | rloo and all: I'm update the spec tomorrow morning with refreshed mind :-) so I go to bed, good night all, have a nice day! | 15:14 |
| jroll | see ya naohirot :) | 15:15 |
| rloo | naohirot: no worries, good night | 15:15 |
| naohirot | bye | 15:15 |
| *** naohirot has quit IRC | 15:15 | |
| NobodyCam | have a good night na | 15:15 |
| NobodyCam | tab fail | 15:16 |
| jroll | lol | 15:16 |
| kfox1111 | rloo: Let me clean up the docs a bit and then see if they will work for you. I should have all the steps in the docs on how to fully setup the agent. | 15:19 |
| kfox1111 | Right now a bit of pxe setup is also in the docs. I need to remove that bit. | 15:20 |
| rloo | kfox1111: jroll thanks you :-) | 15:20 |
| NobodyCam | kfox1111: that would great :) | 15:20 |
| jroll | I wonder if we should have a separate page for the two deploy drivers? | 15:20 |
| *** lazy_prince is now known as killer_prince | 15:21 | |
| jroll | if it's in one page, there's going to be too many branches | 15:21 |
| kfox1111 | In fact, really Its a bunch of shell. With a bit of tweaking, it probably could be made into a shellscript to do all the deployment. | 15:21 |
| rloo | jroll: is this the page you're thinking of? http://docs.openstack.org/developer/ironic/deploy/drivers.html | 15:22 |
| rloo | jroll: oh, or the install_guide itself. | 15:22 |
| jroll | kfox1111: yeah, though it needs to be run in different places etc, I'd rather just open source our ansible stuff :D | 15:22 |
| jroll | rloo: kind of... | 15:22 |
| jroll | but like, agent has different image requirements right now, etc | 15:22 |
| kfox1111 | jroll: that would work too. | 15:23 |
| * jroll wonders how possible that is | 15:23 | |
| devananda | morning, all | 15:23 |
| NobodyCam | good morning devananda | 15:24 |
| jroll | heya devananda | 15:24 |
| kfox1111 | in the mean time, if you want a workable ironic deployment using rdo, this script should be pretty easy to use. | 15:24 |
| kfox1111 | devananda: morning | 15:24 |
| jroll | kfox1111: awesome | 15:24 |
| kfox1111 | is there an existing repo I could put it? | 15:24 |
| kfox1111 | if its a patch to an existing repo, its easier for me to get it through legal. | 15:25 |
| jroll | we have some red hat folks here, maybe you can just give it to them and they can add it to rdo :P | 15:25 |
| jroll | more realistically, maybe it fits into our tools/ directory | 15:25 |
| *** anderbubble has joined #openstack-ironic | 15:25 | |
| jroll | or could put it in the docs and just add some text around the code blocks | 15:25 |
| NobodyCam | omg - gota love the news fox digs up: http://www.myfoxchicago.com/story/27465528/nurse-injects-lethal-dose-of-coffee-milk-into-patients-iv-drip | 15:26 |
| kfox1111 | jroll: tools ... that sounds good. | 15:26 |
| kfox1111 | jroll: the docs could almost be a curl | sh with this script. | 15:26 |
| jroll | NobodyCam: wow, and I thought a coffee IV would be a good thing | 15:28 |
| kfox1111 | heh. | 15:28 |
| kfox1111 | or cafinated soap. ;0 | 15:28 |
| jroll | kfox1111: I mean, for that it would need to be multiple scripts, one for api, one for conductor, etc | 15:28 |
| kfox1111 | ;) | 15:28 |
| jroll | that's... genius | 15:29 |
| jroll | I think | 15:29 |
| kfox1111 | yeah. this is to just get a single host, easy to deploy ironic up and running. kind of a seed ironic. | 15:29 |
| kfox1111 | if you have just a little hardware, its probably enough. for a big cluster, it won't be. | 15:29 |
| devananda | jroll, adam_g, others working on testing -- https://review.openstack.org/#/c/137054/ may interest you | 15:30 |
| dtantsur | jroll, have you seen https://review.openstack.org/#/c/137363/ ? | 15:30 |
| dtantsur | and morning devananda | 15:30 |
| jroll | whoa | 15:30 |
| jroll | when did that happen | 15:30 |
| jroll | (and why does it need a spec) | 15:30 |
| jroll | devananda: thanks | 15:31 |
| dtantsur | devananda, \o/ I will be able have devstack support for discoverd :) | 15:31 |
| devananda | it's a thing we've been talking about for a bit | 15:31 |
| devananda | support for external libs in devstack and tempest is essential as more and more projects want to, you know, test things | 15:32 |
| devananda | this'll also really help detangle / deserialize the gating jobs | 15:32 |
| jroll | yep, I've seen the chatter | 15:32 |
| devananda | jroll: oh - you were asking about the link dtantsur posted :p | 15:33 |
| jroll | haha | 15:33 |
| jroll | yes | 15:33 |
| dtantsur | :) | 15:33 |
| NobodyCam | jroll: I wounder if thats a bad sign for: https://www.gourmet-coffee.com/coffee-iv-drip-kit.html | 15:33 |
| jroll | kylestev: https://review.openstack.org/#/c/137363/1/specs/kilo/partial-image-support-for-agent-driver.rst | 15:33 |
| jroll | NobodyCam: :P | 15:34 |
| jroll | NobodyCam: I hope you saw the disclaimer there | 15:34 |
| jroll | under the share buttons | 15:34 |
| NobodyCam | lol | 15:34 |
| kylestev | jroll: I’ll take a look in a bit. thanks! | 15:35 |
| jroll | :) | 15:35 |
| * jroll wonders if faizan was also planning to write the code | 15:36 | |
| devananda | jroll: asignee: faizan ... | 15:37 |
| devananda | so I think we need to split the boot and deploy interfaces before this | 15:37 |
| jroll | right | 15:37 |
| jroll | agree | 15:37 |
| jroll | well, do we? | 15:37 |
| devananda | is there a spec up for that yet? | 15:37 |
| jroll | this will only change deploy code | 15:37 |
| jroll | and actually will be a minimal change in ironic | 15:37 |
| jroll | most of the work will be in IPA | 15:37 |
| devananda | 39 Post deploy, agent driver prepares the config for subsequent boot, either40 using pxe or vmedia as defined by the driver. | 15:38 |
| jroll | or actually, most of it will be in ripping the partitioning code into a library | 15:38 |
| jroll | gah | 15:38 |
| devananda | so, putting the partition code in a lib is one chunk of work | 15:38 |
| devananda | decoupling PXE or vmedia boot is another | 15:38 |
| *** alexpilotti has joined #openstack-ironic | 15:39 | |
| jroll | yeah | 15:41 |
| jroll | so that specific part should be post-split | 15:41 |
| rloo | devananda: wrt spec for splitting boot/deploy. lucasagomes said there isn't. >lucasagomes: rloo, there's a blueprint and some patches (PoC) | 15:41 |
| jroll | devananda: well, that doesn't need to be after, either | 15:42 |
| jroll | there's a separate agent/ilo deploy class | 15:43 |
| jroll | doing the boot/deploy split first will make the change slightly less complicated | 15:43 |
| jroll | btu I really don't think it will be a big deal | 15:43 |
| devananda | rloo: I think we all agree the work should be done, and I think the only question is how to maintain backwards compat in doing the split | 15:43 |
| jroll | pretty straightforward https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/ilo/deploy.py#L349 | 15:44 |
| devananda | jroll: hmm. this will create a separate agent/pxe class that doesn't use the iscsi deploy methods | 15:44 |
| jroll | devananda: eh? how? | 15:44 |
| jroll | we can share the code for "pxe boot an instance image" between agent/pxe deploy drivers | 15:45 |
