Wednesday, 2015-04-29

« Tuesday, 2015-04-28 Index Thursday, 2015-04-30 »
*** ijw has quit IRC00:00
*** dttocs has quit IRC00:09
*** dttocs has joined #openstack-ironic00:15
*** Haomeng|2 has joined #openstack-ironic00:16
*** Haomeng has quit IRC00:18
*** mtanino_ has quit IRC00:21
*** ijw has joined #openstack-ironic00:22
*** achanda has joined #openstack-ironic00:23
*** dttocs has quit IRC00:29
*** david-lyle has quit IRC00:32
*** ijw has quit IRC00:35
*** dttocs has joined #openstack-ironic00:39
*** Marga_ has quit IRC00:42
*** rloo has quit IRC00:51
*** saripurigopi has joined #openstack-ironic01:03
*** kan_afk is now known as kan__01:05
*** dttocs has quit IRC01:07
*** Marga_ has joined #openstack-ironic01:09
*** ijw has joined #openstack-ironic01:11
*** ijw has quit IRC01:16
*** david-lyle has joined #openstack-ironic01:20
*** saripurigopi has quit IRC01:21
*** zhenguo has joined #openstack-ironic01:38
*** rfchapman has quit IRC01:50
*** bnemec has quit IRC01:56
*** bnemec has joined #openstack-ironic01:59
*** harlowja is now known as harlowja_away02:03
Haomeng|2morning ironic:)02:19
*** david-lyle has quit IRC02:30
*** david-lyle has joined #openstack-ironic02:31
*** david-lyle has quit IRC02:36
openstackgerritMichael Davies proposed openstack/python-ironicclient: Cache negotiated api microversion for this server  https://review.openstack.org/17367402:44
*** dttocs has joined #openstack-ironic02:46
mrdaMorning Haomeng|202:46
Haomeng|2mrda: morning:)02:47
*** dttocs has quit IRC02:50
*** chenglch has joined #openstack-ironic03:19
*** coolsvap|afk is now known as coolsvap03:27
*** david-lyle has joined #openstack-ironic03:32
*** ramineni has joined #openstack-ironic03:46
openstackgerritMichael Davies proposed openstack/python-ironicclient: Cache negotiated api microversion for this server  https://review.openstack.org/17367403:54
openstackgerritMerged openstack/ironic: supports alembic migration for db2  https://review.openstack.org/17372103:57
openstackgerritMichael Davies proposed openstack/python-ironicclient: Add in support for a tox pypy target  https://review.openstack.org/17850103:58
*** Haomeng has joined #openstack-ironic04:33
*** Haomeng|2 has quit IRC04:36
*** yog_ has joined #openstack-ironic04:55
openstackgerritMerged openstack/ironic: Log attempts while trying to sync power state  https://review.openstack.org/16712205:06
*** sabah has joined #openstack-ironic05:09
*** sabah has quit IRC05:13
*** dttocs has joined #openstack-ironic05:29
*** Marga_ has quit IRC05:34
*** yog_ has quit IRC05:47
*** yog_ has joined #openstack-ironic06:05
*** igordcard_ has quit IRC06:06
*** ukalifon1 has joined #openstack-ironic06:08
*** pcaruana has joined #openstack-ironic06:09
*** jlvillal has quit IRC06:14
*** igordcard_ has joined #openstack-ironic06:14
*** jlvillal has joined #openstack-ironic06:16
dtantsurMorning Ironic06:27
*** yog_ has quit IRC06:36
*** pas-ha has joined #openstack-ironic06:43
*** jcoufal has joined #openstack-ironic06:49
*** dttocs has quit IRC06:53
openstackgerritMerged stackforge/ironic-discoverd: Remove profile matching from eDeploy plugin  https://review.openstack.org/17835307:08
*** ifarkas has joined #openstack-ironic07:11
*** rameshg87 has joined #openstack-ironic07:30
rameshg87good afternoon ironic07:30
*** dguerri is now known as _dguerri07:44
*** chenglch|2 has joined #openstack-ironic07:47
*** chenglch has quit IRC07:47
*** achanda has quit IRC07:48
*** Haomeng has quit IRC07:50
*** Haomeng has joined #openstack-ironic07:50
*** _dguerri is now known as dguerri07:52
*** dttocs has joined #openstack-ironic07:54
*** achanda has joined #openstack-ironic07:56
*** dttocs has quit IRC07:58
*** jistr has joined #openstack-ironic07:58
*** derekh has joined #openstack-ironic08:01
*** edcragg has joined #openstack-ironic08:08
*** ndipanov has quit IRC08:12
*** ndipanov has joined #openstack-ironic08:12
*** lucasagomes has joined #openstack-ironic08:14
kejiahumorning ironic08:19
openstackgerritMerged openstack/ironic: Improve strictness of DRAC test cases error checking  https://review.openstack.org/17769708:27
openstackgerritMerged openstack/ironic: Add logical name example to install-guide  https://review.openstack.org/17736708:28
*** achanda has quit IRC08:33
*** achanda has joined #openstack-ironic08:33
*** achanda has quit IRC08:35
*** dmellado has quit IRC08:41
*** kbs1 has joined #openstack-ironic08:49
*** dmellado has joined #openstack-ironic08:51
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Update the Vendor Passthru documentation  https://review.openstack.org/17828908:52
*** kbs has quit IRC08:52
*** MattMan has quit IRC09:00
*** MattMan has joined #openstack-ironic09:01
*** dtantsur is now known as dtantsur|brb09:08
*** athomas has joined #openstack-ironic09:12
*** dttocs has joined #openstack-ironic09:24
openstackgerritAnusha Ramineni proposed openstack/ironic: Delete neutron ports when the node cleaning fails  https://review.openstack.org/17857009:25
