Monday, 2020-09-14

« Sunday, 2020-09-13 Index Tuesday, 2020-09-15 »
jandersgood morning Ironic o/00:02
jandersTheJulia ftarasenko rpioso ajya|afk I spotted the discussion on SSD secure erase in the logs and it's a topic I'm interested in so wanted to share some thoughts/questions. Is there any way in which we can use SSD trim/discard in cleaning? I think this would be the right tool for the job both in terms of speed and maximising SSD lifetime. Do Dell00:24
janderscontrollers support this? If so, can we leverage this in ironic via supported BMCs?00:24
rpiosojanders: iDRAC supports secure erase of SSDs. Please see the description of the WS-Man DCIM_RAIDService.SecureErase method in https://downloads.dell.com/solutions/general-solution-resources/White%20Papers/Dell_SimpleRAID_Profile.pdf. It is in section 8.30. DCIM_RAIDService.SecureErase (), starting on p. 73. Similar is offered via Redfish, too --00:38
rpiosohttps://topics-cdn.dell.com/pdf/idrac9-lifecycle-controller-v4x-series_api-guide_en-us.pdf.00:38
rpiosoajya|afk, ftarasenko, janders, TheJulia: Are you interested in discussing secure erase during the PTG?01:01
jandersrpioso thank you for the links, will have a look through01:35
jandersrpioso regarding PTG - I am! :)01:35
jandersrpioso reading through the document I see "8.30. DCIM_RAIDService.SecureErase () The Secure Erase operation or full initialization on a PCIe SSD overwrites all blocks in an encrypted way and all01:38
jandersthe data on the PCIe SSD device will be permanently lost."01:38
jandersis it possible to secure-erase the data by trim/discard as opposed to overwrite ( https://en.wikipedia.org/wiki/Trim_(computing) )?01:38
jandersidea = maximising SSD lifetime01:38
rpiosojanders: yw :-)01:39
rpiosojanders: Dunno. I would have to check. Got it.01:39
jandersrpioso thanks again, that would be awesome :)01:40
janders(and would go very nicely with fastrack deploys, kexec, etc - we could cut deployment/cleaning overheads very very slim)01:41
jandersif this is useful at some point, maybe it's possible test this out using an SSD running in pass-thru mode and this https://man7.org/linux/man-pages/man8/blkdiscard.8.html01:43
*** zzzeek has quit IRC02:20
*** zzzeek has joined #openstack-ironic02:22
*** rh-jelabarre has joined #openstack-ironic02:33
openstackgerritwu.shiming proposed openstack/ironic master: Remove install unnecessary packages  https://review.opendev.org/75163603:10
*** Qianbiao has joined #openstack-ironic03:17
*** mkrai has joined #openstack-ironic03:23
*** rh-jelabarre has quit IRC03:25
*** zzzeek has quit IRC03:50
*** zzzeek has joined #openstack-ironic03:51
*** uzumaki has joined #openstack-ironic04:49
*** mkrai has quit IRC04:59
*** mkrai has joined #openstack-ironic05:00
*** Qianbiao has quit IRC05:15
*** abdysn has joined #openstack-ironic05:40
*** Qianbiao has joined #openstack-ironic05:50
openstackgerritwu.shiming proposed openstack/python-ironicclient master: Remove install unnecessary packages  https://review.opendev.org/75164906:05
openstackgerritSteve Baker proposed openstack/ironic master: Add status_code argument to @method.expose  https://review.opendev.org/74367706:18
openstackgerritSteve Baker proposed openstack/ironic master: Add @method.body decorator  https://review.opendev.org/74367806:18
openstackgerritSteve Baker proposed openstack/ironic master: Duplicate trait validation with jsonschema  https://review.opendev.org/74577806:18
openstackgerritSteve Baker proposed openstack/ironic master: New argument validate decorator  https://review.opendev.org/74279506:18
openstackgerritSteve Baker proposed openstack/ironic master: WIP Utility functions for REST API JSON handling  https://review.opendev.org/75116006:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert allocations endpoint to plain JSON  https://review.opendev.org/74367906:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert bios endpoint to plain JSON  https://review.opendev.org/74547506:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert chassis endpoint to plain JSON  https://review.opendev.org/74547606:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert conductors endpoint to plain JSON  https://review.opendev.org/74577906:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert deploy_templates endpoint to plain JSON  https://review.opendev.org/74578006:18
