NobodyCam | adam_g: it sure could :) | 00:02 |
---|---|---|
*** penick has joined #openstack-ironic | 00:02 | |
*** bensonmiller has quit IRC | 00:04 | |
jroll | NobodyCam: heh, np | 00:17 |
NobodyCam | adam_g: like this: http://paste.openstack.org/show/tOVdkOaaVu2sD49utrla/ | 00:22 |
NobodyCam | hey jroll | 00:22 |
* jroll ducks | 00:22 | |
jroll | what's up? | 00:22 |
NobodyCam | huh | 00:23 |
NobodyCam | :-p | 00:23 |
jroll | oh, thought you needed something :P | 00:23 |
NobodyCam | its after 5 | 00:23 |
NobodyCam | nope | 00:23 |
jroll | right, beer time | 00:24 |
* jroll goes home | 00:24 | |
jroll | have a good night :) | 00:24 |
NobodyCam | have a good night jroll :) | 00:24 |
*** dlaube has quit IRC | 00:30 | |
*** aswadr has joined #openstack-ironic | 00:31 | |
*** tatyana has joined #openstack-ironic | 00:32 | |
*** mikedillion has quit IRC | 00:33 | |
*** ChuckC has quit IRC | 00:41 | |
*** nosnos has joined #openstack-ironic | 00:59 | |
*** harlowja is now known as harlowja_away | 01:00 | |
*** penick has quit IRC | 01:05 | |
openstackgerrit | Motohiro Otsuka proposed a change to openstack/ironic: Set ssh_virt_type to vmware https://review.openstack.org/116151 | 01:19 |
openstackgerrit | Motohiro Otsuka proposed a change to openstack/ironic: Add boot_device support for vbox https://review.openstack.org/116152 | 01:19 |
*** harlowja_away is now known as harlowja | 01:40 | |
*** penick has joined #openstack-ironic | 01:44 | |
*** rloo_afk has quit IRC | 01:52 | |
*** penick has quit IRC | 01:54 | |
*** nosnos has quit IRC | 02:03 | |
*** nosnos has joined #openstack-ironic | 02:04 | |
*** nosnos_ has joined #openstack-ironic | 02:08 | |
*** nosnos has quit IRC | 02:08 | |
*** Poornima has joined #openstack-ironic | 02:30 | |
*** rushiagr_away is now known as rushiagr | 02:37 | |
*** harlowja is now known as harlowja_away | 02:49 | |
*** nosnos_ has quit IRC | 02:56 | |
*** yuanying has quit IRC | 03:08 | |
*** krtaylor has joined #openstack-ironic | 03:21 | |
*** yuanying has joined #openstack-ironic | 03:24 | |
openstackgerrit | Ruby Loo proposed a change to openstack/ironic: properly format user guide in RST https://review.openstack.org/116168 | 03:31 |
*** Poornima has quit IRC | 03:38 | |
*** nosnos has joined #openstack-ironic | 03:40 | |
openstackgerrit | Motohiro Otsuka proposed a change to openstack/ironic: Add boot_device support for vbox https://review.openstack.org/116152 | 03:40 |
openstackgerrit | Motohiro Otsuka proposed a change to openstack/ironic: Set ssh_virt_type to vmware https://review.openstack.org/116151 | 03:40 |
*** tatyana has quit IRC | 03:45 | |
openstackgerrit | Ruby Loo proposed a change to openstack/ironic: properly format user guide in RST https://review.openstack.org/116168 | 03:48 |
*** nikunj2512 has joined #openstack-ironic | 04:04 | |
*** Poornima has joined #openstack-ironic | 04:04 | |
*** pcrews has quit IRC | 04:14 | |
*** tatyana has joined #openstack-ironic | 04:40 | |
openstackgerrit | Gregory Haynes proposed a change to openstack/ironic: Add conductor_hostname property to nodes https://review.openstack.org/109688 | 04:43 |
greghaynes | ok, that patch ended up much larger than expected :/ | 04:43 |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Take iSCSI deploy out of pxe driver https://review.openstack.org/111232 | 04:49 |
*** rakesh_hs has joined #openstack-ironic | 04:53 | |
*** pradipta_away is now known as pradipta | 04:54 | |
*** shakamunyi has quit IRC | 05:03 | |
*** rakesh_hs has quit IRC | 05:03 | |
*** rakesh_hs has joined #openstack-ironic | 05:05 | |
*** rakesh_hs has quit IRC | 05:09 | |
*** rakesh_hs has joined #openstack-ironic | 05:10 | |
*** rushiagr is now known as rushiagr_away | 05:21 | |
*** Poornima has quit IRC | 05:26 | |
*** tatyana has quit IRC | 05:27 | |
*** rushiagr_away is now known as rushiagr | 05:29 | |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Add support for creating vfat and iso images https://review.openstack.org/105413 | 05:33 |
*** rushiagr is now known as rushiagr_away | 05:40 | |
*** harlowja_away has quit IRC | 05:44 | |
*** tatyana has joined #openstack-ironic | 05:52 | |
*** killer_prince is now known as lazy_prince | 06:00 | |
openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Imported Translations from Transifex https://review.openstack.org/115878 | 06:12 |
*** lazy_prince is now known as killer_prince | 06:12 | |
*** killer_prince is now known as lazy_prince | 06:14 | |
*** shakamunyi has joined #openstack-ironic | 06:29 | |
*** shakamunyi has quit IRC | 06:34 | |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Add support for interacting with swift https://review.openstack.org/105795 | 06:34 |
*** rushiagr_away is now known as rushiagr | 06:34 | |
*** k4n0 has joined #openstack-ironic | 06:35 | |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: IloVirtualMediaIscsi deploy driver https://review.openstack.org/113865 | 07:04 |
*** Poornima has joined #openstack-ironic | 07:06 | |
*** viktors|afk is now known as viktors | 07:31 | |
*** derekh has joined #openstack-ironic | 08:14 | |
*** dtantsur|afk is now known as dtantsur | 08:15 | |
dtantsur | Morning Ironic, TGIF! | 08:16 |
*** lucasagomes has joined #openstack-ironic | 08:32 | |
yuriyz | morning Ironic dtantsur | 08:45 |
lucasagomes | morning yuriyz dtantsur | 08:50 |
*** zz_naotok has quit IRC | 08:50 | |
lucasagomes | gate is broken? | 08:50 |
lucasagomes | I see a couple of failures, some random ones | 08:50 |
dtantsur | lucasagomes, yuriyz, morning | 08:50 |
dtantsur | seems like yes | 08:50 |
