| Pradip_ | TheJulia: I am still stuck on that issue. I am not able to figure out why vm bare metal is not able to connect. I am getting confused. I checked the tftp server. it is running. the the vm seems to be not able to connect to it. any suggestions | 00:19 |
|---|---|---|
| TheJulia | Pradip_: so how did you create the VMs? | 00:20 |
| Pradip_ | using the adding the lines in the localrc file : IRONIC_VM_COUNT=3 IRONIC_VM_SSH_PORT=22 IRONIC_BAREMETAL_BASIC_OPS=True | 00:22 |
| openstackgerrit | Ruby Loo proposed openstack/ironic: use versionutils from oslo_utils https://review.openstack.org/195805 | 00:24 |
| TheJulia | is there a local firewall? | 00:25 |
| Pradip_ | you mean the host machin, right? | 00:25 |
| TheJulia | yes | 00:26 |
| Pradip_ | yes | 00:27 |
| Pradip_ | yes | 00:27 |
| Pradip_ | but it allows any tftp traffic | 00:28 |
| Pradip_ | one thing maybe the vm is not able to find any route to the host machine | 00:29 |
| Pradip_ | i checked the route table. it doesn't include any entry for the host machines ip | 00:30 |
| Pradip_ | network | 00:30 |
| *** ijw_ has quit IRC | 00:32 | |
| TheJulia | well, it needs to allow all sorts of traffic on the ?virbr0? interface | 00:33 |
| TheJulia | dhcp, tftp, interaction to the Ironic API | 00:33 |
| Pradip_ | ok thanks | 00:35 |
| Pradip_ | I will try that | 00:35 |
| *** rloo has quit IRC | 00:36 | |
| *** Pradip_ has quit IRC | 00:38 | |
| *** naohirot has joined #openstack-ironic | 00:43 | |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Adds port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 00:43 |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Add additional note dependent upon os_ironic_facts https://review.openstack.org/195111 | 00:57 |
| *** david-ly_ has joined #openstack-ironic | 01:06 | |
| *** UForgotten has quit IRC | 01:08 | |
| *** UForgotten has joined #openstack-ironic | 01:08 | |
| *** david-lyle has quit IRC | 01:09 | |
| *** ijw_ has joined #openstack-ironic | 01:32 | |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Adds port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 01:36 |
| *** boris-42 has quit IRC | 01:42 | |
| openstackgerrit | Merged openstack/ironic: Make task_manager logging more helpful https://review.openstack.org/194619 | 01:43 |
| *** mitchjameson has quit IRC | 01:48 | |
| *** radek_ has joined #openstack-ironic | 01:52 | |
| *** radek__ has quit IRC | 01:55 | |
| *** Haomeng has joined #openstack-ironic | 01:59 | |
| *** chenglch has joined #openstack-ironic | 02:02 | |
| *** d0ugal has quit IRC | 02:29 | |
| *** d0ugal has joined #openstack-ironic | 02:29 | |
| *** d0ugal is now known as Guest5335 | 02:30 | |
| *** rameshg87 has joined #openstack-ironic | 02:34 | |
| *** alexpilotti has quit IRC | 02:36 | |
| *** r-daneel has quit IRC | 02:39 | |
| *** rameshg87 has quit IRC | 02:54 | |
| openstackgerrit | Sinval Vieira Mendes Neto proposed openstack/ironic: Adds port creation passing the name of the node instead of the UUID of the node https://review.openstack.org/193439 | 02:58 |
| *** radek_ has quit IRC | 03:13 | |
| *** zz_natorious is now known as natorious | 03:21 | |
| *** boris-42 has joined #openstack-ironic | 03:22 | |
| *** rameshg87 has joined #openstack-ironic | 03:49 | |
| *** david-ly_ is now known as david-lyle | 03:51 | |
| openstackgerrit | Tan Lin proposed openstack/ironic: Add a new boot section 'trusted_boot' for PXE https://review.openstack.org/191661 | 03:52 |
| *** hblixt has joined #openstack-ironic | 03:57 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: Change dict() to {} https://review.openstack.org/195860 | 03:58 |
| *** blixt_ has joined #openstack-ironic | 03:59 | |
| *** hblixt has quit IRC | 03:59 | |
| *** rameshg871 has joined #openstack-ironic | 04:07 | |
| *** rameshg87 has quit IRC | 04:08 | |
| *** puranamr has joined #openstack-ironic | 04:20 | |
| *** Marga_ has joined #openstack-ironic | 04:20 | |
| *** ramineni has joined #openstack-ironic | 04:21 | |
| *** chenglch|2 has joined #openstack-ironic | 04:25 | |
| *** chenglch has quit IRC | 04:25 | |
| *** achanda has joined #openstack-ironic | 04:26 | |
| *** Marga_ has quit IRC | 04:30 | |
| *** Marga_ has joined #openstack-ironic | 04:31 | |
| *** puranamr has quit IRC | 04:32 | |
| *** yuikotakada has joined #openstack-ironic | 04:37 | |
| *** puranamr has joined #openstack-ironic | 04:53 | |
| *** mdbooth has quit IRC | 04:54 | |
| *** puranamr has quit IRC | 04:57 | |
| *** Nisha has joined #openstack-ironic | 05:00 | |
| *** achanda has quit IRC | 05:00 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC vendor passthru for soft reboot and NMI https://review.openstack.org/187082 | 05:17 |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Switch to oslo.service https://review.openstack.org/195008 | 05:19 |
| *** yog_ has joined #openstack-ironic | 05:19 | |
| *** dasm|afk is now known as dasm | 05:20 | |
| *** Sukhdev has joined #openstack-ironic | 05:21 | |
| *** jefabo has quit IRC | 05:32 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Switch to oslo.service https://review.openstack.org/195008 | 05:33 |
| *** Nisha_away has joined #openstack-ironic | 05:36 | |
| *** Nisha has quit IRC | 05:36 | |
| *** rameshg87 has joined #openstack-ironic | 05:38 | |
| *** yog_ has quit IRC | 05:39 | |
| *** rameshg871 has quit IRC | 05:39 | |
| *** mtanino has joined #openstack-ironic | 05:43 | |
| *** achanda has joined #openstack-ironic | 05:43 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add iRMC Virtual Media Deploy module for iRMC Driver https://review.openstack.org/151958 | 05:44 |
| *** stegranet has joined #openstack-ironic | 05:44 | |
| *** boris-42 has quit IRC | 05:52 | |
| *** chenglch|2 has quit IRC | 05:54 | |
| *** chenglch|2 has joined #openstack-ironic | 05:54 | |
| *** links has joined #openstack-ironic | 05:55 | |
| *** puranamr has joined #openstack-ironic | 05:58 | |
| *** Sukhdev has quit IRC | 05:58 | |
| *** Nisha_away has quit IRC | 05:59 | |
| *** mtanino is now known as mtanino_zzz | 06:00 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add localboot support for iscsi_irmc driver https://review.openstack.org/163738 | 06:06 |
| *** dasm is now known as dasm|brb | 06:06 | |
| openstackgerrit | Ramakrishnan G proposed stackforge/proliantutils: Check if virtual media is attached before ejecting https://review.openstack.org/195432 | 06:11 |
| *** radek_ has joined #openstack-ironic | 06:12 | |
| *** mtanino_zzz has quit IRC | 06:14 | |
| *** achanda_ has joined #openstack-ironic | 06:14 | |
| *** yog_ has joined #openstack-ironic | 06:15 | |
| *** achanda has quit IRC | 06:16 | |
| *** radek_ has quit IRC | 06:18 | |
| *** dasm|brb is now known as dasm | 06:18 | |
| *** achanda_ has quit IRC | 06:24 | |
| *** boris-42 has joined #openstack-ironic | 06:26 | |
| openstackgerrit | Merged stackforge/proliantutils: Check if virtual media is attached before ejecting https://review.openstack.org/195432 | 06:26 |
| *** dasm is now known as dasm|afk | 06:30 | |
| *** radek_ has joined #openstack-ironic | 06:32 | |
| *** Guest5335 is now known as d0ugal | 06:34 | |
| *** d0ugal has quit IRC | 06:34 | |
| *** d0ugal has joined #openstack-ironic | 06:34 | |
| openstackgerrit | Yuiko Takada proposed openstack/ironic-inspector: Migrate to oslo_db https://review.openstack.org/181905 | 06:38 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Improve warning message in conductor.utils.node_power_action() https://review.openstack.org/195893 | 06:42 |