*** dttocs has quit IRC09:29
*** romcheg has joined #openstack-ironic09:30
*** chenglch|2 has quit IRC09:35
*** yog_ has joined #openstack-ironic09:40
openstackgerritShivanand Tendulker proposed openstack/ironic: Update iLO documentation for UEFI secure boot  https://review.openstack.org/17858209:56
openstackgerritGopi Krishna S proposed openstack/ironic: Add pxe_ucs and agent_ucs drivers to manage Cisco UCS servers  https://review.openstack.org/15973410:01
openstackgerritMerged openstack/ironic: Add maintenance mode example with reason  https://review.openstack.org/17789510:03
*** dtantsur|brb is now known as dtantsur10:18
*** arif-ali has quit IRC10:20
*** saripurigopi has joined #openstack-ironic10:20
*** dttocs has joined #openstack-ironic10:25
*** romcheg has quit IRC10:28
*** dttocs has quit IRC10:29
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Override PXE kernel cmdline via Glance property  https://review.openstack.org/17822910:29
*** romcheg has joined #openstack-ironic10:35
*** romcheg has quit IRC10:51
rameshg87lucasagomes: hi11:00
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Fix leftover of edeploy options removal  https://review.openstack.org/17860911:01
rameshg87lucasagomes: just one question - https://review.openstack.org/#/c/177859/4/specs/liberty/override-pxe-options-via-glance-property.rst11:01
lucasagomesrameshg87, hi there11:01
lucasagomessure11:01
rameshg87lucasagomes: do you want to consider the driver_info as well for some overrides ?11:01
rameshg87lucasagomes: some kernel command lines are dependent on the hardware11:01
rameshg87lucasagomes: for example console=/dev/ttyX11:01
rameshg87lucasagomes: it depends on which COM port the console is connected to11:01
lucasagomesright, but today we don't consider it right?11:02
lucasagomesI mean, we don't have anything in the driver_info for setting such options11:02
lucasagomeswould this be one addition on top of that work? Like extra parameters per node11:02
lucasagomesnode specific and not image specific11:03
lucasagomesbecause the node specific ones should also be applied for all other kernel command lines (the default in the template, and maybe the deploy ramdisk)11:03
lucasagomesrameshg87, &11:03
lucasagomes^*11:03
rameshg87lucasagomes: yeah11:04
rameshg87lucasagomes: so may be driver_info['kernel_command_line'] something like that11:05
lucasagomesrameshg87, sure I think it's a good idea, but you think it fits in the same spec?11:05
rameshg87lucasagomes: in a mixed environment, i think the consoles will differ and it's impossible to give a single argument in pxe_append_params11:05
lucasagomeslooks like another problem11:05
rameshg87lucasagomes: okay, so the spec is for image-related problem11:06
lucasagomesyeah, the image is things like root filesystem comes from a different source11:06
rameshg87lucasagomes: okay, may be another one, agree..11:06
lucasagomesI understand the node specific ones, but I see it's not related11:06
lucasagomesrameshg87, I will open a bug about the node specific parameters11:06
rameshg87lucasagomes: okay11:07
lucasagomesrameshg87, btw, I left a comment in the boot interface spec11:10
rameshg87lucasagomes: yeah i just replied now11:11
*** rameshg87 has left #openstack-ironic11:12
*** rameshg87 has joined #openstack-ironic11:12
rameshg87dtantsur: lucasagomes: let me know your thoughts on my reply to your comments11:13
lucasagomeshttps://bugs.launchpad.net/ironic/+bug/1449979 if you have a time, what you guys thing of?11:13
openstackLaunchpad bug 1449979 in Ironic "Less strictness for node names" [Undecided,New] - Assigned to Lucas Alvares Gomes (lucasagomes)11:13
lucasagomesrameshg87, will do11:13
rameshg87i will check after some time11:13
* rameshg87 goes home 11:13
openstackgerritMerged stackforge/ironic-discoverd: Fix leftover of edeploy options removal  https://review.openstack.org/17860911:13
*** rameshg87 has left #openstack-ironic11:13
*** kbs1 has quit IRC11:15
lucasagomesramineni, gotcha. So deploy interfaces validate their specific commands and give it to the boot interface11:16
lucasagomesso the boot knows how to pass it to the deploy ramdisk depending on the method it's booted11:17
lucasagomesyeah it looks good :-)11:17
*** romcheg has joined #openstack-ironic11:17
*** pas-ha has quit IRC11:18
*** dttocs has joined #openstack-ironic11:26
dtantsurI guess if 2 cores were confused by the spec, it's worth updating...11:29
dtantsurlucasagomes, do you feel like updating it as rameshg went home?11:29
*** pas-ha has joined #openstack-ironic11:30
*** dttocs has quit IRC11:31
lucasagomesdtantsur, sure, I'll get a quick sandwich soon, if ramesh doesn't update when I come back I will put a new patch-set up11:31