*** zz_naotok has joined #openstack-ironic | 08:51 | |
*** rameshg87 has joined #openstack-ironic | 08:52 | |
*** igordcard has joined #openstack-ironic | 08:57 | |
*** lazy_prince is now known as killer_prince | 09:15 | |
*** killer_prince is now known as lazy_prince | 09:19 | |
openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/ironic: Add iPXE Installation Guide documentation https://review.openstack.org/106809 | 09:21 |
*** ndipanov has joined #openstack-ironic | 09:22 | |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: IloVirtualMediaAgent deploy driver https://review.openstack.org/115885 | 09:22 |
*** pradipta is now known as pradipta_away | 09:23 | |
*** pelix has joined #openstack-ironic | 09:28 | |
*** yuanying has quit IRC | 09:30 | |
*** lazy_prince is now known as killer_prince | 09:45 | |
*** killer_prince is now known as lazy_prince | 09:49 | |
rameshg87 | dtantsur, hi | 10:44 |
dtantsur | rameshg87, hi! | 10:45 |
rameshg87 | dtantsur, is there some problem with jenkins recently, i have been facing jenkins failures in all my submittals :( | 10:46 |
rameshg87 | dtantsur, http://logs.openstack.org/32/111232/16/check/gate-ironic-docs/20806a9/console.html says build_sphinx failed | 10:46 |
dtantsur | rameshg87, yeah, seems like that. I didn't investigate. | 10:46 |
rameshg87 | dtantsur, okay, seemed not related to my change. it's failing from yesterday.i thought it might get fixed in a day. but no submittals have gone in yesterday :) | 10:47 |
*** yuanying has joined #openstack-ironic | 10:50 | |
*** yuanying has quit IRC | 11:06 | |
openstackgerrit | Vladyslav Drok proposed a change to openstack/ironic: Add driver name on driver load exception https://review.openstack.org/112049 | 11:08 |
*** nosnos has quit IRC | 11:17 | |
*** nosnos has joined #openstack-ironic | 11:17 | |
*** nosnos has quit IRC | 11:21 | |
*** dtantsur is now known as dtantsur|lunch | 11:25 | |
lucasagomes | yeah this docs problem seems to be something in the wsme extension for sphinx | 11:26 |
lucasagomes | jobs in ceilometer are failing for the same reason | 11:26 |
rameshg87 | lucasagomes, okay | 11:28 |
*** rakesh_hs has quit IRC | 11:29 | |
rameshg87 | lucasagomes, i have one question in agent driver | 11:30 |
rameshg87 | lucasagomes, https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/agent.py#L221 - agent seems to use management interface to set the boot device | 11:31 |
lucasagomes | rameshg87, ack, (I know very little about it) | 11:31 |
rameshg87 | lucasagomes, but agent driver does have management interface - https://github.com/openstack/ironic/blob/master/ironic/drivers/agent.py#L34-L41 | 11:31 |
rameshg87 | lucasagomes, or am i missing something ?? | 11:31 |
lucasagomes | rameshg87, oh... looks like a bug | 11:31 |
lucasagomes | jroll, ^ | 11:31 |
lucasagomes | rameshg87, so that utils method will check if the mgmt interface is present | 11:31 |
lucasagomes | before setting the boot device | 11:32 |
lucasagomes | that's why it didn't fail | 11:32 |
rameshg87 | lucasagomes, yeah seemed a bug to me | 11:32 |
lucasagomes | +1 | 11:32 |
rameshg87 | lucasagomes, i see that for agent ssh driver - https://github.com/openstack/ironic/blob/master/ironic/drivers/agent.py#L82 | 11:32 |
rameshg87 | lucasagomes, so must have missed it :) | 11:32 |
lucasagomes | yup but not for ipmitool/ipminative ones | 11:33 |
*** nikunj2512 has quit IRC | 11:34 | |
*** yuanying has joined #openstack-ironic | 11:37 | |
*** lucasagomes is now known as lucas-hungry | 11:39 | |
*** yuanying has quit IRC | 11:41 | |
jroll | morning ironic, happy friday! :) | 11:57 |
jroll | lucas-hungry: turns out I was right :P | 11:57 |
jroll | rameshg87: want to add a bug, fix it, both, neither? | 11:58 |
rameshg87 | jroll, i can :) | 11:58 |
rameshg87 | jroll, do both | 11:58 |
rameshg87 | jroll, wanted to confirm with you | 11:58 |
jroll | rameshg87: yeah I agree :) | 11:58 |
jroll | thanks! | 11:58 |
rameshg87 | jroll, okay, i will take care of it | 11:58 |
* jroll is going to work harder on getting check jobs in | 11:58 | |
dtantsur|lunch | jroll, morning! | 12:10 |
rameshg87 | jroll, i have a question | 12:13 |
rameshg87 | jroll, why doesn't agent driver validate whether image_source has been specified or not: https://github.com/openstack/ironic/blob/master/ironic/drivers/modules/agent.py#L190-L205 | 12:13 |
rameshg87 | jroll, like pxe driver | 12:13 |
openstackgerrit | Ramakrishnan G proposed a change to openstack/ironic: Add management interface for agent drivers https://review.openstack.org/116241 | 12:26 |
openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/115600 | 12:28 |
*** rameshg87 has quit IRC | 12:28 | |
*** linggao has joined #openstack-ironic | 12:29 | |
openstackgerrit | Vladyslav Drok proposed a change to openstack/ironic: Remove gettextutils _ injection https://review.openstack.org/110634 | 12:30 |
*** Poornima has quit IRC | 12:32 | |
openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/python-ironicclient: Updated from global requirements https://review.openstack.org/116254 | 12:34 |
jroll | ram<tab> :( | 12:34 |
jroll | morning dtantsur|lunch :) | 12:35 |
dtantsur|lunch | oh, I'm not on lunch any more :) | 12:35 |
*** dtantsur|lunch is now known as dtantsur | 12:35 | |
jroll | :P | 12:35 |
dtantsur | hmmm... though I can have some tea! brb | 12:35 |
*** yuanying has joined #openstack-ironic | 12:37 | |
*** yuanying has quit IRC | 12:42 | |
*** rushiagr is now known as rushiagr_away | 12:51 | |
*** lucas-hungry is now known as lucasagomes | 12:55 | |
*** lazy_prince has quit IRC | 12:58 | |
jroll | vdrok: thanks for splitting that up :) | 12:59 |
jroll | oh, right, our gate is broken | 13:00 |
jroll | whee. | 13:00 |
jroll | fix is +A'd though | 13:00 |
jroll | bbiab | 13:00 |
openstackgerrit | Syed Ismail Faizan Barmawer proposed a change to openstack/ironic: Add UEFI based deployment support in Ironic https://review.openstack.org/114357 | 13:00 |
openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: EXPERIMENTAL Endpoint for accepting discovered data https://review.openstack.org/116003 | 13:17 |
*** Poornima has joined #openstack-ironic | 13:21 | |
*** kincl has joined #openstack-ironic | 13:22 | |
vdrok | thanks jroll! | 13:28 |
*** aswadr has quit IRC | 13:32 | |
*** pcrews has joined #openstack-ironic | 13:33 | |
NobodyCam | good morning Ironic, TGIF :) | 13:34 |
lucasagomes | morning NobodyCam | 13:34 |
NobodyCam | morning lucasagomes | 13:35 |
linggao | Good morning NobodyCam lucasagomes jroll dtantsur vdrok | 13:35 |
dtantsur | NobodyCam, linggao, morning! | 13:35 |
lucasagomes | morning linggao :) | 13:35 |
NobodyCam | morning linggao :) | 13:35 |
NobodyCam | morning dtantsur | 13:36 |
vdrok | morning linggao NobodyCam | 13:36 |
vdrok | afternoon dtantsur lucasagomes | 13:37 |
lucasagomes | vdrok, afternoon/evening | 13:37 |
dtantsur | vdrok, g'afternoon | 13:37 |
*** jasondotstar has joined #openstack-ironic | 13:37 | |
NobodyCam | good morning vdrok :) | 13:38 |
*** yuanying has joined #openstack-ironic | 13:38 | |
NobodyCam | lucasagomes: question ? did you see my paste's from last night? | 13:39 |
lucasagomes | NobodyCam, the ones about the nova commands proxy? | 13:39 |
NobodyCam | ya | 13:40 |
lucasagomes | yup | 13:40 |
NobodyCam | that look ok? | 13:40 |
lucasagomes | NobodyCam, I remember you were getting an error | 13:40 |
lucasagomes | so I commented about it | 13:40 |
lucasagomes | because afair it was doing a client.node.list() | 13:41 |
*** Alexei_987 has joined #openstack-ironic | 13:41 | |
lucasagomes | and appending the returned list to another list | 13:41 |
lucasagomes | afaici, we should get the ironic node list (client.node.list()) and covert each element to a "nova bm" object/dict | 13:42 |
lucasagomes | and then return the list of converted objects | 13:42 |
NobodyCam | lucasagomes: http://paste.openstack.org/show/Em861wMwFvrFlsWkugfX/ | 13:42 |
lucasagomes | NobodyCam, w00t! | 13:43 |
*** yuanying has quit IRC | 13:43 | |
NobodyCam | please look at the fields. | 13:43 |
NobodyCam | if we have to convert username/address/it gets mucky | 13:44 |
lucasagomes | oh because of the different drivers right? | 13:44 |
NobodyCam | yep | 13:44 |
lucasagomes | well nova bm only supported ipmitool + pxe | 13:44 |
lucasagomes | we can constraint it | 13:44 |
lucasagomes | only fill the username/password if the driver is == pxe_ipmitool | 13:45 |
lucasagomes | because that's what is supported by nova bm | 13:45 |
lucasagomes | so you can look at ipmi_username/ipmi_password if driver == pxe_ipmitool | 13:45 |
NobodyCam | ya if we limit to one we're good | 13:45 |
lucasagomes | I think it's fair | 13:45 |
NobodyCam | what about nova's interface field | 13:47 |
NobodyCam | note show vs list in the output | 13:47 |
lucasagomes | nova only supported 1 NIC per node? | 13:47 |
NobodyCam | ?? | 13:48 |
NobodyCam | been so long | 13:48 |
NobodyCam | lol | 13:48 |
lucasagomes | I see like MAC Address | 13:48 |
lucasagomes | on the baremetal-node-list | 13:48 |
lucasagomes | and for baremetal-node-show idk what "interfaces" looked like... hmm maybe we should start a devtest with nova bm and check that out | 13:49 |
lucasagomes | or look at the code, may be faster | 13:49 |
NobodyCam | lucasagomes: http://docs.openstack.org/user-guide/content/cli_nova_baremetal-details.html | 13:50 |
lucasagomes | oh | 13:51 |
NobodyCam | brb | 13:54 |
*** russellb is now known as rustlebee | 13:54 | |
*** k4n0 has quit IRC | 13:55 | |
*** teju has joined #openstack-ironic | 13:56 | |
NobodyCam | ok had to run out for a quick smoke :-p | 13:59 |
lucasagomes | ack | 14:00 |
NobodyCam | oh lucasagomes dtantsur did you see https://review.openstack.org/#/c/68942/32 | 14:02 |
lucasagomes | nop | 14:03 |
lucasagomes | lemme see | 14:03 |
dtantsur | NobodyCam, yeah, we have a patch somewhere I guess | 14:03 |
dtantsur | NobodyCam, https://review.openstack.org/#/c/116093/ | 14:03 |
dtantsur | should we approve it? | 14:04 |
NobodyCam | dtantsur: what failed in the test | 14:04 |
NobodyCam | and we will need mrda to approve | 14:05 |
dtantsur | trying to figure out... | 14:05 |
dtantsur | seems like still tempest problem | 14:06 |
*** rameshg87 has joined #openstack-ironic | 14:07 | |
*** teju has quit IRC | 14:08 | |
rameshg87 | jroll, hi | 14:14 |
jroll | hey rameshg87 | 14:17 |
jroll | rameshg87: the image_source validation (or lack thereof) was probably just an oversight | 14:17 |
rameshg87 | hi jroll | 14:18 |
NobodyCam | morning jroll, rameshg87 | 14:18 |
jroll | morning NobodyCam :) | 14:18 |
NobodyCam | :) | 14:18 |
rameshg87 | good morning NobodyCam :-) | 14:18 |
rameshg87 | jroll, so i think it makes sense to add image_source validation since agent driver can't deploy without it | 14:19 |
rameshg87 | jroll, and regarding the tests for https://review.openstack.org/#/c/116241/ . do you have any particular things in mind that we can assert | 14:19 |
jroll | rameshg87: sure, it would be good | 14:20 |
*** rameshg87 has quit IRC | 14:21 | |
jroll | . | 14:22 |
jroll | whyyyyyy | 14:22 |
jroll | adam_g: I'm tempted to recheck 116093 with no bug unless you think that's a real failure | 14:31 |
*** yuanying has joined #openstack-ironic | 14:39 | |
NobodyCam | jroll: see error here:http://logs.openstack.org/93/116093/3/check/check-tempest-dsvm-virtual-ironic/acaf2f9/logs/testr_results.html.gz | 14:40 |
jroll | NobodyCam: yeah, saw the error, looks like a tempest issue but idk | 14:40 |
jroll | I like to recheck for funsies | 14:40 |
NobodyCam | jroll: ++ on recheck | 14:41 |
jroll | ok | 14:41 |
* jroll does it | 14:41 | |
jroll | also | 14:41 |
jroll | other than failing tests, why would we need mrda to approve it? | 14:41 |
jroll | it... doesn't work without that change | 14:41 |