openstackgerritSteve Baker proposed openstack/ironic master: Convert drivers endpoint to plain JSON  https://review.opendev.org/74186506:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert event endpoint to plain JSON  https://review.opendev.org/74662306:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert nodes endpoint to plain JSON  https://review.opendev.org/74782006:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert ports endpoint to plain JSON  https://review.opendev.org/75012006:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert portgroups endpoint to plain JSON  https://review.opendev.org/75086406:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert ramdisk endpoint to plain JSON  https://review.opendev.org/75085706:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert volume endpoint to plain JSON  https://review.opendev.org/75085806:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert volume/connectors endpoint to plain JSON  https://review.opendev.org/75085906:19
openstackgerritSteve Baker proposed openstack/ironic master: Convert volume/targets endpoint to plain JSON  https://review.opendev.org/75086006:19
openstackgerritSteve Baker proposed openstack/ironic master: Expunge the internal version of WSME  https://review.opendev.org/75086106:19
arne_wiebalckGood morning, ironic!06:42
jandersgood morning arne_wiebalck06:43
arne_wiebalckHey janders o/06:44
*** jtomasek has joined #openstack-ironic06:58
*** mkrai has quit IRC07:16
*** mkrai_ has joined #openstack-ironic07:16
iurygregorygood morning arne_wiebalck janders and Ironic!07:19
arne_wiebalckhey iurygregory o/07:19
ftarasenkogm, Ironic!  janders, rpioso, regarding secure erase via Dell RAID Controller, I think, DCIM_RAIDService.ReKey method can be used. The bad is that security is not supported on 330 series controller, so it's not a solution for all Dell servers.07:32
ftarasenkoPutting SSD's in pass-through mode seems a solution for my servers, but haven't tested yet.07:33
*** mkrai_ has quit IRC07:35
*** jtomasek has quit IRC07:39
*** jtomasek has joined #openstack-ironic07:41
uzumakigood morning ironic! o/07:45
iurygregorymorning uzumaki o/07:45
uzumakihow you doing iurygregory ? como estas?07:45
iurygregorydoing good =) and you?07:46
uzumakiestoy bien, gracias =)07:46
uzumakiany interesting holidays in CZ?07:46
iurygregoryI think we have one on 28 Sep07:47
openstackgerritArne Wiebalck proposed openstack/ironic-inspector master: Add NVIDIA V100 to known accelerator devices  https://review.opendev.org/75166007:55
*** uzumaki has quit IRC07:56
iurygregoryreally ironic-inspector was the only one that failed to merge the l-c changes -.-' facepalm07:58
*** mkrai_ has joined #openstack-ironic08:24
openstackgerritMerged openstack/ironic master: Follow-up patch for One Button Secure Erase clean step  https://review.opendev.org/75062408:27
*** Abdallahyas has joined #openstack-ironic08:30
*** abdysn has quit IRC08:34
*** derekh has joined #openstack-ironic08:42
*** lucasagomes has joined #openstack-ironic08:55
openstackgerritMerged openstack/ironic-python-agent-builder master: Use converted ipa jobs  https://review.opendev.org/74972308:58
QianbiaoGood morning ironic.09:08
QianbiaoMorning arne_wiebalck iurygregory uzumaki09:08
QianbiaoMorning everyone09:09
QianbiaoHi, i have an issue while using ironic-inspector to add infiniband ports.09:09
iurygregorymorning Qianbiao09:09
QianbiaoI create the inspect ipa with "disk-image-create ironic-python-agent-ramdisk mellanox centos-minimal devuser -o ipa-centos-8-mlnx"09:10
QianbiaoBut IB ports is not detected.09:10