| devananda | class IloPXEDeploy(pxe.PXEDeploy): | 15:45 |
| devananda | def deploy(self, task): | 15:45 |
| devananda | return super(IloPXEDeploy, self).deploy(task) | 15:45 |
| *** jcoufal has joined #openstack-ironic | 15:46 | |
| jroll | what class will it create? | 15:46 |
| jroll | maybe I'm misunderstanding something | 15:46 |
| devananda | https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/pxe.py#L311 | 15:46 |
| devananda | IloPXEDeploy subclasses PXEDeploy, which is really "pxe boot and iscsi deploy" | 15:46 |
| jroll | but I would think the existing classes would do: if image_is_partial(): pxe_boot_it(); else: local_boot_it() | 15:47 |
| jroll | why would that change | 15:47 |
| devananda | the PXE deploy driver doesn't ever local boot it | 15:47 |
| devananda | but that's not my point | 15:47 |
| jroll | this spec doesn't want the PXE deploy driver to local boot an image | 15:47 |
| jroll | ok | 15:47 |
| jroll | what's your point? | 15:48 |
| devananda | subclassing the PXEDeploy driver won't work with the Agent, because the PXEDeploy driver should actually be called PXEBootAndIscsiDeploy | 15:48 |
| jroll | why would we subclass the PXEDeploy driver? | 15:49 |
| devananda | ooh. duh. | 15:49 |
| devananda | you'd subclass AgentDeploy | 15:49 |
| jroll | maybe, or just build it in to AgentDeploy | 15:49 |
| devananda | never mind. I was reading the wrong class | 15:49 |
| jroll | ok :) | 15:50 |
| jroll | one more thing | 15:50 |
| *** dlpartain has quit IRC | 15:50 | |
| * devananda wants to rename all the PXEDeploy things to iSCSIDeploy, but won't, because compatibility | 15:50 | |
| jroll | ++ | 15:50 |
| jroll | we don't want this spec to encompass laying the bootloader onto the disk, correct? | 15:51 |
| jroll | and local booting that | 15:51 |
| jroll | (as another option) | 15:51 |
| devananda | right | 15:51 |
| jroll | ok | 15:51 |
| devananda | it looks like it's aiming for feature equivalence with the iscsi deployment | 15:51 |
| openstackgerrit | Shivanand Tendulker proposed openstack/ironic-specs: Ironic Management Interfaces to support UEFI Secure Boot https://review.openstack.org/135845 | 15:52 |
| jroll | yeah | 15:52 |
| *** stendulker has joined #openstack-ironic | 15:56 | |
| *** pcrews has joined #openstack-ironic | 15:56 | |
| devananda | NobodyCam: what do you think of loosening the formatting check in your license header check patch? | 16:00 |
| devananda | NobodyCam: right now it's enforcing white space and line count, not just the presense of the license string | 16:01 |
| NobodyCam | devananda: I good with any changes there. That was just a quick patch to add the check. | 16:02 |
| NobodyCam | j* are you guys good with a short spec for Partial image support or would you like a full spec? | 16:03 |
| jroll | NobodyCam: that spec looks fine, I don't see what else would go in there | 16:04 |
| jroll | oh, work items would be nice | 16:05 |
| NobodyCam | I'd kinda like to see work items and Docn impact sections | 16:05 |
| *** kfox1111 has quit IRC | 16:05 | |
| jroll | yeah | 16:05 |
| jroll | I'll leave a review | 16:05 |
| NobodyCam | you rock :) | 16:06 |
| *** ryanpetrello has quit IRC | 16:06 | |
| jroll | and done | 16:08 |
| stendulker | Hi jroll,jayF have raised review for spec related to management interfaces for uefi secure boot https://review.openstack.org/#/c/135845/1 Can you please have a look at the same. | 16:08 |
| jroll | stendulker: I could look, but I don't know enough about UEFI to actually be useful here :) | 16:10 |
| stendulker | jroll : Thanks. Your ironic perspective would be helpful to me. | 16:12 |
| jroll | there's a joke in there :P | 16:12 |
| jroll | will look | 16:12 |
| devananda | NobodyCam: new version coming. was a small change | 16:12 |
| stendulker | jroll :) | 16:13 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic-specs: Add a test for license header https://review.openstack.org/137215 | 16:13 |
| jroll | devananda: one quick thing there | 16:15 |
| NobodyCam | devananda: :) thank you :) | 16:15 |
| *** zigo has quit IRC | 16:16 | |
| devananda | jroll: ? | 16:17 |
| jroll | devananda: on that patch? | 16:17 |
| stendulker | ifarkas: I have raised review for spec related to management interfaces for uefi secure boot https://review.openstack.org/#/c/135845/1 Can you please have a look at the same. | 16:17 |
| jroll | devananda: existing specs have a new line above the link to the license | 16:17 |
| *** BertieFulton has quit IRC | 16:17 | |
| devananda | yep. test it -- it's fine | 16:17 |
| devananda | this is ignoring blank lines | 16:17 |
| devananda | and linebreaks | 16:17 |
| NobodyCam | which my version did not ignore | 16:18 |
| devananda | NobodyCam: exactly | 16:18 |
| NobodyCam | :) | 16:18 |
| * jroll looks | 16:18 | |
| jroll | devananda: where are you ignoring blank lines? | 16:18 |
| devananda | also, this should still work if someone adds a page title or other ReST metadata to the header | 16:18 |
| jroll | am I completely blind right now? | 16:18 |
| devananda | for i, line in enumerate(raw.split("\n")): | 16:19 |
| devananda | header_check = header_check + line | 16:19 |
| jroll | oh, I get it | 16:19 |
| devananda | it's not adding the "\n" back in | 16:19 |
| jroll | right | 16:19 |
| NobodyCam | :) | 16:19 |
| jroll | aha | 16:19 |
| NobodyCam | yep | 16:19 |
| jroll | thanks | 16:19 |
| devananda | jroll: I'll add a comment :) | 16:19 |
| jroll | devananda: I think I just need to add coffee | 16:19 |
| jroll | +2'd | 16:20 |
| devananda | jroll: still want a comment inline? | 16:21 |
| jroll | meh | 16:21 |
| jroll | can't hurt | 16:22 |
| jroll | I'll still +2 it if you add one :P | 16:22 |
| NobodyCam | I added a +1 | 16:22 |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic-specs: Add a test for license header https://review.openstack.org/137215 | 16:22 |
| devananda | done | 16:22 |
| jroll | thanks | 16:23 |
| * devananda tries out ttx's new spec2bp revision | 16:23 | |
| devananda | ooh, shiny! | 16:25 |
| * devananda marvels at all the LP updates | 16:26 | |
| NobodyCam | oh their working on LP again.. I thought all dev work stoped years ago | 16:26 |
| *** zigo has joined #openstack-ironic | 16:27 | |
| devananda | NobodyCam: it's a tool that automates updating blueprint status for me | 16:28 |
| devananda | based on our spe files | 16:28 |
| devananda | spec files | 16:28 |
| NobodyCam | oh :) | 16:28 |
| devananda | the old version only worked on approved specs -- this one even works on unapproved specs! | 16:29 |
| devananda | makes keeping the blueprint status up to date much faster (so I might actually do it) | 16:29 |
| NobodyCam | nice.. so we spec's and blueprints will be in sync | 16:29 |
| NobodyCam | :) | 16:29 |
| NobodyCam | lol | 16:29 |
| *** k4n0 has quit IRC | 16:30 | |
| *** kfox1111 has joined #openstack-ironic | 16:30 | |
| JayF | OK with me landing that test now then? | 16:31 |
| *** ryanpetrello has joined #openstack-ironic | 16:31 | |
| * JayF jfdi | 16:31 | |
| jroll | the question is, are you ok with landing it :P | 16:31 |