jroll | afaik | 14:41 |
jroll | like, just do it and poke mrda | 14:42 |
*** yuanying has quit IRC | 14:43 | |
NobodyCam | :) only because he's doing the nova patches | 14:45 |
NobodyCam | brb | 14:46 |
lucasagomes | jroll, yeah if that's required (a must) we should then just approve and let him know | 14:46 |
jroll | yep, +1 | 14:46 |
* jroll bbl | 14:49 | |
*** rushiagr_away is now known as rushiagr | 14:49 | |
openstackgerrit | Dmitry Tantsur proposed a change to openstack/ironic: EXPERIMENTAL Endpoint for accepting discovered data https://review.openstack.org/116003 | 15:00 |
openstackgerrit | Vladyslav Drok proposed a change to openstack/ironic: Cleanup _LW, _LC, _LI translations https://review.openstack.org/116303 | 15:02 |
* NobodyCam is back | 15:05 | |
openstackgerrit | Steve Martinelli proposed a change to openstack/ironic-specs: Update spec theme to show incubation status https://review.openstack.org/116306 | 15:07 |
*** vdrok is now known as vdrok_away | 15:10 | |
*** chuckC has joined #openstack-ironic | 15:10 | |
*** killer_prince has joined #openstack-ironic | 15:14 | |
*** killer_prince is now known as lazy_prince | 15:15 | |
*** amitpp has joined #openstack-ironic | 15:19 | |
*** jgrimm has joined #openstack-ironic | 15:22 | |
*** shakamunyi has joined #openstack-ironic | 15:24 | |
*** igordcard has quit IRC | 15:26 | |
*** amitpp has quit IRC | 15:29 | |
openstackgerrit | Chris Krelle proposed a change to openstack/ironic: Proxy Nova baremetal commands to Ironic https://review.openstack.org/116316 | 15:31 |
NobodyCam | I'm not sure that is the correct place to submit that patch | 15:33 |
NobodyCam | also it needs cleanup and such :-p | 15:33 |
*** amitpp has joined #openstack-ironic | 15:38 | |
*** yuanying has joined #openstack-ironic | 15:40 | |
*** shakamunyi has quit IRC | 15:41 | |
*** yuanying has quit IRC | 15:44 | |
dtantsur | TGIF and I'm leaving now :) | 16:00 |
dtantsur | have a nice weekend :) | 16:00 |
*** dtantsur is now known as dtantsur|afk | 16:00 | |
adam_g | jroll, we're blocked until https://review.openstack.org/#/c/116103/ merges | 16:01 |
adam_g | its been queued for gate checks for 6 hrs now | 16:03 |
NobodyCam | night dtantsur|afk | 16:03 |
NobodyCam | adam_g: gah :( | 16:03 |
*** amitpp has quit IRC | 16:08 | |
*** amitpp has joined #openstack-ironic | 16:11 | |
*** bandicot has joined #openstack-ironic | 16:17 | |
*** bandicot is now known as bensonmiller | 16:20 | |
*** tatyana has quit IRC | 16:20 | |
jroll | adam_g: gotcha | 16:23 |
JayF | adam_g: should we ask -infra to bump it? | 16:24 |
JayF | adam_g: I know they actively encourage folks to have them bump prio on gate-fixing bugs | 16:24 |
*** dlaube has joined #openstack-ironic | 16:29 | |
adam_g | JayF, ha, it just started running as i was typing a ping to -infra to get it bumped | 16:30 |
JayF | :) | 16:30 |
*** rushiagr is now known as rushiagr_away | 16:31 | |
*** openstackgerrit has quit IRC | 16:34 | |
NobodyCam | :-p | 16:34 |
*** derekh has quit IRC | 16:34 | |
*** yuanying has joined #openstack-ironic | 16:41 | |
*** yuanying has quit IRC | 16:45 | |
*** Poornima has quit IRC | 16:54 | |
ryanpetrello | this ironic stable failure look familiar to anybody? | 16:57 |
ryanpetrello | http://logs.openstack.org/98/116198/1/check/gate-pecan-tox-ironic-stable/56b7b42/console.html | 16:57 |
ryanpetrello | 2014-08-22 08:37:36.666 | Traceback (most recent call last): | 16:57 |
ryanpetrello | 2014-08-22 08:37:36.666 | File "ironic/tests/api/v1/test_drivers.py", line 49, in test_drivers | 16:57 |
ryanpetrello | … | 16:58 |
ryanpetrello | 2014-08-22 08:37:43.665 | raise mismatch_error | 16:58 |
ryanpetrello | 2014-08-22 08:37:43.665 | MismatchError: ['fake-host1', 'fake-host2'] != [u'fake-host2', u'fake-host1'] | 16:58 |
adam_g | you may need to run with PYTHONHASHSEED='0' | 16:58 |
adam_g | hmm. maybe zuul was laying, 116103 is queued again at 7 hrs | 16:58 |
adam_g | lying | 16:59 |
JayF | then I'd totes ping infra | 16:59 |
adam_g | ryanpetrello, that looks like a side effect of the recent tox upgrade on the slaves | 17:00 |
NobodyCam | adam_g: it resets with each review that land ahead of it | 17:00 |
NobodyCam | lands even | 17:00 |
ryanpetrello | gotcha | 17:00 |
adam_g | NobodyCam, oh, right | 17:01 |
*** comstud is now known as bearhands | 17:02 | |
*** Alexei_987 has quit IRC | 17:05 | |
*** viktors is now known as viktors|afk | 17:09 | |
adam_g | has anyone seen this failure http://logs.openstack.org/03/116103/2/check/check-tempest-dsvm-virtual-ironic-nv/2fcf03b/logs/testr_results.html.gz ? the ironic job still failed in check on 116103 | 17:11 |
*** amitpp has quit IRC | 17:11 | |
*** lucasagomes is now known as lucas-dinner | 17:15 | |
NobodyCam | have a good weekend lucas-dinner | 17:23 |
adam_g | ugh. theres actually two gate breakers ATM. | 17:28 |
NobodyCam | two? | 17:30 |
adam_g | https://bugs.launchpad.net/tempest/+bug/1360367 | 17:30 |
*** openstackgerrit has joined #openstack-ironic | 17:31 | |
*** pelix has quit IRC | 17:31 | |
NobodyCam | :( | 17:34 |
jroll | btw, I won't be working on ironic today, but if y'all need something from me feel free to poke me | 17:34 |
jroll | I'll be online | 17:35 |
adam_g | the hash ordering thing is just a PITA from the tox upgrade breaking the world, but that second bug is another case of 'we really need to be voting on other projects' | 17:35 |
*** harlowja has joined #openstack-ironic | 17:36 | |
NobodyCam | yes, or cores need to look at what there is a failed job. | 17:40 |
NobodyCam | we could non-voting-stable and non-voting-unstable | 17:41 |
NobodyCam | to indecate if the just is infact stable or not | 17:41 |
NobodyCam | :-p | 17:41 |
*** yuanying has joined #openstack-ironic | 17:41 | |
NobodyCam | indicate even | 17:42 |