dtantsurcool!11:31
* dtantsur is busy with discoverd 1.1.0 release11:32
*** coolsvap is now known as coolsvap|afk11:33
*** arif-ali has joined #openstack-ironic11:40
*** lucasagomes is now known as lucas-hungry11:40
*** thrash|g0ne is now known as thrash11:43
*** kbs has joined #openstack-ironic11:47
*** dprince has joined #openstack-ironic12:04
*** openstackgerrit has quit IRC12:07
*** openstackgerrit has joined #openstack-ironic12:07
*** Marga_ has joined #openstack-ironic12:10
*** Marga_ has quit IRC12:10
*** Marga_ has joined #openstack-ironic12:11
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Add babel.cfg to MANIFEST.in  https://review.openstack.org/17864712:21
*** saripurigopi has quit IRC12:25
*** dttocs has joined #openstack-ironic12:27
*** dttocs has quit IRC12:31
openstackgerritMerged stackforge/ironic-discoverd: Add babel.cfg to MANIFEST.in  https://review.openstack.org/17864712:35
*** romcheg has left #openstack-ironic12:37
*** romcheg has joined #openstack-ironic12:37
*** EmilienM is now known as EmilienM|afk12:41
*** lifeless has quit IRC12:43
*** chlong has joined #openstack-ironic12:44
*** trown|outttypeww is now known as trown12:45
*** romcheg has quit IRC12:45
*** lucas-hungry is now known as lucasagomes12:47
*** jjohnson2 has joined #openstack-ironic12:50
*** romcheg has joined #openstack-ironic12:51
*** rloo has joined #openstack-ironic12:52
*** kkoski has joined #openstack-ironic13:01
*** ir2ivps9 has quit IRC13:01
*** dttocs has joined #openstack-ironic13:28
*** ir2ivps9 has joined #openstack-ironic13:32
*** rameshg87 has joined #openstack-ironic13:34
*** EmilienM|afk is now known as EmilienM13:37
*** lifeless has joined #openstack-ironic13:38
*** dttocs has quit IRC13:45
*** kan_ has joined #openstack-ironic13:47
rameshg87rloo: hi13:52
rloohi rameshg8713:52
rameshg87rloo: i just replied to your comments on the raid configuration spec.  may be after you read it, we can discuss if required.13:54
rloorameshg87: ok, will look in a few minutes.13:54
dtantsurrloo, morning13:55
rloohiya dtantsur, lucasagomes, and everyone else :)13:58
lucasagomesrloo, morning!13:58
rloohi dtantsur, lucasagomes: wrt 1449979 (which is a dup of another bug), do we need to discuss at the meeting? I think it is fine to relax.14:00
*** mgoddard has joined #openstack-ironic14:01
dtantsurwell, the rules were created for a reason, I guess :)14:01
dtantsurso e.g. mrda might object14:01
*** mgoddard1 has quit IRC14:01
*** zz_jgrimm is now known as jgrimm14:01
rloodtantsur: they were in the spec. i suspect mrda is fine with it, but I doubt that he'll be at the meeting anyway, so if that is the concern, just ask him.14:02
lucasagomesrloo, will take a look14:05
*** alexpilotti has joined #openstack-ironic14:05
-openstackstatus- NOTICE: gerrit has been restarted to clear a stuck events queue. any change events between 13:29-14:05 utc should be rechecked or have their approval votes reapplied to trigger jobs14:05
lucasagomesrloo, right... Ok, we can talk to mrda on IRC14:05
rloolucasagomes: i looked at the review https://review.openstack.org/#/c/165666/ and the dup bug https://bugs.launchpad.net/ironic/+bug/143437614:06
openstackLaunchpad bug 1434376 in Ironic "Node name rule relaxation" [Low,Confirmed]14:06
lucasagomesoh mrda was the one writting that bug14:06
lucasagomesyeah, fine14:07
rloolucasagomes: and nothing there seems to indicate that he is opposed to it. but yeah, just check with him.14:07
lucasagomesok14:07
lucasagomesthanks :-)14:07
* lucasagomes removes from the meeting14:07
NobodyCamgood morning ironicers14:07
rloog'morning NobodyCam14:08
lucasagomesNobodyCam, morning14:08
NobodyCammorning lucasagomes14:08
dtantsurNobodyCam, morning14:08
NobodyCammorning dtantsur & rloo too14:09
NobodyCam:)14:09
* NobodyCam needs coffee14:09
NobodyCambrb14:09
rloohi rameshg87. RAID is only part of zapping, not cleaning, right?14:12
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Support setting IPMI credentials in ironic_discoverd_ramdisk  https://review.openstack.org/17819314:15
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Include journalctl output when sending logs  https://review.openstack.org/17869314:15
rameshg87rloo: yeah, it's part of zapping14:15
BadCubmorning folks14:16
*** zhenguo has quit IRC14:17
rloorameshg87: just reviewing the zapping spec. it looks to me that someone could configure things so that they are done as part of zap and/or clean. it doesn't look like we can enforce that something is only used in zapping, not cleaning. which might not be a big deal.14:18
rloomorning BadCub14:18
kejiahumorning BadCub :)14:18
NobodyCammorning BadCub14:18
rameshg87rloo: yeah, i guess they can if the drivers make them configurable14:19