| *** bob_ has joined #openstack-ironic | 16:32 | |
| devananda | jroll: it looks like you have code up for https://blueprints.launchpad.net/ironic/+spec/expose-configdrive -- want to update the spec's status? | 16:32 |
| devananda | s/spec's/launchpad blueprint's/ | 16:32 |
| jroll | devananda: code landed :) | 16:33 |
| *** jcoufal has quit IRC | 16:33 | |
| jroll | there should be code for other drivers as well | 16:33 |
| devananda | jroll: right. so, want to update launchpad? :) | 16:33 |
| openstackgerrit | Merged openstack/ironic-specs: Add a test for license header https://review.openstack.org/137215 | 16:33 |
| jroll | devananda: sure, not sure which status is best | 16:34 |
| jroll | started? good progress? | 16:34 |
| jroll | implemented? :| | 16:34 |
| devananda | implemented == Done | 16:34 |
| *** kfox1111 has quit IRC | 16:34 | |
| devananda | "Good progress" ? | 16:35 |
| jroll | sure | 16:35 |
| jroll | JayF: -1'd you again | 16:35 |
| JayF | jroll: I'm pushing another patchset already | 16:35 |
| jroll | lol | 16:35 |
| devananda | lucasagomes: it looks like all the patches proposed for https://blueprints.launchpad.net/ironic/+spec/extend-vendor-passthru have landed -- is the feature done, or is there more coming? | 16:36 |
| jroll | JayF: we still need the job in projects.yaml, no? to be able to run -nv on ironic? | 16:36 |
| lucasagomes | devananda, it's all done | 16:36 |
| lucasagomes | even docs | 16:36 |
| JayF | jroll: I responded in gerrit already; but if you look agent_ssh still has the -nv designator | 16:37 |
| JayF | jroll: agent_ssh-src does not | 16:37 |
| jroll | aha | 16:37 |
| jroll | nice, ty | 16:37 |
| JayF | jroll: the new patchset readded the agent_ssh-nv to layout; that's what I realized I missed in my own review :) | 16:37 |
| jroll | right | 16:37 |
| bob_ | is there a way to search for node by extra? | 16:41 |
| *** bob_ is now known as kfox1111 | 16:41 | |
| jroll | kfox1111: no, not through the api | 16:41 |
| JayF | kfox1111: When I need to do that; I just pull a full nodes/detail and filter client side | 16:41 |
| kfox1111 | ie, I set extra name=cn0. now Iwant t o find it. | 16:41 |
| devananda | lucasagomes: woot | 16:41 |
| *** romcheg has quit IRC | 16:42 | |
| *** romcheg has joined #openstack-ironic | 16:42 | |
| kfox1111 | hmm... will be tricky from bash then... | 16:42 |
| jroll | python! | 16:43 |
| JayF | Would be simpler from a python script using python-ironicclient | 16:43 |
| devananda | jroll: it looks like https://blueprints.launchpad.net/ironic/+spec/maintenance-reason is also completed? | 16:43 |
| jroll | devananda: yes, that's fully done | 16:43 |
| jroll | devananda: on that note, should release a new client ;) | 16:43 |
| kfox1111 | not when your trying to write a simple bash script. ;) | 16:43 |
| kfox1111 | so detail does give it to you, but puts uuid way at the end... | 16:44 |
| kfox1111 | does ironic have a plugin to the unified client yet? | 16:44 |
| devananda | jroll: thanks for the poke :) | 16:44 |
| kfox1111 | it would be nice to get all the tools using the same cli ui, so things like bash parsable formatting can be easily enabled. :/ | 16:44 |
| jroll | kfox1111: if you're doing it in bash, check out jq: http://stedolan.github.io/jq/ | 16:44 |
| devananda | kfox1111: not afaik, but someone else could be working on it | 16:44 |
| *** ryanpetrello_ has joined #openstack-ironic | 16:45 | |
| *** Marga_ has joined #openstack-ironic | 16:46 | |
| kfox1111 | a bit ugly, but close enough: ironic node-list --detail | grep \'cn0\' | sed 's/.* \([0-9a-Z-]\+\) |[ ]*$/\1/' | 16:46 |
| jroll | wow | 16:48 |
| *** ryanpetrello has quit IRC | 16:48 | |
| *** ryanpetrello_ is now known as ryanpetrello | 16:48 | |
| rloo | devananda: would be nice to get Chris' README changes in client first: https://review.openstack.org/#/c/134541/ | 16:49 |
| jroll | does the README matter at all for a release? | 16:49 |
| rloo | jroll: it shows up in the package info for the client | 16:49 |
| NobodyCam | jroll: only that it shows up on the pypi home page | 16:49 |
| jroll | does it? | 16:49 |
| jroll | oh | 16:50 |
| * jroll reviews | 16:50 | |
| *** romcheg has quit IRC | 16:50 | |
| rloo | i just approved it | 16:50 |
| NobodyCam | jroll: https://pypi.python.org/pypi/python-ironicclient | 16:50 |
| NobodyCam | Ty rloo :) | 16:50 |
| rloo | NobodyCam: I'm not sure how it'll actually render, but we'll find out ;) | 16:50 |
| NobodyCam | rloo: I test with "rst2html.py README.rst README.html" running in the venv | 16:51 |
| NobodyCam | in my setup I dont get the same css but looks ok | 16:52 |
| jroll | I'm going to +A this and add a patch on top of it | 16:52 |
| jroll | oh, ruby ninja'd me | 16:52 |
| *** Masahiro has joined #openstack-ironic | 16:52 | |
| lucasagomes | devananda, :D | 16:52 |
| *** viktors is now known as viktors|afk | 16:53 | |
| *** eghobo has joined #openstack-ironic | 16:54 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/python-ironicclient: Add IRONIC_URL to README https://review.openstack.org/137403 | 16:55 |
| jroll | ^ quick addition to the README, something I use all the time | 16:55 |
| devananda | jroll: yes, the README matters ... in as much as any doc'n in the client matters | 16:55 |
| jroll | devananda: I didn't realize the README was posted anywhere :) | 16:56 |
| *** Masahiro has quit IRC | 16:56 | |
| kfox1111 | I havent done a new review in a while. do I commit, or just add the stuff, then git review? | 16:57 |
| jroll | git add, git commit, git review | 16:57 |
| kfox1111 | k. thx. | 16:57 |
| jroll | np | 16:57 |
| NobodyCam | kfox1111: git add blah; git commit <enter message>; git review to post to gerrit | 16:57 |
| *** Marga_ has quit IRC | 16:57 | |
| *** Marga_ has joined #openstack-ironic | 16:58 | |
| devananda | jroll: https://github.com/openstack/python-ironicclient -- scroll down :) | 16:58 |
| kfox1111 | ugg.. gota switch networks to post the review... may drop out for a sec... | 16:58 |
| lucasagomes | devananda, jroll +1 for README being important.... in fact it would be nice to make the README as part of our docs (when generated) | 16:58 |
| *** Marga_ has quit IRC | 16:59 | |
| NobodyCam | jroll: landing | 16:59 |
| openstackgerrit | Kevin Fox proposed openstack/ironic: A simple script to deploy ironic using rdo. https://review.openstack.org/137404 | 16:59 |
| jroll | devananda: the readme on the github repo is from master, not latest release :P | 16:59 |
| devananda | lucasagomes: hmm. I think README and published docs serve different purposes | 16:59 |
| devananda | jroll: yes | 16:59 |
| *** Marga_ has joined #openstack-ironic | 16:59 | |
| kfox1111 | stil there? | 16:59 |
| devananda | lucasagomes: IMO, the README should be general info, links to more detailed things. the docs should be -everything- | 16:59 |
| devananda | kfox1111: yep | 16:59 |
| jroll | devananda: right, my point was, I didn't think we needed to wait for readme changes to land before release, didn't realize it was on pypi | 16:59 |
| NobodyCam | kfox1111: si | 16:59 |
| kfox1111 | https://review.openstack.org/137404 | 16:59 |