*** sshnaidm|pto is now known as sshnaidm09:10
openstackgerritMerged openstack/ironic-python-agent master: Fix backup node lookup  https://review.opendev.org/75082309:11
openstackgerritVerification of a change to openstack/ironic-python-agent failed: Move ironic-standalone-ipa-src to Ubuntu Focal  https://review.opendev.org/74821309:11
arne_wiebalckHey Qianbiao o/09:14
QianbiaoHey o/09:14
*** tosky has joined #openstack-ironic09:31
*** uzumaki has joined #openstack-ironic09:50
*** uzumaki has quit IRC10:00
*** uzumaki has joined #openstack-ironic10:04
*** dtantsur|afk is now known as dtantsur10:27
dtantsurmorning ironic10:27
*** Lucas_Gray has joined #openstack-ironic10:30
*** Lucas_Gray has quit IRC10:42
QianbiaoMorning dtantsur10:43
jandersgood morning dtantsur10:45
*** mkrai_ has quit IRC10:47
*** mkrai__ has joined #openstack-ironic10:47
*** Lucas_Gray has joined #openstack-ironic10:47
*** Lucas_Gray has quit IRC10:51
*** Wryhder has joined #openstack-ironic10:51
*** Wryhder is now known as Lucas_Gray10:51
*** uzumaki has quit IRC10:56
openstackgerritVerification of a change to openstack/ironic-python-agent failed: Remove unused jobs  https://review.opendev.org/74972510:58
iurygregorygood morning dtantsur11:01
*** uzumaki has joined #openstack-ironic11:08
*** mkrai__ has quit IRC11:37
dtantsurfolks, could I get a 2nd review on https://review.opendev.org/#/c/747136/ please? it's an API/RPC change, we should get it in ASAP.11:40
patchbotpatch 747136 - ironic - Accept and use a TLS certificate from the agent - 9 patch sets11:40
openstackgerritDmitry Tantsur proposed openstack/ironic-python-agent-builder master: tinyipa: use python 3 for building on all systems  https://review.opendev.org/75176911:48
*** Lucas_Gray has quit IRC11:50
*** rh-jelabarre has joined #openstack-ironic11:50
*** rh-jelabarre has quit IRC11:51
*** rh-jelabarre has joined #openstack-ironic11:51
*** mkrai__ has joined #openstack-ironic11:52
*** Lucas_Gray has joined #openstack-ironic11:53
*** Lucas_Gray has quit IRC11:58
*** Abdallahyas has quit IRC12:00
*** rh-jelabarre has quit IRC12:00
*** rh-jelabarre has joined #openstack-ironic12:00
openstackgerritMerged openstack/tenks master: Fix ansiblelint 208  https://review.opendev.org/74921112:10
*** Goneri has joined #openstack-ironic12:11
*** abdysn has joined #openstack-ironic12:11
*** priteau has joined #openstack-ironic12:15
openstackgerritMerged openstack/ironic-inspector master: Fix l-c and mysql GRANT syntex for Ubuntu Focal  https://review.opendev.org/75121812:17
iurygregoryfinally \o/12:30
iurygregorydtantsur, looking now12:31
janderssee you tomorrow Ironic o/12:34
iurygregorybye janders12:39
iurygregorydtantsur, done12:41
dtantsurthx12:41
* iurygregory goes back testing bifrost with ipe12:42
*** mkrai__ has quit IRC12:50
openstackgerritMerged openstack/ironic-python-agent master: Move ironic-standalone-ipa-src to Ubuntu Focal  https://review.opendev.org/74821312:52
Qianbiaohi dtantsur does inspect support detect infiniband port?12:55
dtantsurQianbiao: I assume so based on the patches I've seen. I don't have any experience with IB though.12:55
Qianbiaomellanox infiniband port. I try to add mellanox element on inspect ipa ramdisk, but not work.12:55
dtantsurQianbiao: you can ask janders tomorrow, he seems to have had some experience12:55
Qianbiaook thanks for the information12:56
jandersdtantsur Qianbiao it used to work12:56
janders(with some catches)12:56
Qianbiaojanders with solution: "add mellanox element on inspect ipa ramdisk" ?12:56
jandersyeah that would be one12:57
QianbiaoI am try working with mellanox CX512:57
janderswhat servers is this on?12:57
Qianbiaocentos712:57
jandershow about server make/model?12:57
janderssometimes there were some hardware-specific catches IIRC12:58
Qianbiaoi am working with huawei ibmc12:58
QianbiaoI am trying make ramdisk by "disk-image-create ironic-python-agent-ramdisk mellanox centos-minimal devuser -o ipa-centos-8-mlnx"12:59
jandersare you booting off IB as well, or just recording IB details?13:00
Qianbiaorecording IB guid13:01