rameshg87rloo: i would expect raid interface have a hard-coded clean step priority as 014:19
BadCubmornin rloo kejiahu NobodyCam14:19
rloorameshg87: so what seems to be missing, at least from the specs, is a notion of 'validating' things before zapping/cleaning, like we have with deploy/power.14:20
rameshg87rloo: yeah14:20
rameshg87rloo: i just left a comment inline about that14:20
NobodyCam:)14:20
rameshg87rloo: we could mimic the same notion by renaming validate_raid_config() to validate()14:20
rameshg87rloo: so that we could invoke it as part of validating all interfaces as well as in the PUT api call itself.14:21
rloorameshg87: zapping is an explicit thing that a user has to invoke. so it would make sense to get an error if some validation occurred. how would you feel if it started zapping, did a few zap steps, and error'd out when RAIDing cuz it was missing the config info?14:21
rameshg87rloo: so, do you mean we could just have ability to call validate() just before zap was invoked ?14:22
* rameshg87 goes back to zapping spec14:23
openstackgerritZhenguo Niu proposed openstack/ironic: A small code rearrange  https://review.openstack.org/17814514:23
rloorameshg87: we have the validate API to validate interfaces. zap and clean aren't interfaces. but RAID will be. so I think it makes sense to have a RAIDInt.validate()14:24
rameshg87rloo: we could do that.  but when do you feel it should get called ?14:24
rloorameshg87: and I think the conductor, when it starts zapping (and maybe cleaning?) to get a list of interfaces involved with the clean steps and validate those interfaces.14:24
rameshg87rloo: hmm .. yeah makes sense14:25
rloorameshg87: so it gets called in the two places: user explicitly does validate API, and conductor calls before starting zapping.14:25
NobodyCammorning rameshg87 :)14:25
rameshg87rloo: right .. that's great14:25
rameshg87morning NobodyCam :)14:25
rloorameshg87: ok, i'll try to summarize in a comment in your spec :-)14:25
NobodyCam:)14:26
rameshg87rloo: sure .. thanks ..14:26
* rameshg87 goes and changes the spec right away14:26
*** Marga_ has quit IRC14:39
openstackgerritRamakrishnan G proposed openstack/ironic-specs: Add new boot interface in Ironic  https://review.openstack.org/16869814:40
rameshg87lucasagomes: dtantsur: ^^^ . made it more descriptive ..14:41
dtantsurthanks14:41
*** dttocs has joined #openstack-ironic14:42
lucasagomesthanks!14:42
openstackgerritRamakrishnan G proposed openstack/ironic-specs: Add generic RAID configuration spec for liberty  https://review.openstack.org/17321414:48
rameshg87rloo: ^^^. fixed the comment. :)14:48
*** alexpilotti has quit IRC14:49
rameshg87yuriyz: do you have code written for ssh reboot driver that you had proposed ?14:53
rameshg87yuriyz: i wanted to try it out.  i got some really old cpus and wanted to see if i could do something with it.14:54
*** tteggel_ has quit IRC14:54
yuriyzhi rameshg87 no and I abandoned the spec because I think now that we should use thing like software IPMI agent for this task14:56
*** tteggel has joined #openstack-ironic14:56
*** dttocs has quit IRC14:56
rameshg87yuriyz: do you mean software ipmi agent will run on the baremetal ?14:57
dtantsurifarkas or trown, hi, could you have a quick look at https://review.openstack.org/#/c/178193 and https://review.openstack.org/#/c/178693 please? both tested locally.14:57
yuriyzrameshg87 yes inside OS if node does not have BMC, idea from NobodyCam14:59
openstackgerritJarrod Johnson proposed stackforge/pyghmi: Add system UUID to inventory  https://review.openstack.org/17871315:03
trowndtantsur: will look shortly15:03
rameshg87yuriyz: oaky15:04
* rameshg87 leaves for the day 15:04
rameshg87good night folks15:04
*** rameshg87 has quit IRC15:04
*** ijw has joined #openstack-ironic15:05
NobodyCamnight rameshg8715:05
NobodyCambrb15:06
*** ijw has quit IRC15:06
*** ijw has joined #openstack-ironic15:06
*** alexpilotti has joined #openstack-ironic15:11
openstackgerritMerged stackforge/pyghmi: Add system UUID to inventory  https://review.openstack.org/17871315:12
ifarkasdtantsur, what's the '2' in ipmitool', 'user', 'set', 'name', '2', user?15:13
dtantsurifarkas, user ID :) they have fixed number of users, user ID 2 is usually the admin15:13
dtantsur(which is meh... but it works)15:14
*** dprince has quit IRC15:16
ifarkasdtantsur, lol, thanks :-)15:16
*** kan_ has quit IRC15:17
*** kan__ is now known as kan_afk15:17
*** ukalifon1 has quit IRC15:19
*** mgoddard has quit IRC15:20
*** mgoddard has joined #openstack-ironic15:21
*** ifarkas has quit IRC15:27
openstackgerritMerged stackforge/ironic-discoverd: Support setting IPMI credentials in ironic_discoverd_ramdisk  https://review.openstack.org/17819315:27
openstackgerritMerged stackforge/ironic-discoverd: Include journalctl output when sending logs  https://review.openstack.org/17869315:29