| *** Marga_ has quit IRC | 16:59 | |
| kfox1111 | very raw yet...... | 17:00 |
| *** Marga_ has joined #openstack-ironic | 17:00 | |
| jroll | kfox1111: nice | 17:00 |
| jroll | lucasagomes: ^ check that out | 17:00 |
| JayF | I'd just be worried about how to prevent bitrot in something like that ... if it's not tested how long will it continue to work? | 17:00 |
| jroll | script for ironic + rdo | 17:00 |
| devananda | hm. so. ironic/tools/ isn't the right place for that | 17:01 |
| devananda | JayF: right | 17:01 |
| NobodyCam | do we need a user contrib folder? | 17:01 |
| lucasagomes | right... jroll checking | 17:01 |
| jroll | devananda: yeah, we were debating it earlier, not sure where to stuff it | 17:01 |
| lucasagomes | jroll, well it's being approved already | 17:02 |
| lucasagomes | haha | 17:02 |
| jroll | lucasagomes: no, https://review.openstack.org/137404 | 17:02 |
| * lucasagomes looks | 17:03 | |
| *** eghobo has quit IRC | 17:04 | |
| devananda | NobodyCam: I don't think this should be bundled with Ironic. For one, it's distro specific | 17:04 |
| dtantsur | ... and I would prefer it be ansible, not bash ... | 17:05 |
| devananda | dtantsur: ++ | 17:05 |
| *** ChuckC_ has quit IRC | 17:05 | |
| NobodyCam | ahh | 17:05 |
| devananda | dtantsur: more than likely, folks will write operational tooling like this in what ever language they like. be it bash, chef, or ansible, or what ever | 17:06 |
| kfox1111 | sure. long term it would be nice to not need to do anythign distro specific. | 17:06 |
| *** ChuckC has joined #openstack-ironic | 17:06 | |
| devananda | that's the way these things go | 17:06 |
| kfox1111 | I just wanted to get someting up and running easy. | 17:06 |
| devananda | kfox1111: totally. and thank you | 17:06 |
| kfox1111 | sure. :) | 17:06 |
| JayF | I'm not sure ironic itself should have a user contrib folder; but I think there's room for a stackforge repo that stuff can go in perhaps? | 17:06 |
| devananda | JayF: yes | 17:06 |
| devananda | I'm looking for precedent | 17:06 |
| dtantsur | kfox1111, why not use instack to get Ironic up and running with RDO? | 17:06 |
| lucasagomes | yeah looks good for that propose, it kinda looks like devstack too | 17:07 |
| devananda | https://github.com/stackforge/openstack-chef-repo | 17:07 |
| JayF | Yeah all the chef stuff is in stackforge as their own repos | 17:07 |
| JayF | but chef cookbooks are more like software than script | 17:07 |
| kfox1111 | dtantsur: because I didnt find docs explaining how? :) | 17:07 |
| devananda | https://github.com/stackforge/packstack | 17:07 |
| dtantsur | kfox1111, https://openstack.redhat.com/Deploying_an_RDO_Undercloud_with_Instack | 17:07 |
| devananda | https://github.com/stackforge/puppet-* -- there are separate repos for each service's puppet module | 17:08 |
| *** eghobo has joined #openstack-ironic | 17:08 | |
| *** eghobo has quit IRC | 17:09 | |
| devananda | kfox1111: I'm going to put a block on that patch, but until there's a better home, you're welcome to iterate and get reviews there | 17:09 |
| devananda | I think we'll want something like an ops-tools / user-contrib repo for things like this | 17:10 |
| jroll | ++ | 17:10 |
| devananda | though honestly, I think the best long-term solution is a separate OpenStack Program for operational tools | 17:10 |
| jroll | JayF: we could likely drop a script or three there as well | 17:10 |
| jroll | devananda: I would think that would fit into the deployments program | 17:11 |
| kfox1111 | devananda: ok, thanks. | 17:11 |
| devananda | tripleo, packstack, all the chef and puppet things, etc, would live under that | 17:11 |
| devananda | jroll: yup | 17:11 |
| *** romcheg has joined #openstack-ironic | 17:11 | |
| devananda | SpamapS: poke, just in case you have opinions or more information on ^ than I do | 17:11 |
| *** ryanpetrello has quit IRC | 17:12 | |
| devananda | gotta step afk for ~30 min | 17:13 |
| kfox1111 | dtantsur: thanks. I'll take a look. | 17:14 |
| kfox1111 | dtantsur: Does instack-undercloud support the agent? | 17:20 |
| dtantsur | kfox1111, IIRC no. You can add it later ofc | 17:21 |
| openstackgerrit | Merged openstack/ironic: Change methods from classmethod to staticmethod https://review.openstack.org/137335 | 17:21 |
| kfox1111 | ok. | 17:21 |
| kfox1111 | is instack redhat's longer term plan for a packstack replacement? | 17:22 |
| openstackgerrit | Merged openstack/python-ironicclient: Fix log_curl_request API version duplication https://review.openstack.org/136773 | 17:22 |
| openstackgerrit | Merged openstack/ironic: Improve docs for running IPA in Devstack https://review.openstack.org/137139 | 17:23 |
| openstackgerrit | Merged openstack/ironic: iLO Management Interface https://review.openstack.org/132746 | 17:23 |
| jroll | so many merges <3 | 17:23 |
| *** Marga_ has quit IRC | 17:24 | |
| kfox1111 | is it safe to install the instack packages to see whats inside? | 17:25 |
| *** Marga_ has joined #openstack-ironic | 17:26 | |
| jroll | devananda: you should approve this https://review.openstack.org/#/c/136934/ | 17:26 |
| SpamapS | devananda: reading | 17:27 |
| dtantsur | kfox1111, 1. I can't answer these kind of questions, sorry :) 2. yes, installing packages does not do anything destructive | 17:28 |
| kfox1111 | ok. thanks. :) | 17:28 |
| *** Marga_ has quit IRC | 17:30 | |
| *** Marga_ has joined #openstack-ironic | 17:30 | |
| SpamapS | kfox1111: regarding the rdo ironic thing.. I think there is a home for stuff like that in the Deployment program. | 17:31 |
| kfox1111 | ok. is there a repo for it yet? | 17:32 |
| SpamapS | kfox1111: generally we ask that you split things into software delivery, configuration, and initialization steps, so that people can choose images, packages, containers, etc. for delivery to systems... so a monolithic script isn't exactly what would work. | 17:33 |
| kfox1111 | how does packstack fit into that, since it does all the above too? | 17:34 |
| SpamapS | kfox1111: Before you do that, take a look at the instack stuff. | 17:34 |
| kfox1111 | yeah. I'm looking at it now. | 17:34 |
| SpamapS | https://openstack.redhat.com/Deploying_RDO_using_Instack | 17:34 |
| *** sambetts has quit IRC | 17:34 | |
| *** andreykurilin_ has quit IRC | 17:34 | |
| kfox1111 | really, I posted it just because I read through the current ironic docs, followed them, got to the end, and then relized I needed to do it differently to do the agent, then flounderd a while trying to figure out how to get swift tmp url's to work with it. | 17:35 |
| kfox1111 | just wanted to post what I did to get a working setup so someone else wont flounder trying to do the same. | 17:35 |
| *** ryanpetrello has joined #openstack-ironic | 17:35 | |
| jroll | and you're awesome for that, that's super helpful | 17:35 |
| jroll | even if it doesn't land, we can point people at it | 17:35 |
| kfox1111 | yup. thats what I figured too. | 17:36 |
| kfox1111 | the swift temp url registration post thing was particularly... unexpected. :) | 17:36 |