jandersgood, this should be simpler13:01
QianbiaoI try to get IB guid when inspecting13:01
jandersdo you have another CX5 ethernet port you're booting off?13:01
Qianbiaojanders we did not dhcp through IB port13:02
Qianbiaowe have ethernet port for pxe13:02
QianbiaoIB port will be used after provision13:02
janderson the CX5 or on the motherboard?13:02
Qianbiaoon CX513:02
Qianbiaolet me take a picture13:03
jandersare you currently able to PXE-boot over the Ethernet port, just not getting the GUID details?13:03
Qianbiaoyes13:03
Qianbiaohttp://paste.openstack.org/show/797833/13:03
jandersok! then adding the element should help13:04
Qianbiaowe are pxe on eno2/eno1.13:04
Qianbiaoand try to get guid of ibs4f113:04
*** Lucas_Gray has joined #openstack-ironic13:04
janderssometimes getting PXE to work over any port is tricky depending on the hardware, but it sounds like this is not a problem, which is good13:04
Qianbiaook i will try again to make sure nothing wrong operated before.13:05
jandersis this your internal lab, or is this for a customer of yours?13:05
Qianbiaointernal lab13:05
janderscool! easier to get debug info if need be13:05
QianbiaoI am testing on it, and will delivery the solution to our customer13:05
QianbiaoThis is the start, i need to using the guid to setup partition for IB network in neutron..13:06
Qianbiaoneed to create a new ml2 driver maybe.13:06
jandersright!13:06
janderslet us know how you go with the custom element, hopefully that will work13:07
jandersI think if it's properly in it should be mentioned in the introspection logs IIRC13:07
jandersallright! it's getting late here, I will see you later13:07
jandersgood luck Qianbiao13:07
QianbiaoIam using https://github.com/openstack/diskimage-builder/tree/master/diskimage_builder/elements/mellanox13:08
Qianbiaothanks, will try. and let you know.13:08
*** uzumaki has quit IRC13:09
*** irclogbot_2 has quit IRC13:19
TheJuliagodo morning13:24
TheJuliagood13:24
iurygregorygood morning TheJulia =)13:24
*** irclogbot_0 has joined #openstack-ironic13:25
*** rloo has joined #openstack-ironic13:26
iurygregorydtantsur, if I want to add a way to enable the ipe from the bifrost CLI would be something like http://paste.openstack.org/show/797834/ that is needed right?13:28
*** mnaser has quit IRC13:32
*** mnaser has joined #openstack-ironic13:32
*** mnaser has quit IRC13:32
*** mnaser has joined #openstack-ironic13:32
dtantsuriurygregory: yep, but I'd use --enable-prometheus-exporter13:33
dtantsurmorning TheJulia, will you believe me if I tell you there is another bug with agent tokens?13:34
iurygregorydtantsur, ack =)13:34
* dtantsur tries to reproduce again13:34
dtantsuryaaaay, power off still does not wipe the token despite me fixing it...13:35
TheJuliaSweet!13:35
TheJuliaSo out of band power off while in a workflow?13:36
TheJuliaOr workflow killing the power?13:36
dtantsurpower off via CLI13:36
*** tkajinam has quit IRC13:37
*** jtomasek has quit IRC13:39
*** tzumainn has joined #openstack-ironic13:41
dtantsurTheJulia: ehmm, we have broken is_agent_token_pregenerated13:42
dtantsurthat's why it regressed13:42
dtantsurno, wait, false alert13:42
*** cdearborn has joined #openstack-ironic13:47
*** ricolin has quit IRC13:52
* dtantsur has spotted the issue and is in deep facepalm13:57
openstackgerritChristopher Dearborn proposed openstack/ironic master: Redfish driver firmware update  https://review.opendev.org/74961914:00
*** uzumaki has joined #openstack-ironic14:01
openstackgerritJulia Kreger proposed openstack/ironic master: Guard conductor from consuming all of the ram  https://review.opendev.org/72648314:01
openstackgerritMerged openstack/tenks master: Use importlib to take place of imp module  https://review.opendev.org/74881814:02
TheJuliaOn the etherpad, https://etherpad.opendev.org/p/IronicWhiteBoard, line 167, is that still relevent?14:02
bfourniehi rpioso, re: discussion on this last week, the missing PowerState with FC640 - https://bugzilla.redhat.com/show_bug.cgi?id=1873305, they tried with iDRAC firmware 4.10.10.10 but are seeing the same issue14:03