*** romcheg has quit IRC15:32
openstackgerritDmitry Tantsur proposed stackforge/ironic-discoverd: Drop unused CLI options  https://review.openstack.org/17873615:37
*** mtanino has joined #openstack-ironic15:38
*** MrAlex has joined #openstack-ironic15:41
NobodyCamlucasagomes: just checking to see if you'll have time today to address the latest comments on 178289. would luv to land that and wrap up the kilo release notes :)15:45
*** jcoufal has quit IRC15:45
*** dttocs has joined #openstack-ironic15:46
* dtantsur announced new version of discoverd and is ready to go home15:50
dtantsurg'night folks15:50
*** dtantsur is now known as dtantsur|afk15:50
trowncongrats dtantsur|afk and good night15:52
openstackgerritZhenguo Niu proposed openstack/ironic: When boot option is not persisted, set boot on next power on  https://review.openstack.org/17764215:54
lucasagomesNobodyCam, sure thing! will do15:55
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Update the Vendor Passthru documentation  https://review.openstack.org/17828915:55
lucasagomesNobodyCam, ^15:55
*** Sukhdev has joined #openstack-ironic15:55
* lucasagomes steps away a little15:56
* BadCub heads of to yet another conference call15:59
*** igordcard has joined #openstack-ironic16:02
*** ramineni has quit IRC16:03
*** dprince has joined #openstack-ironic16:03
*** trown is now known as trown|brb16:04
*** igordcard has quit IRC16:05
openstackgerritMerged openstack/ironic: Update iLO documentation for capabilities  https://review.openstack.org/17525716:05
*** jistr has quit IRC16:11
*** trown|brb is now known as trown16:14
*** chlong has quit IRC16:19
*** alexpilotti has quit IRC16:20
*** pcaruana has quit IRC16:22
NobodyCamawesome lucasagomes will look in a few TY16:28
*** trown is now known as trown|lunch16:30
*** dttocs has quit IRC16:30
*** dttocs has joined #openstack-ironic16:32
*** mgoddard has quit IRC16:43
*** achanda has joined #openstack-ironic16:43
*** mgoddard has joined #openstack-ironic16:45
*** dttocs has quit IRC16:49
*** ndipanov has quit IRC16:56
*** ndipanov has joined #openstack-ironic16:56
*** Shrews has quit IRC16:57
*** derekh has quit IRC16:59
*** Shrews has joined #openstack-ironic17:00
*** mgoddard has quit IRC17:00
*** mgoddard has joined #openstack-ironic17:00
NobodyCambrb17:02
*** igordcard has joined #openstack-ironic17:07
*** dttocs has joined #openstack-ironic17:09
*** pas-ha has quit IRC17:10
*** igordcard has quit IRC17:12
*** igordcard has joined #openstack-ironic17:13
*** igordcard has quit IRC17:13
*** igordcard has joined #openstack-ironic17:14
*** penick has joined #openstack-ironic17:14
*** igordcard has quit IRC17:17
*** dguerri is now known as _dguerri17:18
*** _dguerri is now known as dguerri17:18
NobodyCamanyone what to have a look at https://review.openstack.org/#/c/178289 (Update the Vendor Passthru documentation) before I land it?17:25
*** pelix has quit IRC17:26
NobodyCam+A'd17:29
*** dprince has quit IRC17:30
*** dprince has joined #openstack-ironic17:30
mgagneNobodyCam: god thanks for this update! :O17:31
*** e0ne has joined #openstack-ironic17:31
e0nedevananda: ping17:33
devanandae0ne: hi there!17:33
devanandahey folks - e0ne wants to know who on our side is going t owork on the cinder integration // attend their sessions or sprints on it17:34
e0neit would be good to have all yours concerns in one etherpad17:34
* BadCub goes to hunt food17:34
jrollI'm interested in cinder stuff17:34
e0neyuriyz: i hope you'll be on our session:)17:34
devanandae0ne: I started putting some of htem here https://etherpad.openstack.org/p/ironic-and-cinder17:35
e0nedevananda: great!17:35
e0nethanks17:35
* TheJulia ponders17:35
devanandae0ne: we have this proposal up too: https://review.openstack.org/#/c/149606/17:36
NobodyCammorning devananda jroll mgagne and e0ne17:37
TheJuliaJust an observation/outloud thought, It would be important to tie in those those thoughts/plans with the neutron stuffs since data and storage networks would ideally be separated17:37
jrollhi NobodyCam17:37
devanandaTheJulia: hm. I think these are quite separate concerns17:37
TheJuliaThey are, but ideally they would need to be cogniscent of each other17:37
e0nei'm very new in ironic, need to get deeper on it17:37
devanandaTheJulia: I agree the networks should be separate, but the neutron integratoin work is about managing VLAN configuration of the tenant network17:38
e0nebut i'll do my bes from cinder point of view17:38
devanandaand we can, for now, assume that both the IPMI and storage networks are separate and immutable17:38
devanandaTheJulia: I recognize that's a flawed assumption. many deployments will place storage and tenant network on the same phys interfaces, and use VLANs to separate .... which is probably where your statement comes from, and yea, ... challenging ....17:39