| openstackgerrit | Merged openstack/python-ironicclient: Update README https://review.openstack.org/134541 | 17:38 |
| *** anderbubble has quit IRC | 17:43 | |
| *** rloo has quit IRC | 17:45 | |
| jroll | yeah :| | 17:45 |
| *** dlaube has joined #openstack-ironic | 17:48 | |
| *** Marga_ has quit IRC | 17:52 | |
| *** Marga_ has joined #openstack-ironic | 17:53 | |
| *** Marga_ has quit IRC | 17:54 | |
| NobodyCam | brb | 17:54 |
| *** Marga_ has joined #openstack-ironic | 17:55 | |
| *** Marga_ has quit IRC | 17:55 | |
| *** jistr has quit IRC | 17:56 | |
| *** Marga_ has joined #openstack-ironic | 17:56 | |
| *** derekh has quit IRC | 17:58 | |
| *** ryanpetrello_ has joined #openstack-ironic | 18:01 | |
| *** harlowja_away is now known as harlowja | 18:04 | |
| *** ryanpetrello has quit IRC | 18:04 | |
| *** ryanpetrello_ is now known as ryanpetrello | 18:04 | |
| *** Marga_ has quit IRC | 18:06 | |
| *** Marga_ has joined #openstack-ironic | 18:06 | |
| *** Marga_ has quit IRC | 18:07 | |
| *** Marga_ has joined #openstack-ironic | 18:07 | |
| NobodyCam | oh thats why it hasn't worked, I forgot to include the correct elements in my image build *face+plams+d'oh* | 18:09 |
| jroll | lol | 18:10 |
| NobodyCam | I was adding swift elements but left out the proxy element :-p | 18:10 |
| *** igordcard has joined #openstack-ironic | 18:11 | |
| *** foexle has quit IRC | 18:11 | |
| *** Marga_ has quit IRC | 18:12 | |
| *** igordcard has quit IRC | 18:12 | |
| *** Marga_ has joined #openstack-ironic | 18:12 | |
| *** igordcard has joined #openstack-ironic | 18:12 | |
| *** dtantsur is now known as dtantsur|afk | 18:13 | |
| *** Marga_ has quit IRC | 18:13 | |
| *** Marga_ has joined #openstack-ironic | 18:14 | |
| NobodyCam | have a good holiday dtantsur|afk | 18:14 |
| *** Marga_ has quit IRC | 18:18 | |
| devananda | found a kinesis keyboard in the office.... this is neat. been meaning to try one for a whilei | 18:18 |
| *** Marga_ has joined #openstack-ironic | 18:18 | |
| NobodyCam | kfox1111: sick "TEMP_URL_KEY=`openssl rand -hex 12`" | 18:18 |
| *** harlowja has quit IRC | 18:18 | |
| NobodyCam | slick even | 18:18 |
| NobodyCam | :0p | 18:18 |
| *** stendulker has quit IRC | 18:19 | |
| *** Marga_ has quit IRC | 18:19 | |
| *** harlowja has joined #openstack-ironic | 18:19 | |
| *** Marga_ has joined #openstack-ironic | 18:19 | |
| *** Marga_ has quit IRC | 18:21 | |
| *** Marga_ has joined #openstack-ironic | 18:22 | |
| *** anderbubble has joined #openstack-ironic | 18:24 | |
| *** athomas has quit IRC | 18:24 | |
| *** Marga_ has quit IRC | 18:25 | |
| *** Marga_ has joined #openstack-ironic | 18:25 | |
| *** ifarkas has quit IRC | 18:27 | |
| *** erwan_taf has quit IRC | 18:28 | |
| *** lucasagomes has quit IRC | 18:28 | |
| openstackgerrit | Merged openstack/python-ironicclient: Add IRONIC_URL to README https://review.openstack.org/137403 | 18:30 |
| *** ryanpetrello_ has joined #openstack-ironic | 18:30 | |
| *** Marga_ has quit IRC | 18:32 | |
| *** Marga_ has joined #openstack-ironic | 18:32 | |
| *** ryanpetrello has quit IRC | 18:33 | |
| *** ryanpetrello_ is now known as ryanpetrello | 18:33 | |
| *** Marga_ has quit IRC | 18:35 | |
| *** Marga_ has joined #openstack-ironic | 18:36 | |
| *** foexle has joined #openstack-ironic | 18:36 | |
| *** Marga_ has quit IRC | 18:39 | |
| *** Marga_ has joined #openstack-ironic | 18:39 | |
| *** andreykurilin_ has joined #openstack-ironic | 18:39 | |
| *** Masahiro has joined #openstack-ironic | 18:41 | |
| *** harlowja_ has joined #openstack-ironic | 18:41 | |
| NobodyCam | lol i luv the error message: http://paste.openstack.org/show/7pACL5gTMN90FFLywbS2/ | 18:43 |
| JayF | 410 is a strange http code for that error | 18:43 |
| NobodyCam | its a misconfigured glance backed swift | 18:44 |
| NobodyCam | :-p | 18:44 |
| *** harlowja has quit IRC | 18:45 | |
| *** rushiagr is now known as rushiagr_away | 18:45 | |
| *** Masahiro has quit IRC | 18:46 | |
| *** spandhe has joined #openstack-ironic | 18:52 | |
| kfox1111 | NobodyCam: Yeah. I can't remember where I ran across that trick. I just remember thinking "openssl can do that? oh.. i guess that makes sense." | 18:54 |
| openstackgerrit | Ghe Rivero proposed openstack/ironic: Use olso.config generator https://review.openstack.org/137447 | 19:05 |
| NobodyCam | :) | 19:06 |
| *** anderbubble has quit IRC | 19:08 | |
| *** penick has joined #openstack-ironic | 19:09 | |
| *** kfox1111 has quit IRC | 19:09 | |
| *** mjturek has quit IRC | 19:12 | |
| *** andreykurilin_ has quit IRC | 19:13 | |
| alexm__ | how do I clean things with nova+ironic when an instance failed? | 19:18 |
| alexm__ | « nova delete » has no effect, nor « ironic node-update xxx remove instance_info ». The ironic node appears to still be in use, I still see information attached to it in database. | 19:21 |
| JayF | so likely it's in a state transition still | 19:22 |
| JayF | which means you can't update it | 19:22 |
| alexm__ | | 6ffa7d9c-6773-43dd-9592-e6a43a4c6d36 | test-alex | ERROR | deleting | NOSTATE | | | 19:25 |
| alexm__ | it says instance was in error because no valid host found | 19:25 |
| JayF | see how the task state is deleting | 19:26 |
| JayF | that's why you can't delete it again | 19:26 |
| JayF | if it's in error because no host found, it's really strange that you'd have a node on the backend deployed to | 19:27 |
| *** foexle has quit IRC | 19:27 | |
| JayF | ooooh, unless you hit max reschedule | 19:27 |
| JayF | jroll: ^ no host found is error when all reschedules fail, right? | 19:27 |
| alexm__ | initially I had this | 19:27 |
| alexm__ | | fault | {"message": "No valid host was found. ", "code": 500, "details": " File \"/opt/venvs/ironic/lib/python2.6/site-packages/nova/conductor/manager.py\", line 614, in build_instances | | 19:27 |
| JayF | you have to look at the node on the ironic side as well | 19:28 |
| JayF | and see the logs | 19:28 |
| alexm__ | did a reset-state and force-delete after, didn’t help much :\ | 19:28 |
| JayF | as to why/how it failed | 19:28 |
| alexm__ | everything looks good on ironic conductor, it’s polling the remote node via ipmi | 19:29 |
| *** Haomeng has joined #openstack-ironic | 19:30 | |
| *** Haomeng|2 has quit IRC | 19:31 | |
| *** spandhe has quit IRC | 19:33 | |
| *** spandhe has joined #openstack-ironic | 19:35 | |
| jroll | JayF: correct, but there's lots of ways to be rescheduled | 19:37 |
| alexm__ | by reschedule, you mean max_retries ? | 19:38 |
| *** BertieFulton has joined #openstack-ironic | 19:39 | |
| JayF | basically when nova deploys an instance | 19:39 |
| JayF | it has a number of times it'll try (IDK, it may be max_retries?) to deploy the instance before giving up | 19:40 |
| JayF | my hunch is that something nova->ironic is broken in your installation, and it tried to deploy your instance $retires number of times and failed out throwing the no host found | 19:40 |
| JayF | but without knowing what the ironic node itself is saying and what nova-compute was logging when it was trying to deploy, it's nearly impossible to help :( | 19:40 |