| *** coolsvap|away is now known as coolsvap | 06:44 | |
| *** oomichi has joined #openstack-ironic | 06:52 | |
| *** puranamr has quit IRC | 06:54 | |
| *** coolsvap is now known as coolsvap|away | 06:56 | |
| *** oomichi has quit IRC | 07:01 | |
| *** ijw_ has quit IRC | 07:02 | |
| *** oomichi has joined #openstack-ironic | 07:03 | |
| *** ijw has joined #openstack-ironic | 07:03 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Replace self.assertEqual(None,*) to self.assertIsNone() https://review.openstack.org/195896 | 07:06 |
| *** puranamr has joined #openstack-ironic | 07:09 | |
| *** puranamr has quit IRC | 07:13 | |
| *** lazy_prince has joined #openstack-ironic | 07:14 | |
| *** blixt_ has quit IRC | 07:16 | |
| *** lsmola has joined #openstack-ironic | 07:21 | |
| *** jcoufal has joined #openstack-ironic | 07:25 | |
| *** killer_prince has joined #openstack-ironic | 07:26 | |
| *** lazy_prince has quit IRC | 07:29 | |
| *** achanda has joined #openstack-ironic | 07:32 | |
| *** dtantsur|afk is now known as dtantsur | 07:39 | |
| dtantsur | Morning folks! | 07:39 |
| dtantsur | harlowja, hey, did you suggest me to become a futurist core? I think I've missed that suggestion, sorry. Anyway, I'd like to. :) | 07:41 |
| dtantsur | and \o/ to the release! I'll try switching ironic soonish | 07:42 |
| *** dasm|afk is now known as dasm | 07:43 | |
| yuikotakada | dtantsur, o/ | 07:45 |
| dtantsur | o/ | 07:45 |
| *** oomichi has quit IRC | 07:48 | |
| *** athomas has joined #openstack-ironic | 07:49 | |
| Haomeng | dtantsur: morning | 07:53 |
| dtantsur | Haomeng, o/ | 07:53 |
| Haomeng | dtantsur: I have one finding - "2015-06-26 07:32:28.436 ERROR ironic.drivers.modules.agent [-] node bd5fff55-1a99-4b73-92c4-7a8cbd9b6eca command status errored: {u'message': u'Error downloading image.', u'codeImageDownloadError', u'details': u'Download of image id e18a422d-94d8-4389-b345-21be90b6f745 failed: Unable to write image to /tmp/e18a422d-94d8-4389-b345-21be90b6f745. Error: [left on devic | 07:53 |
| *** rameshg87 has quit IRC | 07:53 | |
| Haomeng | dtantsur: looks like we did not caculate the raw image size to schedule baremetal | 07:54 |
| *** yuikotakada has quit IRC | 07:54 | |
| dtantsur | Haomeng, hmm, I'm not sure how agent driver does it.. | 07:54 |
| *** yuikotakada has joined #openstack-ironic | 07:54 | |
| *** dtantsur is now known as dtantsur|brb | 07:56 | |
| Haomeng | dtantsur: my case is, the flavor is 10G disk, and create agent_ssh driver node to boot the image which is qcow2 which less then 10G, but after it is unziped, will be more then 10G, so can not download the image to local disk which 10G max size | 07:56 |
| dtantsur|brb | sorry, I have to run right now :( we can chat later (or you can catch Lucas who might know more) | 07:57 |
| Haomeng | dtantsur|brb: ok, np, thanks:) | 07:57 |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add whole disk image support for iscsi_irmc driver https://review.openstack.org/181771 | 08:03 |
| *** Marga_ has quit IRC | 08:04 | |
| *** lucasagomes has joined #openstack-ironic | 08:04 | |
| *** v92 has joined #openstack-ironic | 08:06 | |
| *** derekh has joined #openstack-ironic | 08:06 | |
| *** mdbooth has joined #openstack-ironic | 08:15 | |
| openstackgerrit | Naohiro Tamura proposed openstack/ironic: Add IPA support for iscsi_irmc driver https://review.openstack.org/181807 | 08:19 |
| *** yuanying has quit IRC | 08:19 | |
| *** dguerri` is now known as dguerri | 08:20 | |
| *** natorious is now known as zz_natorious | 08:22 | |
| *** gzhai1 has quit IRC | 08:25 | |
| *** rameshg87 has joined #openstack-ironic | 08:25 | |
| *** stegranet1 has joined #openstack-ironic | 08:29 | |
| *** stegranet has quit IRC | 08:31 | |
| *** stegranet1 is now known as stegranet | 08:31 | |
| *** romcheg has joined #openstack-ironic | 08:38 | |
| *** Marga_ has joined #openstack-ironic | 08:43 | |
| sambetts | Morning dtantsur|brb yuikotakada Haomeng o/ | 08:45 |
| yuikotakada | sambetts, o/ | 08:45 |
| *** stegranet has quit IRC | 08:45 | |
| *** athomas has quit IRC | 08:47 | |
| sambetts | yuikotakada: How are you? | 08:48 |
| yuikotakada | sambetts, fine :) how are you? | 08:50 |
| rameshg87 | morning lucasagomes sambetts yuikotakada dtantsur|brb Haomeng and anyone else | 08:50 |
| yuikotakada | rameshg87, g'morning | 08:50 |
| sambetts | yuikotakada: I'm pretty good thanks, glad its Friday though :) | 08:51 |
| sambetts | morning rameshg87 | 08:51 |
| yuikotakada | sambetts, yep, TGIF! | 08:51 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Clean nodes stuck in DEPLOYING state when ir-cond restarts https://review.openstack.org/194132 | 08:51 |
| lucasagomes | good ugt morning rameshg87 yuikotakada sambetts (-: | 08:52 |
| yuikotakada | lucasagomes, good ugt morning :) | 08:52 |
| *** lsmola has quit IRC | 08:53 | |
| sambetts | Hey lucasagomes o/ | 08:53 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Clean nodes stuck in DEPLOYING state when ir-cond restarts https://review.openstack.org/194132 | 08:55 |
| * lucasagomes forgot to save the file | 08:55 | |
| * sambetts has done that before too | 08:56 | |
| rameshg87 | lucasagomes: when you get some time, can you please check if you are okay with whatever we have discussed with regarding to update to raid spe | 08:59 |
| rameshg87 | *spec | 08:59 |
| rameshg87 | https://review.openstack.org/#/c/192936/2/specs/liberty/ironic-generic-raid-interface.rst | 08:59 |
| lucasagomes | rameshg87, ++ | 08:59 |
| *** stegranet has joined #openstack-ironic | 09:00 | |
| lucasagomes | rameshg87, +a | 09:01 |
| rameshg87 | \o/ | 09:03 |
| *** Marga_ has quit IRC | 09:04 | |
| openstackgerrit | Merged openstack/ironic-specs: Fixes for generic RAID interface https://review.openstack.org/192936 | 09:04 |
| *** athomas has joined #openstack-ironic | 09:09 | |
| *** husanu8 has joined #openstack-ironic | 09:09 | |
| *** pelix has joined #openstack-ironic | 09:14 | |
| TheJulia | good mornimg | 09:14 |
| *** bradjones has quit IRC | 09:15 | |
| *** bradjones has joined #openstack-ironic | 09:16 | |
| *** bradjones has quit IRC | 09:16 | |
| *** bradjones has joined #openstack-ironic | 09:16 | |
| sambetts | Morning TheJulia o/ | 09:18 |
| *** husanu8 has quit IRC | 09:18 | |
| yuikotakada | TheJulia, good morning | 09:19 |
| *** romcheg has quit IRC | 09:19 | |
| *** boris-42 has quit IRC | 09:22 | |
| *** ndipanov has joined #openstack-ironic | 09:22 | |
| *** husanu1 has joined #openstack-ironic | 09:23 | |
| *** romcheg has joined #openstack-ironic | 09:25 | |
| *** Haomeng|2 has joined #openstack-ironic | 09:27 | |
| *** husanu1 has quit IRC | 09:28 | |
| *** Haomeng has quit IRC | 09:30 | |
| lucasagomes | TheJulia, g morning | 09:36 |
| openstackgerrit | Merged openstack/ironic: Improve warning message in conductor.utils.node_power_action() https://review.openstack.org/195893 | 09:41 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/ironic: Clean nodes stuck in DEPLOYING state when ir-cond restarts https://review.openstack.org/194132 | 09:44 |
| *** athomas has quit IRC | 09:44 | |
| *** Marga_ has joined #openstack-ironic | 09:48 | |
| *** romcheg has quit IRC | 09:55 | |
| *** naohirot has quit IRC | 09:56 | |
| Haomeng|2 | lucasagomes, TheJulia: good morning:) | 09:59 |
| lucasagomes | Haomeng|2, morning | 10:00 |
| Haomeng|2 | lucasagomes: I have a question, agent driver will download image via glance.swift_temp_url and save to local /tmp/<image-id>, right | 10:00 |