TheJulia:)17:39
devanandabut let's walk before we run17:39
TheJuliaindeed17:39
devanandahemna had a proposal a while back, I haven't found the link yet17:40
devanandathere was a very detailed wiki page17:40
e0nedevananda: https://wiki.openstack.org/wiki/Ironic/blueprints/cinder-integration - but it's outdated17:40
devanandayea, that's the one17:41
jrolle0ne: does nova know about iscsi strings right now?17:41
* jroll brb17:41
devanandarameshg & wanyen are also interested in boot from volume17:42
e0nejroll: not sure what exactly it knows. need to clarofy17:42
e0ne*clarify17:43
devanandatheir approach will be to do some sort of out-of-band firmware-based attach for diskless nodes, IIUC17:43
e0nedevananda: boot from volume could requires some additional features from hardware17:43
devanandae0ne: I know17:43
e0nethat's the main concern why we want to get attach first17:44
devanandae0ne: or rather, there are, again, two ways -- boot form volume is possible to in and out of band17:44
devanandaone can use ipxe to attach an iscsi volume and boot from that17:44
devanandalucasagomes: ^^17:44
e0nehm.. interesting...17:44
devanandaor one can use the BMC to attach at th efirmware layer and boot from that17:44
devanandaso ironic and cinder need to present a consistent API, regardless17:45
devanandaof the two use-cases (attach volume to running instance) & (boot from volume) I see a lot more interest for the latter17:45
devanandaand frankly, think it'll be easier to do17:45
devanandafrom an API design layer, at least, because it doesnt require any agent in the guest17:45
devanandabut I gotta get some food and jump on a call in 15min ...17:46
e0newe're going to provide 3 new APIs: 3 APIs, get_connector, discover_volume, detach_volume17:46
e0neok. i'll head to home. bb in ~1.5-2hours17:47
*** ramineni has joined #openstack-ironic17:49
*** trown|lunch is now known as trown17:53
*** Sukhdev has quit IRC17:55
*** e0ne has quit IRC17:57
*** dttocs has quit IRC17:59
*** dttocs has joined #openstack-ironic17:59
*** harlowja_away is now known as harlowja18:00
*** davidlenwell has quit IRC18:05
*** davidlenwell has joined #openstack-ironic18:05
*** dhellmann has quit IRC18:05
*** mgoddard has quit IRC18:08
*** harlowja has quit IRC18:16
*** harlowja has joined #openstack-ironic18:16
*** harlowja has quit IRC18:17
*** athomas has quit IRC18:18
*** harlowja has joined #openstack-ironic18:20
*** ijw has quit IRC18:23
*** harlowja has quit IRC18:29
*** harlowja has joined #openstack-ironic18:29
*** dguerri is now known as _dguerri18:30
*** e0ne has joined #openstack-ironic18:31
*** e0ne is now known as e0ne_18:31
*** e0ne_ is now known as e0ne18:32
*** e0ne is now known as e0ne_18:32
*** e0ne_ is now known as e0ne18:32
*** e0ne has quit IRC18:32
*** ramineni1 has joined #openstack-ironic18:33
*** _dguerri is now known as dguerri18:34
*** ramineni has quit IRC18:35
NobodyCamlucasagomes: just +A'd 17785918:38
*** hemna has joined #openstack-ironic18:39
*** dguerri has quit IRC18:42
NobodyCamquestion: on the kilo release notes we have: "IPMI Passwords are now obfuscated in REST responses. This may be disabled by a config option." anyone recall what that option is?18:45
*** ramineni1 has quit IRC18:46
*** e0ne has joined #openstack-ironic18:48
rlooNobodyCam: wrt the passwords, i thought it was via policy.json rule18:49
rlooNobodyCam: yeah, there's a 'show_password'18:49
*** kkoski has quit IRC18:54
*** Marga_ has joined #openstack-ironic18:56
NobodyCamrloo: this : https://github.com/openstack/ironic/blob/master/etc/ironic/policy.json#L318:56
rlooNobodyCam: yup18:56
rlooNobodyCam: I forgot how it works. I hope/think we coded it so by default it obfuscates the password.18:57
NobodyCamrloo: devananda: what do you think about changing the working there to s/by a config option/changing the policy.json rule/ ???19:00
NobodyCams/working/wording/19:00
*** kkoski has joined #openstack-ironic19:00
e0nedid i miss anything cinder related?19:01
NobodyCame0ne: nope!19:02
e0nethanks:)19:02
rlooNobodyCam: fine with me. Is 'rule' the right word for that? (I don't know the right terminology for the policy stuff)19:02
*** kkoski has quit IRC19:03
devanandarelease notes are all official and stuff now: https://wiki.openstack.org/wiki/ReleaseNotes/Kilo#OpenStack_Bare_Metal_service_.28Ironic.2919:03
*** kkoski has joined #openstack-ironic19:03
* BadCub likes [stuff]19:05
NobodyCamphew the TODO get removed19:05
rloodevananda: cool. but we can update the release notes still, right?19:05
devanandajroll, NobodyCam: for the neutron integration stuff, I haven't seen a spec yet for the ML2 extension, have you?19:05
NobodyCamI have not19:05
BadCubdevananda: I have not seen one myself. But I have been kinda Ostriched this week19:05