| alexm__ | I’m running debug mode in foreground for all services, so there’s lot of information | 19:42 |
| alexm__ | I’ll just recreate a new ironic node and make a new launch instance | 19:42 |
| alexm__ | BTW, I’m not in devstack | 19:43 |
| JayF | yeah that's good :) | 19:43 |
| *** BertieFulton has quit IRC | 19:44 | |
| *** mrda-away is now known as mrda | 19:44 | |
| alexm__ | ok I was issing the ironic port-create just before, so now it seems to have passed scheduling | 19:45 |
| mrda | Morning Ironic | 19:45 |
| alexm__ | but I’m stucked where I was yesterday, basically the instance is in building/spawning state | 19:46 |
| alexm__ | oh wait, RPC issue just got it | 19:47 |
| jroll | hiya mrda | 19:47 |
| alexm__ | 2014-11-26 19:45:55.760 10224 DEBUG nova.compute.manager [-] [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] Skipping network cache update for instance because it is Building. _heal_instance_info_cache /opt/venvs/ironic/lib/python2.6/site-packages/nova/compute/manager.py:5287 | 19:48 |
| alexm__ | 2014-11-26 19:45:55.761 10224 DEBUG nova.compute.manager [-] Didn't find any instances for network info cache update. _heal_instance_info_cache /opt/venvs/ironic/lib/python2.6/site-packages/nova/compute/manager.py:5339 | 19:48 |
| alexm__ | 2014-11-26 19:45:55.761 10224 DEBUG nova.openstack.common.loopingcall [-] Dynamic looping call <bound method Service.periodic_tasks of <nova.service.Service object at 0x34b60d0>> sleeping for 7.00 seconds _inner /opt/venvs/ironic/lib/python2.6/site-packages/nova/openstack/common/loopingcall.py:132 | 19:48 |
| alexm__ | 2014-11-26 19:46:01.307 10224 ERROR nova.compute.manager [req-e1d56a22-588b-43b0-ac75-baeef00e2dc9 None] [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] Failed to deallocate networks | 19:48 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] Traceback (most recent call last): | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/nova/compute/manager.py", line 2255, in _cleanup_allocated_networks | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] self._deallocate_network(context, instance, requested_networks) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/nova/compute/manager.py", line 1917, in _deallocate_network | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] context, instance, requested_networks=requested_networks) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/nova/network/api.py", line 48, in wrapped | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] return func(self, context, *args, **kwargs) | 19:49 |
| devananda | alexm__: whoa | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/nova/network/api.py", line 299, in deallocate_for_instance | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] requested_networks=requested_networks) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/nova/network/rpcapi.py", line 210, in deallocate_for_instance | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] return cctxt.call(ctxt, 'deallocate_for_instance', **kwargs) | 19:49 |
| devananda | alexm__: please dont do that | 19:49 |
| jroll | whoaaaaaaa | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/rpc/client.py", line 152, in call | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] retry=self.retry) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/transport.py", line 90, in _send | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] timeout=timeout, retry=retry) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/_drivers/amqpdriver.py", line 408, in send | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] retry=retry) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/_drivers/amqpdriver.py", line 397, in _send | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] result = self._waiter.wait(msg_id, timeout) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/_drivers/amqpdriver.py", line 298, in wait | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] reply, ending, trylock = self._poll_queue(msg_id, timeout) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/_drivers/amqpdriver.py", line 238, in _poll_queue | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] message = self.waiters.get(msg_id, timeout) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] File "/opt/venvs/ironic/lib/python2.6/site-packages/oslo/messaging/_drivers/amqpdriver.py", line 144, in get | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] 'to message ID %s' % msg_id) | 19:49 |
| alexm__ | 2014-11-26 19:46:01.307 10224 TRACE nova.compute.manager [instance: a33a761c-ef94-4de4-bea4-b89e86bffb5f] MessagingTimeout: Timed out waiting for a reply to message ID a465d1cc03844e16ac99f6ef0af27692 | 19:49 |
| alexm__ | sorry for long trace | 19:49 |
| alexm__ | hmm should have copied that snippet elsewhere, sorry guys | 19:49 |
| devananda | alexm__: ok -- I was about to kick you | 19:49 |
| devananda | please do not do that again | 19:50 |
| jroll | so that means rabbit isn't working | 19:50 |
| devananda | use paste.openstack.org or something likethat | 19:50 |
| jroll | err, maybe it's not a rabbit issue | 19:50 |
| *** ChanServ sets mode: +o devananda | 19:50 | |
| alexm__ | yes I will next time :) | 19:50 |
| jroll | are you using nova-network? | 19:50 |
| JayF | If you're going to give us that much spam, at least fry it up with some mustard ;) | 19:50 |
| jroll | that looks like nova-network. | 19:50 |
| devananda | also, yikes. hard to type on regular keyboard after an hour on the kinesis | 19:51 |
| alexm__ | ahh damn, must be a misconfiguration between neutron and nova | 19:51 |
| jroll | devananda: lol | 19:51 |
| devananda | quick poll - are there any unapproved specs that yall have on your short list for the next three weeks? | 19:52 |
| alexm__ | using neutron with the new setup, but with a partial nova.conf from old nova-network setup | 19:53 |
| devananda | i've been updating targeting on LP -- https://launchpad.net/ironic/+milestone/kilo-1 is the list in my head | 19:53 |
| NobodyCam | devananda: like https://review.openstack.org/133828 | 19:53 |
| jroll | kilo-1 is in 3 weeks? | 19:53 |
| devananda | there are two approv3ed specs with code up | 19:53 |
| devananda | yes | 19:53 |
| devananda | dec 18 | 19:53 |
| devananda | https://wiki.openstack.org/wiki/Kilo_Release_Schedule | 19:54 |
| JayF | I think a lot of stuff is blocked on state machine | 19:54 |
| devananda | yup | 19:54 |
| JayF | if that's going to be done, it needs to land by k-1 | 19:54 |
| jroll | devananda: add https://review.openstack.org/#/c/137363/ | 19:54 |
| JayF | or else hold up everything else for much too long | 19:54 |
| jroll | devananda: I should say, I'd like to see that one done, but idk if code will make it in 3 weeks | 19:54 |
| NobodyCam | maybe https://review.openstack.org/128927 | 19:55 |
| jroll | NobodyCam: I don't think there's any way code could be done for that in 3 weeks | 19:56 |
| jroll | oh wait | 19:56 |
| jroll | yeah, I think that one could land | 19:57 |
| jroll | if we limit the scope to just "give me the capabilities for this driver" | 19:57 |