openstackbugzilla.redhat.com bug 1873305 in Bare Metal Hardware Provisioning "Failed to power on /inspect node when using Redfish protocol" [High,New] - Assigned to dtantsur14:03
openstackgerritDmitry Tantsur proposed openstack/ironic master: Also wipe agent token on manual power off or reboot  https://review.opendev.org/75181314:03
dtantsurTheJulia: join me in the facepalm ^^^14:03
iurygregorydamm if14:04
TheJuliasigh14:04
bfourniedtantsur: could this be issue with inspection or clean failing and subsequent attempts also fail due to token issues?14:05
dtantsurbfournie: not 100% sure14:05
arne_wiebalckTheJulia for https://review.opendev.org/726483, I am starting systematic tests with 200 physical nodes (without your patch for now, to get the baseline behaviour, and also to see how scaling out controllers helps)14:05
patchbotpatch 726483 - ironic - Guard conductor from consuming all of the ram - 15 patch sets14:05
dtantsurTheJulia: I also wonder if we should wipe agent_url on power off14:06
bfourniedtantsur: k14:06
dtantsursince DHCP may deliver us a new IP next time14:06
rpiosoftarasenko: There's also DCIM_RAID.RemoveControllerKey(). Its doc states, "CAUTION: All encrypted drives shall be erased." I'm under the impression DCIM_RAID.ReKey() does not erase. At least, that profile doesn't state it does.14:06
TheJuliadtantsur: If we have a valid token I think we'll update it now14:06
rpiosoGood morning, ironicers14:06
TheJuliaI _think_14:06
rpiosobfournie: Thank you for the update. Drats!14:07
TheJuliagranted dhcp doing that is a possibility but also means the network config and dhcp is broken14:07
dtantsurTheJulia: nope, we outright reject any changes14:07
TheJuliahmm14:07
TheJuliaWe likely should then14:07
TheJuliabecause people have broken networks14:07
dtantsurokay, updating the patch14:07
TheJuliak14:07
*** sshnaidm is now known as sshnaidm|afk14:07
QianbiaoHello, is this a ipa compatible issue: Failed looking up node with addresses '00:2e:c7:dc:38:c3,74:5a:aa:63:07:62,74:5a:aa:63:07:60,00:2e:c7:dc:38:c2' at http://192.168.43.144/baremetal, status code: 40414:08
QianbiaoAm i building ramdisk with wrong ipa?14:09
dtantsurQianbiao: ironic-inspector cannot match the MAC addresses with anything it has in its cache14:09
QianbiaoHmm. but 74:5a:aa:63:07:62 this mac can be list in port list command14:09
openstackgerritDmitry Tantsur proposed openstack/ironic master: Also wipe agent token on manual power off or reboot  https://review.opendev.org/75181314:10
dtantsurQianbiao: you need to check ironic-inspector logs for what it's going to use for lookup14:10
Qianbiaosure chekcing.14:10
*** ricolin has joined #openstack-ironic14:11
rpiosoarne_wiebalck, rpittau: I'm thinking it's time for a Redfish Interop Profile sync-up. How about late this week or sometime next week? Doodle it?14:13
Qianbiaodtantsur It says:  Look up error: Could not find a node for attributes {'bmc_address': [], 'mac': []}14:14
Qianbiaolooks like a compatible issue?14:14
dtantsuroh, hmm, interesting, I've confused two lookups14:14
dtantsurthe first lookup was from ironic, not ironic-inspector and it does use ports14:14
dtantsurironic-inspector fails.. I don't remember if IB ports are used during lookup14:14
Qianbiaodtantsur nothing about ib port for now.14:15
Qianbiaothose mac address are all eth port14:15
Qianbiaoinspect complians "Failed to load collector {'extra_hardware'}"14:16
Qianbiaois this related?14:16
* rpioso will be right back. Time for a reboot salute :-)14:17
TheJuliarpioso: I think rpittau|afk is on pto.14:19
*** mkrai has joined #openstack-ironic14:19
arne_wiebalckrpioso: this and next week are both fine with me, should be pretty flexible14:20
QianbiaoHe said: four weeks IIRC14:21
openstackgerritJulia Kreger proposed openstack/ironic master: Minor agent version code cleanup  https://review.opendev.org/74955214:22