jroll | no, we just need to be able to vote | 17:43 |
jroll | it's absurd that we can't | 17:43 |
adam_g | yeah, unless our tests are actually voting nobody is going to pay attention | 17:44 |
JayF | I've seen people proudly declare how they ignore nonvoting tests on the ML and otherwise :( | 17:44 |
NobodyCam | uggh doc jobs are failing | 17:44 |
NobodyCam | https://review.openstack.org/#/c/116093 | 17:44 |
JayF | not sure why it's OK to be so blaze' about breaking other folks stuff | 17:44 |
*** yuanying has quit IRC | 17:46 | |
NobodyCam | i'd even be happy with notice that anohter porject is landing something that is going to break us :-p | 17:46 |
NobodyCam | patch ####### failed your test but was +a'd | 17:47 |
SpamapS | Question about concurrency in Ironic | 17:48 |
NobodyCam | shoot | 17:48 |
jroll | adam_g: the best are when they change lib/ironic, our tests fail, and they merge it anyway | 17:48 |
SpamapS | Suppose I tell Ironic to rebuild 100 machines | 17:48 |
jroll | JayF: because f the not blessed projects | 17:48 |
SpamapS | Can you walk me through what you would expect to happen, if I have one conductor. | 17:49 |
jroll | SpamapS: uh huhhhhh | 17:49 |
SpamapS | assume the machines have a 5 minute POST | 17:49 |
jroll | SpamapS: it should work, just slowly, I think. | 17:49 |
NobodyCam | SpamapS: it rebuilds them one by one? | 17:49 |
jroll | SpamapS: if you use the agent driver, it is alllllll goooooood. | 17:49 |
jroll | oh, rebuild, ha | 17:49 |
jroll | I guess it would still be fine | 17:50 |
SpamapS | so, power off... power on.. deploy ramdisk.. deploy root .. power off.. power on. ? | 17:50 |
NobodyCam | yep | 17:50 |
jroll | NobodyCam: one by one? really? | 17:50 |
SpamapS | so 16 hours to rebuild 100 machines | 17:50 |
SpamapS | assuming image transfer is instant. ;) | 17:50 |
NobodyCam | I've actually never tested that, | 17:51 |
jroll | lolol | 17:51 |
jroll | it should be parallel | 17:51 |
SpamapS | What I'd like to see is they are all powered off, powered on, and in the deploy ramdisk. | 17:51 |
SpamapS | if we only do one dd at a time.. thats fine | 17:51 |
jroll | especially with the agent model | 17:51 |
*** igordcard has joined #openstack-ironic | 17:51 | |
NobodyCam | SpamapS: have you actually tested at that scale? | 17:52 |
SpamapS | jroll: can you explain the difference the agent driver would introduce? I'm only familiar with IPMI+PXE | 17:52 |
SpamapS | NobodyCam: no, but we're about to. :) | 17:52 |
SpamapS | NobodyCam: working out a few more kinks and somebody asked how thats likely to work. | 17:52 |
jroll | SpamapS: the image is deployed by: | 17:53 |
jroll | an http call to the agent from conductor | 17:53 |
jroll | agent downloads image from swift and writes it out | 17:53 |
NobodyCam | SpamapS: how are the rebuild commands being issued? 100 nova rebuild <balh> commands? | 17:53 |
jroll | agent calls back to ironic and ironic reboots | 17:53 |
jroll | that's the fundamental difference, anyway | 17:53 |
*** amitpp has joined #openstack-ironic | 17:54 | |
SpamapS | jroll: so how do we know the root partition is o-k to destroy? run agent from ramdisk? | 17:56 |
SpamapS | NobodyCam: correct | 17:56 |
SpamapS | NobodyCam: one after the other | 17:56 |
NobodyCam | I expect that is how the rebuild are going to run | 17:56 |
NobodyCam | one after the other | 17:56 |
NobodyCam | but would love to see the test results | 17:57 |
SpamapS | NobodyCam: they are reported all as in progress by Nova | 17:58 |
*** amitpp has quit IRC | 17:58 | |
SpamapS | NobodyCam: but I haven't paid attention to what state each node is in while they're in that state. | 17:58 |
NobodyCam | :) | 17:59 |
dlaube | interesting stuff.. can't wait to hear the results SpamapS | 18:00 |
NobodyCam | morning dlaube :) | 18:00 |
SpamapS | If they all end up sitting in the deploy ramdisk and are just rebooted as they finish, then thats good, because then 16 hours shrunks to 10 minutes + nnodes*image_transfer_time. | 18:02 |
SpamapS | shrinks | 18:02 |
dlaube | g'morning guys | 18:03 |
NobodyCam | SpamapS: it may get https://github.com/openstack/ironic/blob/master/etc/ironic/ironic.conf.sample#L544 threads | 18:05 |
NobodyCam | minus any periodic task workers | 18:06 |
JayF | jroll: Does agent driver support rebuild yet? | 18:06 |
JayF | SpamapS: fwiw right now agent only supports full disk images, and doesn't honor --preserve-ephemeral (the worst named flag ever) | 18:07 |
SpamapS | JayF: it's amazing isn't it? | 18:07 |
JayF | SpamapS: but I can tell you we've done lots of simultaneous builds before, not with one conductor, but like >500 nodes + 4 conductors, we had someone build around 100 machines near-simultaneously without incident | 18:08 |
SpamapS | JayF: I think we should just call it --http://beyondbasicplay.files.wordpress.com/2013/09/catching-bubbles.jpg | 18:08 |
JayF | we should call it | 18:08 |
JayF | --preserve-snowflake | 18:08 |
JayF | because as an option it completely goes against the cloud-style of deployment and disposable servers :( | 18:08 |
jroll | JayF: I believe rebuild is a thing, but we don't support --preserve-ephemeral | 18:09 |
jroll | JayF: actually we don't support ephemeral at all | 18:09 |
jroll | because whole disk images | 18:09 |
jroll | that said, any disks that are not the boot disk will be preserved in the upstream agent | 18:09 |
SpamapS | JayF: 2TB of swift data isn't really ever going to fit in that model, and it also isn't cost effective to run that over iSCSI. DAS is not going away. | 18:09 |
jroll | morning dlaube :) | 18:10 |
JayF | SpamapS: A man can dream, right? | 18:11 |
SpamapS | I mean we can just dispose of a swift server. But that is going to waste ridiculous amounts of CPU and bandwidth to resync. | 18:11 |