| Haomeng|2 | lucasagomes: I understand the /tmp is in memory, not local host, because at that time, we did not create paration table | 10:00 |
| lucasagomes | Haomeng|2, hi there, I don't know agent drivers that much... but yes it will download it from swift_temp_url (or any other URL you may give) | 10:01 |
| lucasagomes | I don't know where it saves tho | 10:01 |
| * lucasagomes looks | 10:01 | |
| Haomeng|2 | lucasagomes: so if the image size is more than memory size, will raise such exception - ""2015-06-26 07:32:28.436 ERROR ironic.drivers.modules.agent [-] node bd5fff55-1a99-4b73-92c4-7a8cbd9b6eca command status errored: {u'message': u'Error downloading image.', u'codeImageDownloadError', u'details': u'Download of image id e18a422d-94d8-4389-b345-21be90b6f745 failed: Unable to write image to /tmp/e1 | 10:01 |
| Haomeng|2 | lucasagomes: ok | 10:01 |
| Haomeng|2 | lucasagomes: do you think who are agent expert, want to confirm, and how to get agent log | 10:02 |
| Haomeng|2 | lucasagomes: and debug with agent | 10:02 |
| lucasagomes | Haomeng|2, oh yeah... and it will also convert the image while writting to the disk right? (If it's qcow2 for e.g) | 10:02 |
| Haomeng|2 | lucasagomes: I can guess the way, just inject key by manually into agent image | 10:02 |
| lucasagomes | that may require extra ram | 10:02 |
| Haomeng|2 | lucasagomes: yes | 10:02 |
| lucasagomes | jroll, JayF JoshNang ^ | 10:02 |
| Haomeng|2 | lucasagomes: ok, got they are racespace experts:) | 10:02 |
| lucasagomes | yeah | 10:02 |
| Haomeng|2 | lucasagomes: thank you Luca, will check with them:) | 10:02 |
| Haomeng|2 | lucasagomes: this should be a BIG issue:) | 10:03 |
| lucasagomes | Haomeng|2, no problem | 10:03 |
| Haomeng|2 | lucasagomes: for BIG image:) | 10:03 |
| Haomeng|2 | lucasagomes: :) | 10:03 |
| *** chenglch|2 has quit IRC | 10:03 | |
| lucasagomes | yeah would be good to have a way around it, perhaps using the local disk to cache it | 10:03 |
| Haomeng|2 | lucasagomes: yes:) | 10:03 |
| lucasagomes | or write it to the disk while you download | 10:03 |
| Haomeng|2 | lucasagomes: good idea;) | 10:03 |
| Haomeng|2 | lucasagomes: yes, on air:) | 10:03 |
| lucasagomes | yeah | 10:03 |
| Haomeng|2 | lucasagomes: thanks:) | 10:03 |
| lucasagomes | np | 10:04 |
| *** Marga_ has quit IRC | 10:04 | |
| Haomeng|2 | lucasagomes: :) | 10:04 |
| Haomeng|2 | lucasagomes: nice day:) | 10:04 |
| lucasagomes | Haomeng|2, you too, enjoy the day over there! | 10:04 |
| Haomeng|2 | lucasagomes: :) | 10:04 |
| *** romcheg has joined #openstack-ironic | 10:06 | |
| openstackgerrit | Shuichiro MAKIGAKI proposed openstack/ironic: Add sequence diagrams for pxe_ipmi driver https://review.openstack.org/147087 | 10:06 |
| *** ijw has quit IRC | 10:07 | |
| *** yuikotakada has quit IRC | 10:08 | |
| *** athomas has joined #openstack-ironic | 10:13 | |
| *** e0ne has joined #openstack-ironic | 10:15 | |
| *** Marga_ has joined #openstack-ironic | 10:19 | |
| *** Marga_ has quit IRC | 10:22 | |
| *** Marga_ has joined #openstack-ironic | 10:23 | |
| *** stegranet has quit IRC | 10:31 | |
| *** MattMan has quit IRC | 10:33 | |
| *** MattMan has joined #openstack-ironic | 10:34 | |
| *** thrash|g0ne is now known as thrash | 10:34 | |
| *** links has quit IRC | 10:36 | |
| *** dtantsur|brb is now known as dtantsur | 10:42 | |
| dtantsur | sambetts, morning! | 10:43 |
| sambetts | Hey dtantsur o/ | 10:43 |
| dtantsur | TheJulia, morning you too | 10:43 |
| *** achanda_ has joined #openstack-ironic | 10:44 | |
| *** ijw has joined #openstack-ironic | 10:44 | |
| *** achanda has quit IRC | 10:47 | |
| *** ramineni has quit IRC | 10:47 | |
| *** ijw has quit IRC | 10:50 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Use oslo_utils.versionutils https://review.openstack.org/195987 | 10:50 |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Use oslo_log https://review.openstack.org/195989 | 10:52 |
| *** ijw has joined #openstack-ironic | 11:04 | |
| *** e0ne is now known as e0ne_ | 11:08 | |
| *** ijw has quit IRC | 11:08 | |
| *** yog_ has quit IRC | 11:10 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Use oslo_utils.versionutils https://review.openstack.org/195987 | 11:10 |
| *** lazy_prince has joined #openstack-ironic | 11:13 | |
| *** slagle_ is now known as slagle | 11:14 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add node fields for raid configuration https://review.openstack.org/155230 | 11:14 |
| *** killer_prince has quit IRC | 11:14 | |
| *** kbyrne has quit IRC | 11:16 | |
| *** Marga_ has quit IRC | 11:17 | |
| *** links has joined #openstack-ironic | 11:17 | |
| *** lazy_prince has quit IRC | 11:17 | |
| *** lazy_prince has joined #openstack-ironic | 11:18 | |
| *** e0ne_ has quit IRC | 11:18 | |
| *** links has quit IRC | 11:19 | |
| *** kbyrne has joined #openstack-ironic | 11:20 | |
| *** kbyrne has quit IRC | 11:21 | |
| *** kbyrne has joined #openstack-ironic | 11:21 | |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add node fields for raid configuration https://review.openstack.org/155230 | 11:22 |
| *** lazy_prince has quit IRC | 11:22 | |
| openstackgerrit | Sergey Vilgelm proposed openstack/ironic: Update ironic.conf.sample https://review.openstack.org/196002 | 11:27 |
| *** lucasagomes is now known as lucas-afk | 11:28 | |
| lucas-afk | folks, I will take this afternoon off | 11:28 |
| lucas-afk | I got a summer cold and I'm feeling quite shite heh | 11:28 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add RAIDInterface for RAID configuration https://review.openstack.org/196003 | 11:28 |
| lucas-afk | so will take a rest | 11:28 |
| rameshg87 | lucas-afk: have a great weekend | 11:28 |
| lucas-afk | enjoy the weekend everyone! | 11:29 |
| lucas-afk | rameshg87, you too | 11:29 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add RPCAPIs for RAID configuration https://review.openstack.org/196006 | 11:35 |
| openstackgerrit | Ramakrishnan G proposed openstack/ironic: Add APIs for RAID configuration https://review.openstack.org/196007 | 11:39 |
| *** e0ne has joined #openstack-ironic | 11:39 | |
| dtantsur | lucas-afk, get better! | 11:42 |
| *** bradjones has quit IRC | 11:42 | |
| *** ijw has joined #openstack-ironic | 11:44 | |
| *** bradjones has joined #openstack-ironic | 11:45 | |
| *** bradjones has quit IRC | 11:45 | |
| *** bradjones has joined #openstack-ironic | 11:45 | |
| * rameshg87 is done for the day | 11:46 | |
| rameshg87 | have a great weekend folks | 11:46 |
| *** rameshg87 has quit IRC | 11:46 | |
| openstackgerrit | Cameron.C proposed openstack/ironic: Requirement broken fix. https://review.openstack.org/196010 | 11:49 |
| *** ijw has quit IRC | 11:50 | |
| *** achanda_ has quit IRC | 11:58 | |
| *** stegranet has joined #openstack-ironic | 12:04 | |
| *** dguerri is now known as dguerri` | 12:10 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-ironicclient: Bump default retry timeout to match Nova's https://review.openstack.org/196020 | 12:15 |
| *** trown|outttypeww is now known as trown | 12:16 | |
| *** stegranet has quit IRC | 12:28 | |
| *** dguerri` is now known as dguerri | 12:28 | |
| *** stegranet has joined #openstack-ironic | 12:34 | |
| *** ijw has joined #openstack-ironic | 12:44 | |
| *** smoriya has quit IRC | 12:48 | |
| *** ijw has quit IRC | 12:50 | |
| *** puranamr has joined #openstack-ironic | 12:51 | |
| *** puranamr_ has joined #openstack-ironic | 12:53 | |
| *** puranamr has quit IRC | 12:53 | |