*** dhellmann has joined #openstack-ironic19:06
*** Marga_ has quit IRC19:07
devanandarloo: sorta. I mean, the wiki stays up, but the release notes may get copied elsewhere, once it is published19:08
*** jjohnson2_ has joined #openstack-ironic19:08
devanandarloo: so if there are changes you want made, best do them very soon19:08
NobodyCamdevananda: the only thing I had left was link lucasagomes vendor passthru docs19:09
devanandaNobodyCam: is that landed?19:09
NobodyCamis that still worth doing?19:09
NobodyCamdevananda: landing now19:09
devanandaif we have docs for it, yah19:09
rloodevananda: oh, I thought 'under the hood' it was pointing to https://wiki.openstack.org/wiki/Ironic/ReleaseNotes/Kilo19:09
NobodyCamhttps://review.openstack.org/#/c/178289/19:09
devanandarloo: nope19:10
devanandai'll stub that page and point back19:10
devananda*to the main release notes now19:10
NobodyCamdevananda: I will link as soon a we ^^^ merges19:10
*** jjohnson2 has quit IRC19:11
NobodyCams/a we/as/19:11
rloodevananda: ok. I had thought about putting together a list of things that were deprecated in kilo, in the upgrade notes, but i didn't get past 'thinking about'.19:11
devanandarloo: heh. we should actually do that19:12
devanandarloo: care to "think about them" in irc?19:12
rloodevananda: well, i seriously doubt i'll do that today.19:13
NobodyCambrb19:14
openstackgerritOpenStack Proposal Bot proposed openstack/ironic: Updated from global requirements  https://review.openstack.org/17723119:17
openstackgerritMerged openstack/ironic: Update the Vendor Passthru documentation  https://review.openstack.org/17828919:20
devanandarloo: i found only two things19:20
devanandaDeprecate pxe_deploy_{ramdisk, kernel} parameters to node.driver_info in favor of deploy_{ramdisk, kernel}19:20
devanandaDrivers implementing their own version of vendor_passthru() has been deprecated. Please update the code to use the @passthru decorator.19:20
openstackgerritOpenStack Proposal Bot proposed openstack/python-ironicclient: Updated from global requirements  https://review.openstack.org/17881519:23
rlooThese exceptions are deprecated, and will be removed after Kilo: CatalogUnauthorized (use KeystoneUnauthorized), CatalogFailure (use KeystoneFailure).19:23
rloodevananda: in a commit, maybe don't need to mention deprecated exceptions?19:24
devanandayea, not worth mentioning in release notes IMO19:25
devanandathere are several other deprecated things coming in from common openstack libs19:25
lucasagomesNobodyCam, thanks!19:25
lucasagomesdevananda, morning19:25
lucasagomessorry I had to leave before didn't say good night :-)19:25
lucasagomesso good night everyone19:26
rloolucasagomes: you're so polite :) Good night!19:26
lucasagomes:-) night19:26
*** lucasagomes is now known as lucas-dinner19:26
devanandalucas-dinner: g'night!19:27
rloodevananda: ohh, yuck. I suppose we should mention the deprecated stuff from libs too.19:27
devanandarloo: meh19:27
devanandarloo: I think that's a much higher level of detail than folks are looking for in tihs19:28
rloodevananda: so the operator or whoever should know to look at .. I guess our .sample has 'deprecated' or something like that next to deprecated options.19:28
devanandarloo: yes. common deprecated config options (eg, comfing from an oslo lib) will affect many projects ,and show up in all our config files19:28
devanandathat should be noted in documentation, sure, but not in the release notes19:29
rloodevananda: well, i think if someone is upgrading, they may need to know what changed in the configs. maybe we could mention see ..sample for any deprecated config options?19:29
devanandarloo: in operator documentatoin - definitely. anyone want to write some? :)19:29
* rloo didn't see what devananda just asked.19:29
devanandarelease notes should capture and summarize the most important changes, not every single change.19:30
rloodevananda: right. we just need the supporting documentation that isn't there. Anyway, that's why I thought the deprecated stuff could be mentioned in the upgrade notes (which I guess are part of the release notes now, or at least on the same wiki)19:31
NobodyCamdevananda: release notes updated.. wanta dbl check and make sure I did it in correct location19:31
*** lucas-dinner has quit IRC19:31
NobodyCamhttps://wiki.openstack.org/wiki/ReleaseNotes/Kilo#Hardware_Driver_Changes19:32
devanandaNobodyCam: ++19:36
devanandathere's one thing still missing from that, which is how to discover what the driver exposes19:37
jroll19:05:25       devananda | jroll, NobodyCam: for the neutron integration stuff, I haven't seen a spec yet for the ML2 extension, have you? <- I have no idea what ML2 extension you mean, so no? :)19:38
devanandajroll: the "its not a mechanism driver" thing19:38