JayF | I think we should just accept that it's a workaround because some things don't fit in that model yet | 18:11 |
jroll | NobodyCam: did your soft power patch land? | 18:11 |
JayF | that being said, like jroll said, the agent will not image any extra disks | 18:11 |
JayF | the logic for which disk to image is a little weird | 18:11 |
JayF | I think it's something like smallest disk >8GB | 18:11 |
jroll | 4GB | 18:11 |
SpamapS | I'd like to have the disks and the partitions of the box addressible using a set of UUID's. | 18:12 |
JayF | patches welcome :) | 18:12 |
SpamapS | Thats something I think we're wanting to do soon-ish. | 18:12 |
JayF | it's not perfect but it works right now for many cases | 18:12 |
JayF | we have a stack of things for k we want to do soonish | 18:12 |
SpamapS | A few of our engineers at HP are a bit nervous about the fact that preserve_ephemeral is not defaulted to True. | 18:13 |
jroll | what | 18:13 |
JayF | and by that I mean we're doing it now in our production but just couldn't get it through "The Process(tm)" in time | 18:13 |
* jroll flips tables | 18:13 | |
jroll | sorry | 18:13 |
jroll | 18:12:24 SpamapS | I'd like to have the disks and the partitions of the box addressible using a set of UUID's. <- in ironic? | 18:14 |
SpamapS | jroll: yes in Ironic. | 18:14 |
SpamapS | Just basically as it is claiming the disks, feed back the UUID's it has claimed. | 18:14 |
jroll | ironic shouldn't know about those things, this is not an inventory system :| | 18:14 |
jroll | well | 18:15 |
jroll | s/know about/keep track of | 18:15 |
SpamapS | And then I can say "deploy image X to wherever the disk with uuid==xxxxx is" | 18:15 |
SpamapS | It knows where MACs are. | 18:16 |
NobodyCam | jroll: nope, but adam_g put this up: https://review.openstack.org/#/c/116093 | 18:16 |
jroll | NobodyCam: where's your soft power patch, then? | 18:16 |
SpamapS | jroll: It's not inventory. It's just a bread crumb. | 18:16 |
NobodyCam | mine or Ng's? | 18:16 |
jroll | ehhhh | 18:16 |
jroll | NobodyCam: whichever we want to land | 18:16 |
jroll | SpamapS: I don't believe in preserve_ephemeral, and I barely believe in rebuild, so I'm likely the wrong person to talk to about this stuff | 18:17 |
jroll | I can see ironic keeping track of disks | 18:17 |
jroll | but not letting you say "I want to use disk x" | 18:17 |
jroll | idk | 18:17 |
jroll | I would want to hear more, I guess | 18:18 |
SpamapS | jroll: well oddly enough, I'm talking about how to get out of the rebuild business. :) | 18:18 |
jroll | right, but you're still in the snowflake business at that point :) | 18:18 |
jroll | and I get the use case, but | 18:18 |
jroll | this is the cloud, folks should use it like a cloud :/ | 18:18 |
SpamapS | jroll: I want to express "it would be most efficient if you deployed to the box with the data I left behind" | 18:18 |
NobodyCam | jroll: I thought we came up with neither: but https://review.openstack.org/#/c/107864/ <-Ng's https://review.openstack.org/#/c/107778 <- Nobodycam's | 18:19 |
JayF | SpamapS: IMO the concept of 'I left data behind' scares the crap out of me, as a multi-tenant operator of Ironic ;) | 18:19 |
SpamapS | jroll: so would you put massive databases on iscsi/ceph/SAN and just take the complexity and CAPEX hit? | 18:19 |
JayF | I would upgrade massive databases in-band | 18:20 |
JayF | instead of using provisioning tooling to do it | 18:20 |
jroll | NobodyCam: hmm, I thought we had one in the right direction, idk | 18:20 |
jroll | right, I wouldn't reinstall the OS to upgrade a thing | 18:20 |
SpamapS | JayF: we both know multi-tenant hardware sharing is not a good idea. But anyway, the data is there on the platters so you have to remove it if you don't want it left behind anyway. :) | 18:20 |
JayF | SpamapS: which we do. | 18:20 |
NobodyCam | jroll: I liked Ng version, | 18:20 |
SpamapS | JayF: right so you'd not have any UUID's for the user. | 18:21 |
NobodyCam | but what I had may fit better with what landed in nova, if thats what your thinking? | 18:21 |
JayF | SpamapS: one of the stack of things for K is decom, implemented and working in our deployment of Ironic, with a hardware manager that does all the wiping :) | 18:21 |
SpamapS | JayF: I'd treat these like volumes. I want to boot from the volume that has UUID=XXXX, which I still own because I asked you not to delete it. | 18:21 |
JayF | SpamapS: fwiw code is up as well ): | 18:21 |
JayF | s/):/:)/ | 18:21 |
JayF | ^ aka the 'turn that frown upside-down' regexp | 18:21 |
jroll | NobodyCam: yeah, maybe, idk. see SpamapS' email to the list | 18:21 |
jroll | lol | 18:21 |
SpamapS | jroll: regarding snowflakes btw, I don't see it as snowflakes. It is a cost savings to use DAS and to be able to use image based deployments on real servers with DAS. | 18:23 |
jroll | SpamapS: I'm curious what the use case is for using 'rebuild' to upgrade things | 18:24 |
SpamapS | The reality is that we're not suggesting using MySQL with a single node and saying if you don't preserve my data, our business dies. | 18:24 |
NobodyCam | SpamapS: why not ssh switch to run level 1 and shurdown / unmount everything then issue halt? | 18:24 |
jroll | or more, if "that's how tripleo works" is the use case | 18:24 |
jroll | NobodyCam: ironic powers it back on :P | 18:25 |
JayF | SpamapS: my thoughts are kinda different than jroll's on this. Mostly just as an operational guy, the idea of swapping out my OS and leaving my data and hoping that it works when it comes back up frightens me | 18:25 |
JayF | SpamapS: I much prefer to replicate data, stand up parallel clusters and migrate, or do upgrades in a rolling fashion | 18:25 |
SpamapS | NobodyCam: runlevel 1 is no SSH, but even if I leave SSH up, at some point the OS shuts everything down so it can unmount all remaining disks. | 18:25 |