| *** Sukhdev has joined #openstack-ironic | 12:54 | |
| *** puranamr_ has quit IRC | 12:58 | |
| openstackgerrit | KATO Tomoyuki proposed openstack/ironic: changes and edit of the ironic install guide https://review.openstack.org/191900 | 13:06 |
| *** vdrok has quit IRC | 13:07 | |
| *** dasm is now known as dasm|afk | 13:08 | |
| *** Marga_ has joined #openstack-ironic | 13:09 | |
| *** vdrok has joined #openstack-ironic | 13:18 | |
| dtantsur | so folks, do you want to facepalm with me? https://bugs.launchpad.net/devstack/+bug/1469160 | 13:25 |
| openstack | Launchpad bug 1469160 in devstack "Cinder tries to start even if it wasn't requested - and fails" [Undecided,In progress] - Assigned to Dmitry Tantsur (divius) | 13:25 |
| *** dtantsur_ has joined #openstack-ironic | 13:27 | |
| *** dtantsur has quit IRC | 13:29 | |
| *** ifarkas has joined #openstack-ironic | 13:29 | |
| *** dtantsur_ has quit IRC | 13:32 | |
| *** dguerri is now known as dguerri` | 13:33 | |
| *** dtantsur has joined #openstack-ironic | 13:38 | |
| *** Marga_ has quit IRC | 13:39 | |
| *** ifarkas has quit IRC | 13:40 | |
| sambetts | dtantsur: oh dear... | 13:40 |
| dtantsur | :) | 13:40 |
| *** ijw has joined #openstack-ironic | 13:44 | |
| *** achanda has joined #openstack-ironic | 13:45 | |
| *** e0ne is now known as e0ne_ | 13:46 | |
| openstackgerrit | Dmitry Tantsur proposed openstack/ironic: Allow deleting even associated and active node in maintenance mode https://review.openstack.org/191736 | 13:47 |
| *** boris-42 has joined #openstack-ironic | 13:47 | |
| *** dguerri` is now known as dguerri | 13:48 | |
| *** dguerri is now known as dguerri` | 13:48 | |
| *** ijw has quit IRC | 13:50 | |
| *** dguerri` is now known as dguerri | 13:50 | |
| openstackgerrit | Zhenguo Niu proposed openstack/ironic: When boot option is not persisted, set boot on next power on https://review.openstack.org/177642 | 13:53 |
| *** e0ne_ is now known as e0ne | 13:56 | |
| NobodyCam | Good MOrning Ironicers | 13:56 |
| NobodyCam | and ... | 13:57 |
| NobodyCam | TGIF | 13:57 |
| NobodyCam | :) | 13:57 |
| sambetts | Morning NobodyCam o/ | 13:57 |
| *** ijw has joined #openstack-ironic | 13:58 | |
| openstackgerrit | Merged openstack/ironic-inspector: Deprecate authenticate opt in favor of auth_strategy https://review.openstack.org/195082 | 14:00 |
| *** achanda has quit IRC | 14:01 | |
| dtantsur | NobodyCam, morning, really TGIF! | 14:02 |
| NobodyCam | morning sambetts and dtantsur :) | 14:03 |
| NobodyCam | :) | 14:05 |
| *** krtaylor has quit IRC | 14:10 | |
| *** e0ne is now known as e0ne_ | 14:14 | |
| *** e0ne_ is now known as e0ne | 14:16 | |
| *** r-daneel has joined #openstack-ironic | 14:26 | |
| *** Sukhdev has quit IRC | 14:29 | |
| *** puranamr has joined #openstack-ironic | 14:31 | |
| *** alexpilotti has joined #openstack-ironic | 14:31 | |
| *** alexpilotti_ has joined #openstack-ironic | 14:35 | |
| *** lazy_prince has joined #openstack-ironic | 14:35 | |
| *** alexpilotti has quit IRC | 14:36 | |
| *** alexpilotti_ is now known as alexpilotti | 14:36 | |
| *** ijw has quit IRC | 14:44 | |
| *** krtaylor has joined #openstack-ironic | 14:46 | |
| *** lazy_prince has quit IRC | 14:46 | |
| openstackgerrit | Merged openstack/ironic: Replace self.assertEqual(None,*) to self.assertIsNone() https://review.openstack.org/195896 | 14:49 |
| openstackgerrit | Merged openstack/ironic: use versionutils from oslo_utils https://review.openstack.org/195805 | 14:59 |
| openstackgerrit | Merged openstack/bifrost: Add delete role for dynamic inventory testing https://review.openstack.org/193690 | 15:02 |
| openstackgerrit | Merged openstack/bifrost: Individual playbook for creating a test VM https://review.openstack.org/193691 | 15:02 |
| openstackgerrit | Merged openstack/bifrost: Update the dynamic test playbook https://review.openstack.org/193692 | 15:02 |
| openstackgerrit | Chris Krelle proposed openstack/bifrost: Follow up to 193690 https://review.openstack.org/196096 | 15:05 |
| *** Marga_ has joined #openstack-ironic | 15:06 | |
| *** syh has quit IRC | 15:10 | |
| jlvillal | Good morning Ironic | 15:24 |
| *** e0ne is now known as e0ne_ | 15:25 | |
| *** e0ne_ is now known as e0ne | 15:25 | |
| NobodyCam | morning jlvillal | 15:29 |
| NobodyCam | last day before your month away | 15:29 |
| dtantsur | jlvillal, morning | 15:30 |
| dtantsur | month away, mmmmm | 15:30 |
| dtantsur | :) | 15:30 |
| NobodyCam | :) | 15:30 |
| jlvillal | NobodyCam: Thanks. I will actually work Monday & Tuesday also | 15:30 |
| jlvillal | dtantsur: NobodyCam: Off to sunny Moscow :) | 15:30 |
| dtantsur | lol, good luck :) | 15:31 |
| *** achanda has joined #openstack-ironic | 15:31 | |
| jlvillal | dtantsur: Thanks | 15:32 |
| jlvillal | dtantsur: Just upgraded to Fedora 21. I just noticed when a command completes in my terminal and I have focus elsewhere, a little popup will happen. Kind of cool. It said 'Command completed: git pull --rebase' | 15:33 |
| * jlvillal assumes dtantsur uses Fedora ;) | 15:33 | |
| dtantsur | cool :) I'm using Fedora, but not the default terminal | 15:34 |
| dtantsur | F22 right now, to be precise | 15:34 |
| jlvillal | dtantsur: Oh yeah, it is F22. I was on F20, so I needed to upgrade. | 15:35 |
| dtantsur | right | 15:35 |
| jlvillal | I like to use Cinnamon | 15:35 |
| dtantsur | ++ | 15:35 |
| NobodyCam | jlvillal: Moscow.. wow... | 15:36 |
| dtantsur | it's a bit buggy sometimes, but still very cool | 15:36 |
| * jlvillal has not adapted yet to Gnome3 | 15:36 | |
| * dtantsur stopped trying after 2nd attempt | 15:36 | |
| jlvillal | NobodyCam: Should be fun :) | 15:36 |
| NobodyCam | :) ya | 15:36 |
| dtantsur | ok folks, I'm planning to start enjoying my weekend | 15:38 |
| NobodyCam | have a great time doing that | 15:38 |
| jlvillal | dtantsur: Woo hoo! :) | 15:38 |
| dtantsur | have a great trip, jlvillal! ping me on hangouts/whatever if you need some chat about moscow :) | 15:38 |
| jlvillal | dtantsur: Will do. I better add you to my circles :) | 15:39 |
| dtantsur | jlvillal, yeah, feel free to | 15:39 |
| dtantsur | g'night | 15:39 |
| *** dtantsur is now known as dtantsur|afk | 15:39 | |
| cinerama | good morning (gst) ironicers | 15:41 |
| jlvillal | cinerama: Good morning | 15:41 |
| cinerama | hi jlvillal how are you? | 15:41 |
| NobodyCam | cinerama: gst? | 15:42 |
| jlvillal | cinerama: Doing very well. Three more work days before my vacation. So I'm well :) | 15:42 |
| jlvillal | cinerama: I hope you are well too! | 15:42 |
| cinerama | hey - i was looking at https://bugs.launchpad.net/ironic/+bug/1398948 (tldr - confusing ipmitool output in debug mode) | 15:42 |
| openstack | Launchpad bug 1398948 in Ironic "ironic-conductor should not print out ipmitool CLI options when in debug mode" [Low,Triaged] - Assigned to Sandhya Balakrishnan (sandhya-balakrishnan) | 15:42 |
| TheJulia | goot morning cinerama | 15:42 |
| cinerama | jlvillal: COOL. i need another vacation already :) | 15:42 |
| jlvillal | :) | 15:42 |
| *** romcheg has quit IRC | 15:42 | |
| * jlvillal goes off to help the high school interns learn OpenStack... | 15:42 | |
| cinerama | anyway so the deal is that we run "ipmitool <options> -h" to confirm that we have certain options available | 15:43 |
| NobodyCam | ya | 15:43 |
| TheJulia | so... I think the context be off | 15:43 |
| jlvillal | cinerama: Is that me you are talking to? | 15:43 |
| *** hemna has joined #openstack-ironic | 15:43 | |