| devananda | so, driver-capabilites is targeted to the backlog right now | 19:58 |
| JayF | that's hardware-capabilities | 19:58 |
| JayF | not driver-capabilities | 19:58 |
| JayF | oh wow that one is on backlog as well | 19:59 |
| devananda | https://review.openstack.org/#/c/128927/ -- specs/backlog/driver-capabilities.rst | 19:59 |
| devananda | ya | 19:59 |
| JayF | it def wasn't last time I looked at it, sorry | 19:59 |
| devananda | rev1 was not | 19:59 |
| jroll | right, I don't think there's disagreement that we want that | 19:59 |
| *** zer0c00l has quit IRC | 19:59 | |
| devananda | sure | 19:59 |
| jroll | I wonder if dtantsur|afk wants to pick that back up | 20:00 |
| jroll | I could help if needed | 20:00 |
| *** ryanpetrello has quit IRC | 20:00 | |
| devananda | https://launchpad.net/ironic/+milestone/kilo-1 now has 2 approved in progress, and two unapproved | 20:00 |
| JayF | jroll: you know that https://review.openstack.org/#/c/131272/ is looking for a home now too, right? :P | 20:01 |
| devananda | one of those looks fairly simple, one is the state machine | 20:01 |
| devananda | which, ya, we des- | 20:01 |
| jroll | JayF: I'm not going anywhere near that :) | 20:01 |
| devananda | definitely need to land the spec ASAP, and then the code very soon thereafter | 20:01 |
| * NobodyCam notices the time and runs out the door for lunch ... lol | 20:01 | |
| JayF | devananda: how do already-fixed bugs work w/r/t targetting? | 20:02 |
| devananda | it has the potential to drag on for the whole cycle otherwise, and stall many other things | 20:02 |
| JayF | devananda: i.e. if I fixed a bug that's worthy of release notes, should I have you target it for k1? | 20:02 |
| devananda | JayF: they get auto targeted when a milestone is tagged | 20:02 |
| JayF | aha okay | 20:02 |
| JayF | cool | 20:02 |
| jroll | devananda: if another core or two is around friday, maybe we can get together and crank that out quickly | 20:02 |
| jroll | the spec, that is | 20:02 |
| devananda | targeting is only necessary for prioritizing work / revie2ws | 20:02 |
| JayF | okay, cool. | 20:02 |
| devananda | jroll: ++ | 20:02 |
| *** ryanpetrello has joined #openstack-ironic | 20:02 | |
| * JayF is gone until Monday in about 30 minutes :D | 20:03 | |
| *** zer0c00l has joined #openstack-ironic | 20:03 | |
| devananda | JayF: also, targeting is used to block releases until critical things are done | 20:03 |
| NobodyCam | JayF: have a great holiday | 20:03 |
| devananda | JayF: cheers, enhjoy the break :) | 20:03 |
| NobodyCam | Is it me or is the state machine review going backwards | 20:04 |
| jroll | yes, it is | 20:04 |
| devananda | how so? | 20:04 |
| jroll | idk where these new states are coming from | 20:04 |
| * devananda hasnt read all the latest discussion | 20:04 | |
| devananda | ???? | 20:05 |
| NobodyCam | devananda: read the last comment | 20:05 |
| jroll | we spent hours on this at the summit, it should essentially be a rubber stamp | 20:05 |
| JayF | I don't agree with that really jroll, lots of things have come up which weren't mentioned at the summit | 20:05 |
| JayF | that being said it should be getting closer to resolution not further | 20:05 |
| jroll | sure, but we shouldn't be changing the actual states we're putting in | 20:06 |
| *** ryanpetrello has quit IRC | 20:10 | |
| *** penick_ has joined #openstack-ironic | 20:11 | |
| *** penick has quit IRC | 20:13 | |
| *** penick_ is now known as penick | 20:13 | |
| NobodyCam | brb | 20:14 |
| *** ryanpetrello has joined #openstack-ironic | 20:20 | |
| *** andreykurilin_ has joined #openstack-ironic | 20:20 | |
| *** Masahiro has joined #openstack-ironic | 20:30 | |
| NobodyCam | added a comment to the state machine spec... please feel free to yell at me | 20:32 |
| dlaube | hey guys, what's the best way to filter a list of nodes by property? | 20:33 |
| * jroll yells at NobodyCam for posting a good comment | 20:33 | |
| jroll | dlaube: you need to filter properties on the client side | 20:34 |
| devananda | dlaube: what jroll said | 20:34 |
| *** Masahiro has quit IRC | 20:34 | |
| NobodyCam | :-p | 20:35 |
| dlaube | I was thinking about tagging some of the nodes with a location under the "extra" property | 20:35 |
| dlaube | no way to filter using ironic CLI or in the API? | 20:35 |
| JayF | dlaube: we store that info there too | 20:36 |
| JayF | dlaube: but generally for our scripts; we just pull down the whole node details object and filter that | 20:36 |
| NobodyCam | dlaube: awk can do great things with text :) | 20:36 |
| * JayF gone for the holiday | 20:36 | |
| NobodyCam | have fun JayF | 20:36 |
| devananda | dlaube: there is no way to filter within a nested resource, like properties, in our API | 20:36 |
| *** penick has quit IRC | 20:37 | |
| *** andreykurilin_ has quit IRC | 20:37 | |
| devananda | dlaube: ive talked with folks about adding arbitrary tag support, in such a way that the tags could be searched performantly | 20:37 |
| devananda | dlaube: but afaik no one is working on that | 20:37 |
| devananda | headin out for lunch, bbi45 | 20:39 |
| NobodyCam | enjoy devananda | 20:39 |
| NobodyCam | :) | 20:39 |
| dlaube | ok, thanks devananda and NobodyCam | 20:41 |
| * dlaube loves awk | 20:41 | |
| NobodyCam | :) | 20:41 |
| NobodyCam | brb | 20:47 |
| dlaube | when I nova boot, I can only provide one SSH key that gets deployed on to the ironic node | 20:47 |
| dlaube | is cloud-init the way to go for multiple ssh keys? | 20:48 |
| jroll | maybe, you could also use automation post-boot | 20:48 |
| dlaube | yeah, I was thinking about rolling out chef-client to handle it but ideally I could handle at provision time | 20:51 |
| jroll | yeah, cloud-init user data is your best bet for that afaik | 20:52 |
| dlaube | I'll have to study up on it | 20:52 |
| dlaube | then its just a matter of nova boot --user-data blah.file | 20:52 |
| jroll | yeah, something like that | 20:52 |
| dlaube | I *think* | 20:52 |
| dlaube | thanks for listening to my rant and for blessing the —user-data idea jroll | 20:53 |
| NobodyCam | dlaube: you use DIB to build your images? | 20:57 |
| mrda | devananda: we should talk about tags when you get back | 20:58 |
| NobodyCam | morning mrda | 20:59 |
| dlaube | NobodyCam: yessir | 20:59 |
| mrda | hey NobodyCam | 20:59 |
| dlaube | g'morning mrda | 20:59 |
| mrda | hi dlaube | 20:59 |
| openstackgerrit | Ghe Rivero proposed openstack/ironic: Use olso.config generator https://review.openstack.org/137447 | 20:59 |
| zer0c00l | What do you guys use for developing ironic? I am trying to use devstack (master branch) and the local.conf on a Fedora 19 machine | 20:59 |
| NobodyCam | dlaube: I know there a couple of salt elelment not sure about chef thou | 20:59 |
| zer0c00l | Almost most of the services end up in error state | 20:59 |
| * NobodyCam uses tripleo devtest scripts | 21:00 | |
| NobodyCam | many folk use devstack | 21:00 |
| zer0c00l | NobodyCam: from master branch? | 21:01 |
| zer0c00l | NobodyCam: can you share your localconf please? | 21:01 |