JayF | but I don't particularly want to restrict your use case :) I just don't understand why you'd want to go down that path | 18:25 |
NobodyCam | jroll: ok ssh in switch to run level 1 and shurdown / unmount everything then hard power off with ironic | 18:25 |
SpamapS | jroll: we do image based deployment. The idea is simple: put a single stream of bits through QA, and then put the exact same stream of bits into production. | 18:25 |
NobodyCam | humm | 18:26 |
JayF | similar, I'm guessing, to the folks who don't understand why we do what we do | 18:26 |
jroll | hrm. | 18:26 |
JayF | SpamapS: I'd implement that pattern with Docker containers ;) | 18:26 |
SpamapS | And that is a valid, possible next step. | 18:26 |
SpamapS | That leaves us with kernel updates, which might be isolated enough to not want an image based update pattern. | 18:27 |
SpamapS | I believe this is precisely the CoreOS model. | 18:27 |
SpamapS | That also puts us in a bad spot for cinder-volume, which cannot run inside a docker container | 18:28 |
SpamapS | (because iscsi is not namespaced properly) | 18:29 |
JayF | That's unfortunate about iscsi not being namespaced :( | 18:29 |
JayF | although you can still use containers, even if not docker. | 18:29 |
SpamapS | Yeah I kind of hoped a year ago when we figured that out that we would just wave our hands until they fixed it. | 18:29 |
JayF | I think we are the 'they' | 18:29 |
JayF | people working for big companies on open source enabling what we need to do :) | 18:29 |
JayF | there is no they anymore, we are they :P | 18:30 |
SpamapS | My version of "they" was realy Serge Hallyn | 18:30 |
SpamapS | But then he got busy. :-P | 18:31 |
SpamapS | It's realy not iSCSI itself, but the netlink socket control mechanism it relies on. | 18:31 |
SpamapS | If they'd just used ioctls like everybody else, it would be done. | 18:31 |
SpamapS | Anyway, another version of what we do doesn't even bother with a docker container. | 18:32 |
SpamapS | It works like your agent sort of. Download the image, force it over top of the root partition, and restart everything. | 18:32 |
SpamapS | But that still can't update kernels, and pid1 would be haphazard. :-P | 18:33 |
SpamapS | Anyway your time has been much appreciated. Thanks. :) | 18:35 |
NobodyCam | jroll: SpamapS: looking at https://review.openstack.org/#/c/107778 and https://review.openstack.org/#/c/116093 we could make use of the nova timeout value and attempt a soft power off. | 18:37 |
*** yuanying has joined #openstack-ironic | 18:42 | |
Ng | I sort of want to think that maintenance mode or something like it, could be a way out here | 18:42 |
Ng | can you rebuild() a node that's in maintenance mode? | 18:43 |
Ng | i.e. if we could just tell Ironic to stop enforcing power state on a per-node basis, that would be sufficient | 18:43 |
SpamapS | @Ng | 18:44 |
SpamapS | gah | 18:44 |
* SpamapS stabs hipchat | 18:44 | |
NobodyCam | Ng: oh | 18:44 |
Ng | mwaha | 18:44 |
Ng | so like, we say "stop asserting power", ssh in and do "shutdown -h now", poll Ironic for power state == OFF, call rebuild, flip power assertion back on (or maybe have that be implicit in the rebuild) | 18:45 |
SpamapS | Ng: thats a workaround for sure, but it stands to reason that most of the time if I'm doing a 'nova stop' I don't mean "and do nothing to avoid corrupting the root filesystem" | 18:45 |
SpamapS | kvm issues a soft stop | 18:45 |
NobodyCam | thats true a node in maintenance mode skips the power stuff | 18:45 |
Ng | SpamapS: what does nova-kvm do if the guest OS ignores the ACPI signal? | 18:46 |
*** yuanying has quit IRC | 18:46 | |
Ng | I think my proposed patch added another power state, so you could distinguish between sending a node a hard or a soft power off | 18:47 |
Ng | but I was literally asleep when I wrote it, so I'm not really sure what it did. probably implemented Counting The Moon as a Service | 18:47 |
JayF | Ng: def count_the_moon() | 18:48 |
JayF | Ng: return 1 | 18:48 |
JayF | Ng: ## todo(jayf) support for other planets | 18:48 |
SpamapS | #winning | 18:48 |
Ng | hehe | 18:48 |
Ng | I would be so stunned if anyone got that reference :) | 18:49 |
SpamapS | Ng: I don't really know. Easy enough to test, just kill -9 pid 1 and then issue a stop.. | 18:49 |
Ng | I don't often test things, but when I do, it involves sending -9 to init | 18:49 |
NobodyCam | lol | 18:51 |
NobodyCam | brb | 18:53 |
*** ndipanov has quit IRC | 18:57 | |
dlaube | has anyone tried playing with ipv6 in ironic? | 19:26 |
dlaube | in terms of provisioning baremetal servers with v6 addresses I mean | 19:26 |
jroll | isn't that more about the switch than ironic itself? | 19:27 |
jroll | not sure what you mean | 19:27 |
JayF | I sadly do not have experience with it :-( | 19:28 |
jroll | ^^ | 19:28 |
dlaube | well, I'm mostly interested in hearing if anyone is using v6 addresses with IPMI as it relates to ironic…. and any ipv6 support/limitations in neutron | 19:29 |
dlaube | been reading up over here -> https://wiki.openstack.org/wiki/Neutron/IPv6 | 19:30 |
jroll | ahh | 19:30 |
jroll | I think neutron supports it ok | 19:30 |
jroll | I wouldn't trust ipmi to do so but idk | 19:30 |
dlaube | gotcha | 19:31 |
dlaube | yeah, I've been looking into it and a bunch of vendors let you use v6 for IPMI but usually not by default and not easily :P | 19:31 |
dlaube | http://ipvsix.me/?p=159 | 19:32 |
jroll | yeah | 19:32 |
jroll | I mean | 19:32 |
jroll | those should be on an isolated network anyhow | 19:32 |
dlaube | yeah | 19:32 |
*** Hefeweizen has quit IRC | 19:34 | |
NobodyCam | dlaube: only enough to file this: https://bugs.launchpad.net/ironic/+bug/1325942 | 19:39 |
dlaube | ty | 19:41 |