| TheJulia | reading the original submitted subject, the person is indicating that they don't get the config dump of what hte conductor loaded. When triaged, the subject was changed, I think because the first output was indicated. | 15:44 |
| cinerama | if it fails, it will return non-zero, otherwise it returns the help text. the problem is that the help text appearing right at the beginning of the log is confusing because people might think it is a sign something is broken | 15:44 |
| cinerama | jlvillal: not in particular, but feel free to weigh in :) | 15:44 |
| jlvillal | I agree with the people that it is confusing... | 15:44 |
| cinerama | TheJulia: yeah, i was reading the bug as being about the subject | 15:44 |
| * jlvillal now goes to find the interns... | 15:44 | |
| openstackgerrit | Merged openstack/bifrost: Follow up to 193690 https://review.openstack.org/196096 | 15:46 |
| cinerama | TheJulia: anyway so if we're fixing the bug in the subject, we'd either need the ability to suppress logging to debug in our execute wrapper, or maybe even just print a message telling people to ignore the ipmitool output | 15:47 |
| cinerama | TheJulia: we could also just check the ipmitool version and do stuff based on that but checking the actual flags might be a better test | 15:48 |
| JayF | Haomeng|2: lucas-afk: yeah; that's mostly a known limitation we don't hit b/c we run machines with absurd quantities of ram | 15:49 |
| devananda | g'morning, all (says he, as he gets on a conference call) | 15:55 |
| TheJulia | cinerama: ignore ipmitool and look at the ticket again :) | 15:56 |
| *** romcheg has joined #openstack-ironic | 15:58 | |
| TheJulia | good morning devananda | 15:58 |
| cinerama | TheJulia: so i have been going off what the ticket was triaged as, which I also feel is kind of a bug | 15:59 |
| TheJulia | I think we have two issues then | 16:01 |
| cinerama | perhaps we should readjust that bug, and then create a new one? | 16:01 |
| *** puranamr has quit IRC | 16:02 | |
| *** puranamr has joined #openstack-ironic | 16:02 | |
| *** achanda has quit IRC | 16:02 | |
| *** achanda has joined #openstack-ironic | 16:02 | |
| *** romainh has joined #openstack-ironic | 16:10 | |
| *** e0ne is now known as e0ne_ | 16:14 | |
| *** gabriel-bezerra has quit IRC | 16:15 | |
| *** e0ne_ has quit IRC | 16:15 | |
| *** stegranet has quit IRC | 16:16 | |
| *** romcheg has quit IRC | 16:17 | |
| *** derekh has quit IRC | 16:18 | |
| *** hblixt has joined #openstack-ironic | 16:18 | |
| *** achanda has quit IRC | 16:22 | |
| *** puranamr has quit IRC | 16:24 | |
| *** puranamr has joined #openstack-ironic | 16:25 | |
| cinerama | well, i've done that | 16:25 |
| *** romcheg has joined #openstack-ironic | 16:27 | |
| TheJulia | cinerama: :) | 16:31 |
| *** absubram has joined #openstack-ironic | 16:35 | |
| *** absubram has quit IRC | 16:36 | |
| *** absubram has joined #openstack-ironic | 16:37 | |
| *** jcoufal has quit IRC | 16:37 | |
| *** gabriel-bezerra has joined #openstack-ironic | 16:38 | |
| *** romainh has left #openstack-ironic | 16:43 | |
| *** e0ne has joined #openstack-ironic | 16:43 | |
| *** dguerri is now known as dguerri` | 16:44 | |
| *** e0ne has quit IRC | 16:45 | |
| *** achanda has joined #openstack-ironic | 16:47 | |
| *** mitchjameson has joined #openstack-ironic | 16:47 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Update Readme based on feedback from 193709 https://review.openstack.org/196146 | 16:55 |
| NobodyCam | cinerama: see that Mr Jenkins -1'd patch: 195759 | 17:00 |
| cinerama | NobodyCam: i'm just rebasing that now | 17:00 |
| NobodyCam | :) | 17:00 |
| cinerama | it was not a smooth rebase | 17:01 |
| NobodyCam | :( | 17:01 |
| openstackgerrit | Stephanie Miller proposed openstack/bifrost: Disambiguate the roles of ci_testing, ci_testing_zuul etc https://review.openstack.org/195759 | 17:03 |
| *** achanda has quit IRC | 17:07 | |
| *** achanda has joined #openstack-ironic | 17:07 | |
| *** achanda has quit IRC | 17:11 | |
| openstackgerrit | Merged openstack/bifrost: Add dynamic inventory description to README https://review.openstack.org/193709 | 17:12 |
| openstackgerrit | Merged openstack/bifrost: Addition of dynamic enrollment and deployment playbooks https://review.openstack.org/193737 | 17:12 |
| *** trown is now known as trown|lunch | 17:20 | |
| *** e0ne has joined #openstack-ironic | 17:22 | |
| devananda | for those that weren't following -- there was just a very lengthy discussion in -infra about versioning | 17:23 |
| harlowja | dtantsur|afk ok dokie, http://lists.openstack.org/pipermail/openstack-dev/2015-June/068118.html | 17:25 |
| *** EmilienM is now known as EmilienM|brb | 17:25 | |
| *** hblixt has quit IRC | 17:26 | |
| openstackgerrit | Merged openstack/bifrost: Update Readme based on feedback from 193709 https://review.openstack.org/196146 | 17:29 |
| *** ijw has joined #openstack-ironic | 17:30 | |
| NobodyCam | TheJulia: ImportError: No module named shade on http://logs.openstack.org/85/193885/6/gate/gate-bifrost/e0b7671/console.html ???? | 17:37 |
| TheJulia | ouch | 17:39 |
| TheJulia | I'm surprised that didn't... oh | 17:39 |
| TheJulia | That makes sense | 17:39 |
| TheJulia | way too much sense | 17:39 |
| TheJulia | I know what to do, need to add some more smartness to the inventory | 17:41 |
| *** ndipanov has quit IRC | 17:42 | |
| *** lucas-afk has quit IRC | 17:44 | |
| *** zz_natorious is now known as natorious | 17:45 | |
| TheJulia | NobodyCam: its because the addition of shade becoming a dependency got added after the fact, and realistically it needs the extra logic to allow it to still run but disable polling ironic if shade can't load. | 17:48 |
| TheJulia | s/after the fact/before this change that failed/ | 17:48 |
| TheJulia | since it was not in the test path :( | 17:48 |
| *** Marga_ has quit IRC | 17:51 | |
| openstackgerrit | Clif Houck proposed openstack/ironic: Add support for image caching for the agent driver https://review.openstack.org/161832 | 17:54 |
| *** e0ne is now known as e0ne_ | 17:55 | |
| *** stegranet has joined #openstack-ironic | 17:55 | |
| *** lucas-afk has joined #openstack-ironic | 17:56 | |
| *** pelix has quit IRC | 17:58 | |
| * devananda takes a break to find food | 17:59 | |
| *** achanda has joined #openstack-ironic | 17:59 | |
| NobodyCam | mmmm food | 17:59 |
| NobodyCam | :) | 17:59 |
| *** Marga_ has joined #openstack-ironic | 18:02 | |
| *** stegranet has quit IRC | 18:03 | |
| *** EmilienM|brb is now known as EmilienM | 18:04 | |
| *** Marga_ has quit IRC | 18:07 | |
| *** Marga_ has joined #openstack-ironic | 18:07 | |
| openstackgerrit | Julia Kreger proposed openstack/bifrost: Add logic to allow bifrost_inventory to run without shade https://review.openstack.org/196170 | 18:09 |
| TheJulia | NobodyCam: ^^^ | 18:09 |
| *** e0ne_ is now known as e0ne | 18:10 | |
| TheJulia | w/r/t 193885 | 18:10 |
| * NobodyCam looks | 18:13 | |
| *** Marga_ has quit IRC | 18:14 | |
| *** Marga_ has joined #openstack-ironic | 18:14 | |
| *** Marga_ has quit IRC | 18:15 | |
| *** Marga_ has joined #openstack-ironic | 18:15 | |
| *** achanda has quit IRC | 18:16 | |
| *** athomas has quit IRC | 18:17 | |
| *** Pradip_ has joined #openstack-ironic | 18:18 | |
| *** trown|lunch is now known as trown | 18:22 | |
| *** puranamr has quit IRC | 18:22 | |
| *** puranamr has joined #openstack-ironic | 18:26 | |
| *** arif-ali has quit IRC | 18:31 | |
| *** arif-ali has joined #openstack-ironic | 18:36 | |