devanandajroll: item 12 on https://etherpad.openstack.org/p/liberty-neutron-summit-topics19:39
jrolldevananda: I didn't know that was a thing yet, fwiw19:39
jrollother than what you've mentioned19:39
devanandajroll: oh. so yes. there are a couple companies that are, apparently, working on somehting19:40
devanandajroll: neutron's going to have a session to try to get it sorted before we all just fork your mechanism driver19:41
jrolllol19:41
devanandaoh, also item 63 on that 'pad. same thing afaict19:42
*** david-lyle has quit IRC19:44
*** david-lyle has joined #openstack-ironic19:44
NobodyCamoh wow item 64 on that pad is ummm.... ya... :/19:49
NobodyCamoh also item #8 under lightning sessions19:52
*** dprince has quit IRC19:52
*** arif-ali has quit IRC19:56
*** romcheg has joined #openstack-ironic20:01
devanandaironic<>neutron integration is apparently a thing20:03
*** romcheg1 has joined #openstack-ironic20:11
*** romcheg has quit IRC20:13
*** dguerri has joined #openstack-ironic20:18
NobodyCam:)20:19
*** romcheg has joined #openstack-ironic20:23
*** e0ne has quit IRC20:26
*** romcheg1 has quit IRC20:26
*** kkoski has quit IRC20:31
*** e0ne has joined #openstack-ironic20:36
*** kkoski has joined #openstack-ironic20:43
*** romcheg has quit IRC20:44
*** e0ne has quit IRC20:44
*** jcoufal has joined #openstack-ironic20:45
*** jcoufal has quit IRC21:03
*** jjohnson2_ has quit IRC21:03
*** kkoski has quit IRC21:11
*** Sukhdev has joined #openstack-ironic21:14
*** thrash is now known as thrash|g0ne21:15
*** arif-ali has joined #openstack-ironic21:19
mrdaMorning21:21
*** achanda has quit IRC21:22
*** yog_ has quit IRC21:24
BadCubmornin mrda21:33
*** yog_ has joined #openstack-ironic21:36
*** achanda has joined #openstack-ironic21:36
*** kkoski has joined #openstack-ironic21:36
*** kkoski has quit IRC21:36
*** harlowja is now known as harlowja_away21:36
*** harlowja_away is now known as harlowja21:37
*** gridinv has joined #openstack-ironic21:39
*** gridinv_ has joined #openstack-ironic21:44
*** gridinv has quit IRC21:46
NobodyCamjlvillal: ugg 800+ lines :-p21:48
jlvillalNobodyCam: What did I do??? :)21:49
jlvillalNobodyCam: Oh that patch :D21:49
jlvillalNobodyCam: On the bright side, almost zero logic changes.21:49
NobodyCamlol...21:50
* jlvillal has been sick the last few days and therefore quiet. Also playing with Abstract Syntax Trees....21:50
NobodyCamcan I ask why you added iter() in test_ssh to the .side_effect's?21:51
* NobodyCam hopes jlvillal feels better21:52
*** bnemec has quit IRC21:52
jlvillalNobodyCam: When adding autospec=True have to make side-effects iterables21:52
jlvillalNobodyCam: Done in other patches also.21:53
jlvillalNobodyCam: Thanks, I hope I get better soon too21:53
NobodyCam:)21:53
*** bnemec has joined #openstack-ironic22:02
*** dguerri is now known as _dguerri22:18
openstackgerritJohn L. Villalovos proposed openstack/ironic: ironic/tests/drivers: Add autospec=True and spec_set=  https://review.openstack.org/17704122:33
jlvillalNobodyCam: Thanks for catching that :)22:34
NobodyCam:-p22:34
bkero.oO (What are these guys on about at Acquia?)22:35
*** dttocs has quit IRC22:39
NobodyCambkero: ??22:42
bkeroOops, sorry. Wrong channel?22:42
NobodyCamhehehe ya don't think we've looked at drupal intergration22:43
NobodyCamoh but think if it each request spawns a whole Baremetal node to answer22:44
bkeroSounds like that time I rigged up xinetd which spun up hosts to respond to queries22:45
mrdaNobodyCam: Perhaps we can use drupal as deployment driver?22:45
NobodyCamlol22:46
NobodyCammorning mrda22:46
mrda;)22:46
NobodyCambkero: 1 vm per reply?22:46
bkeroNobodyCam: of course22:46
bkerobring-up time needs to be pretty fast. It's how you know if your cloud is pathetically slow22:46
NobodyCam:)22:47
bkero< 10ms should be the goal22:47
NobodyCamdon't think we're that quick ... yet...22:48
NobodyCam:)22:48
*** dttocs has joined #openstack-ironic22:49
NobodyCam<10min we can do ... no prob22:49
bkeroJeez, it sounds like you have some work to do then. Do you think you have higher priorities than getting this done? :P22:49
NobodyCamlol22:49
NobodyCamhehehee22:49
bkero.oO (maybe if I could get somebody to pay me to do it...)22:50
*** ndipanov has quit IRC22:54
*** penick has quit IRC22:59
*** dttocs has quit IRC23:01
*** Sukhdev has quit IRC23:14
*** penick has joined #openstack-ironic23:21
NobodyCamhumm https://review.openstack.org/#/c/15760223:27
NobodyCamGheRivero: happen to be around?23:28
*** penick has quit IRC23:30
*** chlong has joined #openstack-ironic23:50
* NobodyCam calls it a day. 23:57
NobodyCamg'night ya all23:58
BadCubg'night NobodyCam23:58
NobodyCam:)23:58
* BadCub has one more conference call to go :p23:59
NobodyCamoff to play with Kodi23:59
« Tuesday, 2015-04-28 Index Thursday, 2015-04-30 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!