*** yuanying has joined #openstack-ironic | 19:43 | |
*** yuanying has quit IRC | 19:47 | |
*** penick has joined #openstack-ironic | 19:55 | |
*** igordcard has quit IRC | 20:04 | |
*** penick has quit IRC | 20:11 | |
*** rwsu has quit IRC | 20:13 | |
*** Hefeweizen has joined #openstack-ironic | 20:21 | |
*** rushiagr_away is now known as rushiagr | 20:27 | |
NobodyCam | conf calls done | 20:35 |
NobodyCam | :-p | 20:35 |
NobodyCam | now brb | 20:35 |
*** yuanying has joined #openstack-ironic | 20:44 | |
*** yuanying has quit IRC | 20:48 | |
adam_g | im gonna be disappearing soon but our gate should be good again once https://review.openstack.org/#/c/116353/ + https://review.openstack.org/#/c/116103/ merge | 20:57 |
adam_g | 116103 looks like its about to fail its 10 hr gate job :( so it will need a recheck | 20:58 |
adam_g | 116353 will hopefully be approved and sent to gate before EOD | 20:58 |
NobodyCam | adam_g: I'll keep an eye on them and recheck / reverify if needed | 20:58 |
adam_g | NobodyCam, thanks | 20:58 |
JayF | when a gate is broken in two ways | 21:05 |
JayF | how do you get the patches merged? | 21:05 |
JayF | like obviously unless both are applied the checks still fail, right? | 21:05 |
JayF | like this case, that might not be what's happening, but generally what's the approach? | 21:06 |
NobodyCam | ping infra | 21:06 |
NobodyCam | :-p | 21:06 |
NobodyCam | beg fungi or clarkb for help | 21:06 |
NobodyCam | lol | 21:06 |
JayF | that's a horrible answer :( | 21:06 |
NobodyCam | hehehehe | 21:06 |
NobodyCam | in this case both foxes are to tempest so you could make one dependent on the other | 21:09 |
NobodyCam | s/foxes/fixes/ | 21:09 |
*** linggao has quit IRC | 21:17 | |
adam_g | JayF, in this case the fact that ironic job is non-voting makes it easier | 21:18 |
JayF | aah | 21:18 |
JayF | because it's not 2x ironic patches to fix | 21:18 |
jroll | lol | 21:18 |
JayF | it's 2x tempest patches | 21:19 |
JayF | the fact the irnoic job is non-voting is also a direct root cause | 21:19 |
jroll | right | 21:19 |
jroll | we wouldn't have this problem in the first place if we voted | 21:19 |
*** rushiagr is now known as rushiagr_away | 21:20 | |
adam_g | doh | 21:23 |
adam_g | jroll, yeah, we should propose making the job voting again in the tempest check pipeline, on the basis we are almost a week from FF and can't afford to have things blocked like this | 21:25 |
NobodyCam | adam_g: ++ I would support that | 21:25 |
NobodyCam | is reverify not working anymore? | 21:26 |
NobodyCam | or Mr J just that backed up | 21:26 |
NobodyCam | mrda: been on today? | 21:30 |
adam_g | NobodyCam, i think its backed up on its post-back. i see 116103 running in the check queue now | 21:31 |
NobodyCam | should be just about that time. :) | 21:31 |
NobodyCam | adam_g: sweet :) | 21:31 |
adam_g | i need to run for a while. ill check back later. thanks for kicking 116103, NobodyCam | 21:32 |
*** rushiagr_away is now known as rushiagr | 21:33 | |
NobodyCam | :) | 21:34 |
*** rushiagr is now known as rushiagr_away | 21:36 | |
jroll | adam_g: don't they want us integrated or whatever to do that | 21:40 |
NobodyCam | jroll: we'll never get integrated if they don't stop breaking us LOL :( | 21:43 |
jroll | ikr | 21:43 |
*** yuanying has joined #openstack-ironic | 21:45 | |
*** igordcard has joined #openstack-ironic | 21:46 | |
*** jasondotstar has quit IRC | 21:48 | |
*** yuanying has quit IRC | 21:49 | |
NobodyCam | uggh 116103 going to need a recheck :-p | 22:11 |
*** penick has joined #openstack-ironic | 22:15 | |
*** penick_ has joined #openstack-ironic | 22:19 | |
*** penick has quit IRC | 22:21 | |
*** penick_ is now known as penick | 22:21 | |
*** lucas-dinner has quit IRC | 22:21 | |
*** yuanying has joined #openstack-ironic | 22:45 | |
*** yuanying has quit IRC | 22:50 | |
NobodyCam | 116103 recheck started :-p | 22:50 |
*** igordcard has quit IRC | 23:22 | |
*** chuckC has quit IRC | 23:31 | |
*** penick has quit IRC | 23:36 | |
openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/ironic: Updated from global requirements https://review.openstack.org/115600 | 23:38 |
jroll | much gate | 23:43 |
jroll | such broken | 23:44 |
jroll | wow | 23:44 |
* jroll looks for a bug # to recheck | 23:45 | |
JayF | I don't think we should recheck yet | 23:45 |
JayF | until those tempest things land, right? | 23:45 |
greghaynes | recheck such bug | 23:45 |
*** yuanying has joined #openstack-ironic | 23:46 | |
jroll | I rechecked | 23:47 |
jroll | JayF: we need that to land | 23:47 |
jroll | JayF: that is the tempest thing | 23:47 |
jroll | (one of the) | 23:47 |
JayF | oh. | 23:47 |
JayF | blargh. | 23:47 |
jroll | https://review.openstack.org/#/c/116103/ | 23:47 |
jroll | yeah | 23:47 |
jroll | I'm going to get up early tomorrow and just hammer rechecks on that until it goes through | 23:48 |
jroll | saturday morning means quick gate :P | 23:48 |
jroll | the heck, are "reverify"s not working? | 23:49 |
*** yuanying has quit IRC | 23:51 | |
NobodyCam | jenkins way backed up | 23:53 |
NobodyCam | postback take a long while | 23:53 |
jroll | yeah, I didn't realize it didn't tell you when jobs starts | 23:53 |
NobodyCam | but watch zuul the job will restart in a few | 23:53 |
jroll | yep | 23:54 |
NobodyCam | :) | 23:54 |
NobodyCam | and Ty | 23:54 |
jroll | NobodyCam: jay says it already failed neutron-icehouse though :P | 23:54 |
NobodyCam | ya :( | 23:54 |
jroll | s/:P/:( | 23:54 |
NobodyCam | :'( | 23:54 |
NobodyCam | lol | 23:54 |
jroll | :((((((((((((((((((((((( | 23:54 |
* jroll has had a long week and now it ends with a broken gate | 23:55 | |
*** aswadr has joined #openstack-ironic | 23:55 | |
NobodyCam | :( /me knows what he'll be doing this weekend ... typing recheck bug <RamdomInt^4> | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!