| *** arif-ali has quit IRC | 18:37 | |
| *** arif-ali has joined #openstack-ironic | 18:38 | |
| *** puranamr has quit IRC | 18:40 | |
| *** stegranet has joined #openstack-ironic | 18:40 | |
| *** puranamr has joined #openstack-ironic | 18:41 | |
| *** e0ne has quit IRC | 18:42 | |
| devananda | for those who didn't follow the discussion in -infra, I believe I've identified a bad / broken behavior in our API, though I also anticipate some folks disagreeing with me | 18:53 |
| *** puranamr has quit IRC | 18:54 | |
| *** e0ne has joined #openstack-ironic | 18:54 | |
| devananda | if a node is in a new state (eg, manageable) and an older client requests a list of nodes (eg, GET /v1/nodes) , that node shouldn't be returned -- because it is not in a state which the older client has any way to reason about or manipulate | 18:54 |
| devananda | anyone want to discuss now? if not, I'll hit the ML | 18:55 |
| NobodyCam | devananda: would that imply that once spoken to a node must always be talked to with the same micro version. and if so we'll need to track last version | 18:56 |
| *** puranamr has joined #openstack-ironic | 18:56 | |
| devananda | NobodyCam: nope | 18:56 |
| Pradip_ | hi, is there any limitation of using openvswitch in ironic Q_ML2_PLUGIN_MECHANISM_DRIVERS=openvswitch? | 18:56 |
| JayF | devananda: yuck | 18:56 |
| JayF | devananda: All possible solutions seem gross too, potentially | 18:56 |
| devananda | NobodyCam: a node in state "available" is fully usable by v1.1 | 18:56 |
| openstackgerrit | Merged openstack/bifrost: Add logic to allow bifrost_inventory to run without shade https://review.openstack.org/196170 | 18:56 |
| devananda | JayF: indeed. but this seems to be the least gross | 18:56 |
| JayF | like, from a dev/client standpoint, sure | 18:56 |
| devananda | Pradip_: huh? Ironic does not talk to openvswitch -- you may be looking for "neutron" | 18:57 |
| JayF | but we're going to scare the shit out of some operators when they list their nodes and it's missing | 18:57 |
| devananda | JayF: totally agree | 18:57 |
| devananda | JayF: we could introduce a "state unknown" state -- but that's just making the problem worse | 18:57 |
| JayF | yeah; maybe | 18:57 |
| JayF | I'd almost say reflect it as "ERROR" or something similar | 18:57 |
| TheJulia | but then it is not really in error | 18:58 |
| TheJulia | its a client and user workflow issue at that point | 18:58 |
| devananda | JayF: as it is, a v1.1 client can't /do/ certain things with the node-in-manageable-state -- but it can do other things | 18:58 |
| NobodyCam | ya this is actually a tuff one | 18:58 |
| JayF | from a perspective of v1 states, it's closer to error than anything else | 18:58 |
| devananda | so it's a very poor experience for said client | 18:58 |
| JayF | I'm just not hugely a fan of making those nodes disappear | 18:58 |
| JayF | I don't have a better alternative though | 18:58 |
| devananda | JayF: heh | 18:58 |
| TheJulia | make them appear in maintenance state? | 18:58 |
| NobodyCam | i'd almost say inaccessible | 18:59 |
| devananda | TheJulia: that's also misleading, but in a worse way | 18:59 |
| TheJulia | maintenance_reason set as "Unable to act on node, upgrade client" | 18:59 |
| JayF | TheJulia: I was kinda thinking that, yeah, but because maint still exists I think devananda is right about it being worse | 18:59 |
| NobodyCam | maintenance seems wrong because of what that implys | 18:59 |
| TheJulia | oh eyah | 18:59 |
| JayF | TheJulia: what's maintenance reason? | 18:59 |
| devananda | TheJulia: because a v1.1 client could move them out of maintenance mode if they were ACTUALLy in maintenance mode | 18:59 |
| TheJulia | yup | 19:00 |
| devananda | i'd rather hide them from a full list, and if a v1.1 client tries to access the node directly by UUID, it gets an incompatible request response | 19:00 |
| devananda | rather than a "node not found" | 19:00 |
| devananda | so the lcient knows tha tsomething is there | 19:00 |
| TheJulia | well, the api would have to basially reject the cleint for any action because it's version is too low to act upon the node | 19:00 |
| TheJulia | which means, as soon as someone upgrades their server, they will possibly hit a situation where they can add nodes, but then nothing else until they have a matching client | 19:00 |
| JayF | TheJulia: no, that's not the case I don't think | 19:01 |
| *** ijw__ has joined #openstack-ironic | 19:01 | |
| devananda | eg, version:1.1 GET /nodes/ ===> return a list of nodes which could be acted upon by said client | 19:01 |
| JayF | TheJulia: because you don't get into any of the states that don't translate without positive action; I *think* | 19:01 |
| NobodyCam | or worse add a buch of node and then have htem disapear because the client cant see the enrolled state | 19:01 |
| JayF | devananda: we have to communicate somehow to the person making the API call that we excluded incompatible items from the list | 19:01 |
| devananda | version:1.1 GET /nodes/NNN ===> HTTP 409 incompatible request | 19:01 |
| JayF | NobodyCam: ++ | 19:01 |
| JayF | devananda: I almost prefer that tbh | 19:01 |
| devananda | NobodyCam: that won't happen | 19:01 |
| devananda | so | 19:01 |
| devananda | an older client adding nodes will ALWAYS result in nodes that client can both see and interact with. | 19:02 |
| JayF | devananda: but all my clients are "latest" and we embrace the break here so I might not be a valid user in this idea :) | 19:02 |
| devananda | if we don't do that ^ we should quit our jobs and farm goats | 19:02 |
| devananda | JayF: you're a special snowflake :) | 19:02 |
| NobodyCam | devananda: (super edge case but) two clients one new one old | 19:02 |
| devananda | NobodyCam: not edge case at all | 19:03 |
| JayF | yeah, that's not an edge case | 19:03 |
| JayF | that's called having too many virtualenvs | 19:03 |
| JayF | hehe | 19:03 |
| devananda | NobodyCam: that's the inverse of the main situation (one client, two clouds). both are completely realistic | 19:03 |
| Pradip_ | devananda: sorry wrong question. I mean I am trying to config devstack with Ironic. I am not able to make the pxe_boot work. the ironic nodes are not able to reach the sftp server. So there must be something wrong with the networking. I have the line Q_ML2_PLUGIN_MECHANISM_DRIVERS=openvswitch in the localrc file. So, I am wondering whether that may be the reason? | 19:03 |
| NobodyCam | new client adds node and leaves in enrolled | 19:03 |
| JayF | Pradip_: That's a configuration value for Neutron; not for Ironic. | 19:03 |
| devananda | Pradip_: ahh, questoin makes sense now. however, i do not know the answer off hand, sorry | 19:04 |
| *** ijw has quit IRC | 19:04 | |
| Pradip_ | devananda: Okay that fine. thanks | 19:04 |
| devananda | NobodyCam: client A adds node, puts it in manageable state. client B is using v1.1 and doesn't see the node at all | 19:05 |
| devananda | NobodyCam: client A moves node to available state. now client B can see it! | 19:05 |
| Pradip_ | JayF: ok. thanks | 19:05 |
| devananda | NobodyCam: later on, client A moves it back to manageable state. now client B can't see it any more. | 19:05 |
| devananda | if client B was maintaining a local cache of resource UUIDs, and polling them directly, it's going to start getting HTTP 409 errors for that node | 19:06 |
| JayF | devananda: I know this is a problem on paper; have we seen instances of people actively broken by returning "unknown" (to client) states from a node list? | 19:06 |