*** kashifkhan has joined #openstack-ironic14:22
*** uzumaki has quit IRC14:24
Qianbiaodtantsur I found :: [-] Not updating iptables - no changes in MAC list [u'28:a6:db:a9:92:27', u'84:a9:c4:6f:3f:c5', u'70:c7:f2:69:07:62'] {{(pid=12303) sync /opt/stack/ironic-inspector/ironic_inspector/pxe_filter/iptables.py:120}}14:26
Qianbiaothe update list does not contains the target mac14:26
Qianbiaosomething wrong?14:26
*** uzumaki has joined #openstack-ironic14:28
QianbiaoIt just success...14:30
Qianbiaoi changed nothing.14:30
rpiosoTheJulia: Thank you for informing me rpittau|afk is out.14:32
rpiosoTheJulia: Do you expect him back next week?14:33
Qianbiaorpioso he said 4 weeks IIRC14:33
bfournierpioso: he will be out for 3 more weeks14:33
TheJuliadtantsur: I was thinking 2 more weeks, but it could be 314:34
TheJuliadtantsur: anything from bifrost that you want to add for potential review items?14:34
rpiosobfournie, TheJulia, Qianbiao: Thank you!14:34
rpiosoarne_wiebalck: Looks like rpittau|afk will be out for a while. How about we proceed and sync up with him later?14:36
rpiosoarne_wiebalck: We could doodle Wednesday-Friday this week and next and see where it falls :-)14:37
arne_wiebalckrpioso: sounds good!14:37
TheJuliaDoes anyone remember if last week we reached consensus on the make automated clean work in both directions rfe?14:41
TheJuliadtantsur: ^^^14:42
dtantsurTheJulia: I think JayF was "whatever" in the end, so we sort of did?14:42
dtantsurI do intend to make this feature opt-in from the operator's perspective after that conversation14:42
TheJuliaokay, then I think that is fine and I'm totally cool with that. I think that is the right way to head in14:43
dtantsurbifrost items.. lemme see what we've already got14:43
rpiosoarne_wiebalck: I created the doodle: https://doodle.com/poll/s3dttwbbuvc2n6dx14:43
rpiosoarne_wiebalck: Could we use your video bridge?14:44
arne_wiebalckrpioso: sure, once we settle on the slot, I will create a meeting14:45
rpiosoarne_wiebalck: Thank you!14:49
QianbiaoHello, is there a method to confirm whether a diskimage element is correctly append.14:49
arne_wiebalck rpioso: Thanks for driving this!14:49
rpiosoarne_wiebalck: :-)14:50
*** abdysn has quit IRC14:53
*** stendulker has joined #openstack-ironic14:57
JayFdtantsur: TheJulia: I mean, I'm not happy with the way things are. I won't be happy with the way things will be. We need to take this ask -- for an anti-feature of leaving data preserved on "available" nodes -- as a call to action to handle data-preservation cases better within Ironic, without doing things like that.14:59
iurygregoryo/15:00
TheJulia#startmeeting ironic15:00
TheJuliao/15:00
openstackMeeting started Mon Sep 14 15:00:51 2020 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.15:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:00
*** openstack changes topic to " (Meeting topic: ironic)"15:00
openstackThe meeting name has been set to 'ironic'15:00
JayFo/15:00
iurygregoryo/15:01
mgoddard\o15:01
rpioso\o15:01
bfournieo/15:01
erbarro/15:01
stendulkero/15:01
TheJuliaGood morning everyone!15:01
arne_wiebalcko/15:01
TheJuliaOur Ironic meeting agenda can be found on the wiki.15:01
TheJulia#link https://wiki.openstack.org/wiki/Meetings/Ironic#Agenda_for_next_meeting15:01
TheJulia#topic CFP is open for the Forum15:01
*** openstack changes topic to "CFP is open for the Forum (Meeting topic: ironic)"15:01
*** kashifkhan has quit IRC15:01
TheJulia#undo15:01
openstackRemoving item from minutes: #topic CFP is open for the Forum15:01
TheJulia#topic Announcements / Reminders15:02
*** openstack changes topic to "Announcements / Reminders (Meeting topic: ironic)"15:02
dtantsuro/15:02
rpiosoTheJulia: We have a doodle up for an ironic Redfish Interop Profiles sync-up: https://doodle.com/poll/s3dttwbbuvc2n6dx15:02
TheJulia#info CFP for the forum is supposed to be open. Apparently the submission form is closed but that should hopefully get fixed soon. we need to submit any items _today_15:02
TheJulia#info Since we are in the home stretch for the victoria cycle, reviewers, please review changes this week!15:03