| zer0c00l | neutrol fails with "] ConnectionFailed: Connection to neutron failed: ('Connection aborted.', error(111, 'Connection refused'))" | 21:01 |
| dlaube | ok, ty NobodyCam | 21:02 |
| dlaube | been a fan of chef since the beginning and just recently getting into ansible | 21:02 |
| NobodyCam | zer0c00l: I do most all of my work from devtest's seed vm | 21:02 |
| zer0c00l | NobodyCam: where do i get one? | 21:03 |
| zer0c00l | devtest seed vm? | 21:03 |
| zer0c00l | never heard of it | 21:03 |
| NobodyCam | zer0c00l: its part of the TripleO story | 21:05 |
| NobodyCam | If you've not heard of that I would point you to: http://docs.openstack.org/developer/tripleo-incubator/README.html | 21:05 |
| NobodyCam | as a first read | 21:05 |
| NobodyCam | but I manually modify the devtest scripts to stop where I want them ot | 21:06 |
| NobodyCam | so I tend to not deploy the undercloud and overclouds | 21:06 |
| *** achanda has joined #openstack-ironic | 21:15 | |
| *** andreykurilin_ has joined #openstack-ironic | 21:19 | |
| *** achanda has quit IRC | 21:20 | |
| jroll | dlaube: :P | 21:22 |
| jroll | zer0c00l: I just use devstack when I need to test new code, my localconf is just the one from http://docs.openstack.org/developer/ironic/dev/dev-quickstart.html#deploying-ironic-with-devstack | 21:23 |
| jroll | I unstack and stack often | 21:23 |
| Shrews | jroll: speaking of which, i know we discussed this before, but it's becoming a thing now: https://github.com/Shrews/dockstack | 21:25 |
| jroll | oh snap | 21:26 |
| NobodyCam | Shrews: ahh another one :-p | 21:26 |
| Shrews | hope to have something going after the break | 21:26 |
| jroll | Shrews: does it work? | 21:26 |
| jroll | +1000 for not being bash | 21:26 |
| Shrews | jroll: not quite. getting there. the ground work is done, now just need to add configuring the containers | 21:27 |
| jroll | cool | 21:27 |
| *** amitpp has joined #openstack-ironic | 21:30 | |
| *** romcheg has quit IRC | 21:30 | |
| *** romcheg has joined #openstack-ironic | 21:31 | |
| *** romcheg has quit IRC | 21:31 | |
| *** ryanpetrello has quit IRC | 21:32 | |
| *** romcheg has joined #openstack-ironic | 21:38 | |
| NobodyCam | davidlenwell: refstsck.org down? | 21:39 |
| NobodyCam | *http://refstack.org | 21:40 |
| devananda | mrda: back | 21:41 |
| NobodyCam | wb devananda :) | 21:41 |
| *** achanda has joined #openstack-ironic | 21:48 | |
| *** achanda has quit IRC | 21:49 | |
| *** ifarkas_ has joined #openstack-ironic | 21:56 | |
| mrda | devananda: so tags. You want nova-style tagging of instances? | 21:57 |
| devananda | mrda: i'm not sure wha5t nova style means | 21:57 |
| *** amitpp has quit IRC | 21:58 | |
| *** ifarkas_ has quit IRC | 21:59 | |
| mrda | well, it's an arbitrary 1:N tagging mechanism | 21:59 |
| mrda | which (I believe) can be used for scheduling | 22:00 |
| devananda | gotcha | 22:08 |
| devananda | mrda: oh, also, your logical names spec - i dont see a BP for it?e | 22:09 |
| mrda | devananda: not yet :) | 22:10 |
| mrda | Should I create one now? | 22:11 |
| devananda | mrda: so arbitrary 1:N tagging, which can be queried in ironic's API - that's a thing which the Tuskar folks were asking about a cycle or two ago | 22:11 |
| mrda | (on the assumption we can get agreement on the spec?) | 22:11 |
| devananda | mrda: you should create a BP when ever you create a spec :) | 22:11 |
| mrda | ok, it's my first one :) | 22:11 |
| devananda | mrda: if you create one now, i will update it now too | 22:11 |
| devananda | ah, neat! | 22:11 |
| * mrda wanders off to look at bp's | 22:12 | |
| devananda | mrda: name must be the same as the spec file. you can just copy the first para from the spec | 22:13 |
| devananda | mrda: dont worry about targeting, etc - that's my job for now | 22:13 |
| NobodyCam | j*: as a known swift user should I see any objects I have in glance with a swift list command? | 22:15 |
| jroll | NobodyCam: maybe? | 22:15 |
| *** alexpilotti has quit IRC | 22:15 | |
| jroll | NobodyCam: it might be a service account thing, not sure | 22:15 |
| devananda | NobodyCam: is glance configured with swift as its backend? | 22:15 |
| jroll | that is to say, I'm not sure which account glance uses for images | 22:16 |
| jroll | I have gone digging into the sqlite db (swift, wat) to figure those things out | 22:16 |
| jroll | it's pretty simple | 22:16 |
| mrda | devananda: https://blueprints.launchpad.net/ironic/+spec/logical-names | 22:18 |
| *** Masahiro has joined #openstack-ironic | 22:18 | |
| devananda | mrda: updated :) | 22:19 |
| mrda | ta | 22:19 |
| mrda | or should I say | 22:19 |
| mrda | \o/ | 22:19 |
| devananda | :)u | 22:19 |
| devananda | urgh | 22:19 |
| devananda | hitting too man6y randm keys still | 22:19 |
| jroll | lolol | 22:19 |
| NobodyCam | devananda: I think so... Thats what I'm checking now | 22:20 |
| NobodyCam | lol | 22:20 |
| JoshNang | NobodyCam: are you doing swift list as the same user as glance is configured to use? | 22:22 |
| NobodyCam | JoshNang: as admin | 22:22 |
| *** Masahiro has quit IRC | 22:23 | |
| JoshNang | NobodyCam: i'd assume you can see them then | 22:23 |
| JoshNang | but i'm not a swift guru | 22:23 |
| mrda | JoshNang: you're a slow guru? | 22:24 |
| jroll | lol | 22:24 |
| JoshNang | mrda: hah | 22:24 |
| JoshNang | too bad JayF is already gone for the day | 22:24 |
| JoshNang | i'm sure he'd have a few more puns to add :) | 22:25 |
| NobodyCam | JoshNang: swift --os-username glance --os-password unset --os-auth-url http://127.0.0.1:5000/v2.0/ --os-tenant-name service list worked like it should have ... TY | 22:26 |
| JoshNang | NobodyCam: woo! | 22:26 |
| NobodyCam | http://paste.openstack.org/show/l0pnKjevkajEy5YnZVJa/ :) | 22:27 |
| JoshNang | odd that admin couldn't see them though. maybe it's to avoid seeing potentially millions of objects | 22:27 |
| NobodyCam | that would be my guess | 22:28 |
| devananda | ugh, i see what ya'll meant about the state machine spec | 22:36 |
| devananda | i will finish reviewing the comments later. gotta step away now | 22:36 |
| *** alexpilotti has joined #openstack-ironic | 22:45 | |
| *** EmilienM has quit IRC | 22:47 | |
| *** EmilienM has joined #openstack-ironic | 22:47 | |
| *** achanda has joined #openstack-ironic | 22:50 | |
| NobodyCam | :) have a good holiday devananda | 22:52 |
| *** achanda has quit IRC | 22:55 | |
| zer0c00l | jroll: Do i really need cinder on devstack for testing ironic? | 22:57 |
| zer0c00l | cinder and heat | 22:57 |
| zer0c00l | And why do we need swift? | 22:57 |
| jroll | zer0c00l: you don't need cinder or heat, not sure where you got that idea | 22:58 |
| jroll | and swift is used for certain drivers, you may not need it | 22:58 |
| zer0c00l | okay | 23:04 |
| *** igordcard has quit IRC | 23:04 | |
| *** ryanpetrello has joined #openstack-ironic | 23:12 | |
| *** ryanpetrello has quit IRC | 23:19 | |
| *** andreykurilin_ has quit IRC | 23:22 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic-specs: Support Zapping of Nodes https://review.openstack.org/102685 | 23:49 |
| *** romcheg has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!