| devananda | which tell it "the resource you requested is incompatible with the version you specified" | 19:06 |
| devananda | and I think that's quite reasonable | 19:06 |
| devananda | JayF: yup | 19:06 |
| devananda | JayF: both nova and inspector | 19:07 |
| JayF | I'm nor surprised by that; but it makes sense | 19:07 |
| devananda | JayF: and there's a patch up from TheJulia to avoid a similar thing breaking shade | 19:07 |
| NobodyCam | my fear is a large compaany that has automated actions, run from some server, and then real humans on their own workstations.. humans are liklly to upgrade their cleint while the automated tools bit rot off in the conor | 19:07 |
| * TheJulia feels bad for starting all this discussion with a patch, even though it is actually really good | 19:08 | |
| NobodyCam | * lots of type'ohs | 19:08 |
| NobodyCam | TheJulia: no please do not feel bad | 19:08 |
| NobodyCam | much needed imo | 19:08 |
| *** harlowja has quit IRC | 19:09 | |
| TheJulia | NobodyCam: or the inverse, user doesn't know but the management systems at the company auto-upgrade the server at some point | 19:09 |
| NobodyCam | TheJulia: that too | 19:09 |
| *** ijw__ has quit IRC | 19:09 | |
| NobodyCam | but I also don't have a clear solution | 19:10 |
| NobodyCam | so I'm playing dev(il's) advocate | 19:10 |
| TheJulia | It is almost likely we need an available concept that could deny access based on protocol version, or xyz is in someone elses's pool of nodes that you don't have access to... but thats just makes the problem worse in the short term | 19:16 |
| *** r-daneel has quit IRC | 19:17 | |
| TheJulia | not returning the node in the list based upon the ver if the state, or whatever breaking change that has been to that point, would effectively be the same thing though. | 19:18 |
| NobodyCam | oh hat if we return a 426 | 19:18 |
| JayF | I just come back to; if I did node list on two machines and got different results | 19:19 |
| JayF | I don't think it's natural to think "of course, my client is too old" | 19:19 |
| JayF | we're setting up the pit trap for operators | 19:19 |
| JayF | or, maybe, we already have | 19:19 |
| TheJulia | the trap is already there, its how we navigate around or build a bridge over it | 19:20 |
| JayF | I think dropping nodes from list | 19:20 |
| JayF | is building a bridge with a trapdoor | 19:20 |
| *** puranamr has quit IRC | 19:20 | |
| TheJulia | so 426 with the client would make sense from a lets fail fast standpoint. Just people really are going to have to stop allowing "latest" in their api client usage if they have code behind it | 19:21 |
| *** harlowja has joined #openstack-ironic | 19:22 | |
| TheJulia | JayF: absolutely with a trap door | 19:22 |
| TheJulia | the nodes will be crossing the bridge... and if your version is too low, they silently fall through the trap door | 19:24 |
| NobodyCam | i'm not super found of having filter code in somthing like node list that checking state.. but.. | 19:26 |
| *** alexpilotti has quit IRC | 19:26 | |
| TheJulia | I don't think i've looked at that code, but I could envision being very simple... and turning narly as time went on | 19:27 |
| devananda | I think we need to also pin the client, not bump the client version at every server version change | 19:28 |
| TheJulia | because upon each breaking change, the conditional logic would have to go in | 19:28 |
| NobodyCam | TheJulia: its the mantenance as time goes on that scarry to /me | 19:28 |
| devananda | also, we did not think about the long term maintenance of this in our API code | 19:28 |
| devananda | yea | 19:28 |
| devananda | it's one thing to hide a new REST resource | 19:28 |
| NobodyCam | nova uses a min and max api version | 19:28 |
| NobodyCam | ya | 19:29 |
| devananda | it's entirely different when we change properties/characteristics/behaviors of existing resources | 19:29 |
| NobodyCam | yes! | 19:29 |
| devananda | also - cleaning is badly broken for me still | 19:29 |
| devananda | i can't delete a node, or move a node from manageable->available, without manual DB updates | 19:30 |
| devananda | and it's all cleaning's fault! | 19:30 |
| devananda | AND I can't avoid that by using an older client | 19:30 |
| JoshNang | :( | 19:30 |
| NobodyCam | devananda: do you have enough info to file a bug? | 19:30 |
| devananda | even an older client calling "delete" on a node it provisioned -- well, now that node goes through cleaning | 19:31 |
| NobodyCam | s/info/details/ | 19:31 |
| devananda | NobodyCam: the node gets stuck in cleaning every time. | 19:31 |
| *** ijw has joined #openstack-ironic | 19:31 | |
| NobodyCam | :( | 19:32 |
| devananda | i'm sure something is wrong with my local network - who knows - it's not imporant why. here's how to reproduce it: | 19:32 |
| devananda | PUT {target_state: 'active'} /v1/nodes/NNNN/states/provision | 19:32 |
| devananda | <wait> | 19:32 |
| devananda | look, the node is in "wait callback" | 19:32 |
| devananda | PUT {target-state: 'deleted'} /v1/nodes/NNN/states/provision | 19:32 |
| devananda | <wait> | 19:33 |
| devananda | huh ... it's stuck in cleaning and now I cant do anything | 19:33 |
| devananda | doesn't matter how old a client I use | 19:33 |
| devananda | except that my old client wouldn't have even known what "cleaning" was | 19:33 |
| devananda | :( | 19:33 |
| JoshNang | do you have conductor logs? it sounds like its waiting for the agent to boot and callback | 19:35 |
| TheJulia | :( | 19:35 |
| *** ijw has quit IRC | 19:36 | |
| JoshNang | especially if the node doesn't have clean_steps in driver_internal_info or a clean_step other than {} | 19:36 |
| devananda | JoshNang: yes, it is | 19:36 |
| *** e0ne has quit IRC | 19:36 | |
| devananda | JoshNang: i think the root cause is my NUC isn't sending a DHCP BOOTP request. | 19:36 |
| devananda | JoshNang: but regardless of that, the API is broken | 19:37 |
| devananda | so there's some hardware issue -- great, that happens all the time | 19:37 |
| devananda | but I can't fix it from the API :( | 19:37 |
| devananda | s/fix/work around/ | 19:37 |
| JoshNang | :( i've got a patch started to add 'fail' as a verb for cleaning. | 19:38 |
| JoshNang | i'd probably be useful for a few other *ing states | 19:38 |
| devananda | yah | 19:39 |
| devananda | JoshNang: is that patch up? | 19:40 |
| JoshNang | devananda: not yet. let me see if i can get it finished today | 19:41 |
| *** achanda has joined #openstack-ironic | 19:41 | |
| *** Marga_ has quit IRC | 19:46 | |
| *** Marga_ has joined #openstack-ironic | 19:48 | |
| openstackgerrit | Merged openstack/bifrost: Change Bifrost over to using dynamic inventory https://review.openstack.org/193885 | 19:49 |
| TheJulia | devananda: So your just not seeing any dhcp requests on the wire with your nuc at all when it goes to boot back up? | 19:51 |
| *** Marga_ has quit IRC | 19:53 | |
| devananda | TheJulia: yea, there's some interaction in my driver not sending the right AMT command to trigger it | 19:55 |
| TheJulia | :( | 19:55 |
| devananda | TheJulia: the hardware works -- i can manually make it net boot -- i'm not sure what changed to cause this | 19:55 |
| devananda | dont have the brain today to mess with it -- instead i may go write a new revision of our API version spec | 19:56 |
| devananda | after i have lunch and a drink :) | 19:56 |
| JayF | "a" drink? | 19:57 |
| JayF | I think new revision of API version spec is worth a fifth at a minimum | 19:58 |
| JayF | heh | 19:58 |
| *** achanda has quit IRC | 19:58 | |
| *** ijw has joined #openstack-ironic | 19:58 | |
| * TheJulia seconds JayF's statement | 19:58 | |
| *** e0ne has joined #openstack-ironic | 20:03 | |
| *** zhenguo has quit IRC | 20:03 | |
| devananda | hah | 20:05 |
| * NobodyCam orders a round of shots... | 20:05 | |
| *** romcheg1 has joined #openstack-ironic | 20:06 | |
| *** Marga_ has joined #openstack-ironic | 20:06 | |
| *** romcheg has quit IRC | 20:06 | |
| devananda | wow, it's 1pm | 20:08 |
| devananda | bbiab | 20:08 |
| *** Marga_ has quit IRC | 20:10 | |
| *** Marga_ has joined #openstack-ironic | 20:10 | |
| openstackgerrit | Josh Gachnang proposed openstack/ironic: Allow 'fail' verb in the API https://review.openstack.org/196229 | 20:14 |
| *** radek_ has quit IRC | 20:14 | |
| *** krtaylor has quit IRC | 20:21 | |
| *** Marga_ has quit IRC | 20:22 | |
| *** Marga_ has joined #openstack-ironic | 20:23 | |
| *** e0ne has quit IRC | 20:24 | |
| *** hemna has quit IRC | 20:25 | |
| JoshNang | devananda: ^. i added some comments in the review, because i think there's a few ways we can handle this, and wasn't sure which way to go | 20:25 |
| *** meghal has joined #openstack-ironic | 20:30 | |
| meghal | hello…I was wondering as to what is recommended way for achieving HA with nova-compute ? | 20:32 |
| *** ijw__ has joined #openstack-ironic | 20:32 | |
| meghal | does it work well with ironic if nova-compute goes down on one node and comes up on another node using elect-leader or so ? | 20:32 |
| meghal | or is there any effort going on in this direction ? | 20:33 |
| NobodyCam | meghal: have you seen the cluster compute manager? | 20:33 |
| NobodyCam | https://github.com/openstack/ironic/tree/master/ironic/nova | 20:34 |
| *** ijw has quit IRC | 20:35 | |
| meghal | NobodyCam: I am using juno as of now, is this part of kilo or still in master ? | 20:35 |
| meghal | NobodyCam: looking into what it does | 20:35 |
| meghal | NobodyCam: ok…found it…it is under ironic and not in nova…that is where I got confused | 20:38 |
| *** puranamr has joined #openstack-ironic | 20:38 | |
| *** e0ne has joined #openstack-ironic | 20:39 | |
| NobodyCam | oh ya I pasted a link ^^ | 20:41 |
| meghal | NobodyCam: yep…did see that later…thanks | 20:43 |
| NobodyCam | :) | 20:43 |
| *** puranamr has quit IRC | 20:43 | |
| *** puranamr has joined #openstack-ironic | 20:43 | |
| meghal | NobodyCam: if I understand correctly, if nova-compute is configured to use clustered compute manager, if 1 nova-compute goes down and other joins, it will not update any of the active ironic nodes ? | 20:43 |
| *** e0ne has quit IRC | 20:44 | |
| NobodyCam | meghal: that was discussed at the last summit. I think this: https://etherpad.openstack.org/p/liberty-ironic-nova-virt-driver is the most up to date info I have | 20:47 |
| meghal | NobodyCam: going through that doc. Currently, what does clustered compute manager solve ? | 20:48 |
| NobodyCam | > 1 nova compute process | 20:49 |
| NobodyCam | one min... brb | 20:49 |
| *** Marga_ has quit IRC | 20:51 | |
| *** Marga_ has joined #openstack-ironic | 20:54 | |
| *** ijw__ has quit IRC | 20:58 | |
| *** Marga_ has quit IRC | 20:59 | |
| *** Marga_ has joined #openstack-ironic | 21:01 | |
| *** puranamr has quit IRC | 21:01 | |
| * NobodyCam is back | 21:02 | |
| *** puranamr has joined #openstack-ironic | 21:02 | |
| *** krtaylor has joined #openstack-ironic | 21:05 | |
| *** puranamr has quit IRC | 21:06 | |
| meghal | NobodyCam: so clusteredcomputemanager makes sure that instances are not deleted when new nova-compute comes up…right ? | 21:06 |
| *** puranamr has joined #openstack-ironic | 21:07 | |
| NobodyCam | meghal: I have not tested that expressly but would say defently maybe | 21:07 |
| *** puranamr has quit IRC | 21:09 | |
| *** romainh has joined #openstack-ironic | 21:11 | |
| *** puranamr has joined #openstack-ironic | 21:11 | |
| *** mitchjameson has quit IRC | 21:11 | |
| openstackgerrit | John L. Villalovos proposed openstack/ironic: Refactor check_allow_management_verbs https://review.openstack.org/196251 | 21:12 |
| jlvillal | JoshNang, What do you think of ^^^ Asking because your patch touched that area. | 21:13 |
| *** mitchjameson has joined #openstack-ironic | 21:16 | |
| *** puranamr has quit IRC | 21:23 | |
| *** harlowja has quit IRC | 21:23 | |
| *** puranamr has joined #openstack-ironic | 21:24 | |
| *** openstack has joined #openstack-ironic | 21:26 | |
| openstackgerrit | Devananda van der Veen proposed openstack/ironic-specs: Create 'devref' folder and move version specs to it https://review.openstack.org/196254 | 21:26 |
| *** ijw has joined #openstack-ironic | 21:26 | |
| *** mitchjameson has joined #openstack-ironic | 21:29 | |
| *** meghal has quit IRC | 21:32 | |
| *** meghal has joined #openstack-ironic | 21:32 | |
| *** puranamr has quit IRC | 21:35 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Log configuration options on ironic-conductor startup https://review.openstack.org/196256 | 21:35 |
| *** harlowja has joined #openstack-ironic | 21:35 | |
| *** hemna has joined #openstack-ironic | 21:38 | |
| *** puranamr has joined #openstack-ironic | 21:48 | |
| *** Marga_ has quit IRC | 21:55 | |
| *** rwsu has quit IRC | 21:56 | |
| *** ijw has quit IRC | 21:59 | |
| *** hemna_ has joined #openstack-ironic | 22:01 | |
| *** hemna has quit IRC | 22:02 | |
| *** thrash is now known as thrash|g0ne | 22:04 | |
| *** Marga_ has joined #openstack-ironic | 22:16 | |
| *** Marga_ has quit IRC | 22:21 | |
| *** stegranet has quit IRC | 22:21 | |
| *** lucas-afk has quit IRC | 22:22 | |
| openstackgerrit | Stephanie Miller proposed openstack/ironic: Explain ipmitool debug log output https://review.openstack.org/196267 | 22:23 |
| *** Marga_ has joined #openstack-ironic | 22:30 | |
| *** meghal has quit IRC | 22:36 | |
| *** meghal has joined #openstack-ironic | 22:36 | |
| *** meghal has quit IRC | 22:45 | |
| *** absubram has quit IRC | 22:48 | |
| *** meghal has joined #openstack-ironic | 22:52 | |
| jlvillal | cinerama, I wonder if ^^^ should really be logging the output. I was thinking of subprocess.check_call() But not sure. | 22:56 |
| cinerama | jlvillal: heh yeah that is kind of a conversation starter as much as anything. so we run it via our convenience wrapper for execute now | 22:59 |
| cinerama | which is going to output to LOG.debug | 22:59 |
| *** romainh has left #openstack-ironic | 23:01 | |
| cinerama | sorry, that should be "oslo's execute" | 23:01 |
| jlvillal | cinerama, Yeah. I wasn't sure if we need the output. | 23:01 |
| jlvillal | As it seems to run ipmitool and try a command. If a non-zero value is returned we assume it is not supported. | 23:01 |
| cinerama | i personally think we don't but others might disagree | 23:01 |
| cinerama | so i started with a less-invasive approach to "test the waters" | 23:02 |
| jlvillal | Okay. I would lean towards less output. Maybe log each failure in the log and not the entire output. | 23:02 |
| jlvillal | But others might hate that idea :) | 23:02 |
| devananda | I believe that ^ is exactly what I just posted on the review :) | 23:04 |
| jlvillal | Cool :) | 23:10 |
| *** Marga_ has quit IRC | 23:17 | |
| *** ekarlso has quit IRC | 23:18 | |
| *** ekarlso has joined #openstack-ironic | 23:18 | |
| *** puranamr has quit IRC | 23:36 | |
| *** puranamr has joined #openstack-ironic | 23:50 | |
| *** romcheg1 has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!