TheJulia#info A gentle reminder (for even me) to self triage bugs!15:03
TheJulia#link https://docs.openstack.org/ironic/latest/contributor/bugs.html#triaging-guide15:03
TheJulia#info vPTG time slots have been selected - 2PM-4PM UTC ?Monday through Wednesday?15:03
TheJulia#info Redfish Inter-op profiles sync-up call doodle has been proposed15:04
TheJulia#link https://doodle.com/poll/s3dttwbbuvc2n6dx15:04
rlooo/15:04
TheJuliaDoes anyone have anything else to announce or remind us of?15:04
QianbiaoTheJulia not sure whether is suitable to remind here. the ramdisk deploy interface issue, may someone take a look at.15:05
QianbiaoI have add it to the white board in vendor list15:05
TheJuliaOkay,15:06
cdearborno/15:06
TheJuliaIt is not really an item for the general community wide reminders, but good to know15:07
TheJuliaAnyone have anything else?15:07
TheJuliaI guess not!15:08
TheJuliaLets carry on!15:08
TheJuliaNo action items from our previous meeting, so we will carry on directly to reviewing subteam status reports15:08
TheJulia#topic Review subteam status reports15:09
*** openstack changes topic to "Review subteam status reports (Meeting topic: ironic)"15:09
TheJulia#link https://etherpad.opendev.org/p/IronicWhiteBoard15:09
TheJuliaStarting at line 282.15:09
TheJuliadtantsur: do you want me to propose a 8.3.1 release for bifrost?15:10
openstackgerritMerged openstack/bifrost master: Let us finish the rabbit hunt  https://review.opendev.org/74826515:10
dtantsurTheJulia: if you don't mind. just remember that it's a manual process for now15:11
TheJuliamanual as in manual tag or manual as in manual edit the file?15:11
dtantsurTheJulia: as in go to #openstack-release and tell them git hash, branch and version15:12
TheJulia#sadness15:12
iurygregoryoh we can't push a patch? O.o15:13
iurygregoryI can do that since TheJulia seems sad =)15:13
TheJuliaiurygregory: by all means, and if they object to it point them to the IRC log!15:13
iurygregoryTheJulia, ack =)15:13
openstackgerritDmitry Tantsur proposed openstack/ironic master: Add redfish options to the generated docs  https://review.opendev.org/75183915:14
dtantsuriurygregory: anyway ping one of us to double-check15:14
iurygregorydtantsur, sure15:14
TheJuliadtantsur: looks like the most important ramdisk tls patch is the eventlet monkey patching? is that correct?15:14
dtantsurmmm, no15:15
JayFthat patch was integrated in the static TLS support15:15
dtantsurwe're no longer blocked by that15:15
JayFit's all one patch now15:15
dtantsuryes, this15:15
JayFI punted on the "we should monkey patch everything" problem, fell back to just monkey patching socket.15:15
dtantsurauto-TLS works in the CI once the IPA patch is approved15:15
TheJuliaokay then, so basically that feature is approved and will hopefully merge today *crosses fingers, toes, etc*15:16
JayFOne question that I thought of: do we think we should also have *manual* TLS in a job? I'm leaning towards no, but wanted that to be an explicit decision.15:16
* dtantsur ponders15:16
TheJuliaiurygregory: line 408 Make CI manageable: You have two patches up. What is the path forward on those?15:16
dtantsurJayF: I'm not sure how much coverage of *our* code it would add15:17
*** JamesBenson has joined #openstack-ironic15:17
* TheJulia is unsure and is leaning towards no on that question because of the amount of resources that would be needed to test it15:17
iurygregoryTheJulia, I'm still fixing things in https://review.opendev.org/74188615:17
patchbotpatch 741886 - ironic - Add tempest default_boot_option and altflavor - 16 patch sets15:17
JayFReally, I think it'd add coverage of the listen_tls agent conf (minimal) and the ipa-tls static config element15:17
iurygregoryto get all green15:17
TheJuliaiurygregory: ack, please add that to the notes.15:17
dtantsurJayF: we could add such a job to IPA-builder only15:18
TheJuliadtantsur: I like that idea15:18
dtantsurbut it will need some devstacking as well15:18
JayFeh, just config for Ironic to ignore validity of IPA certs, then you get the TLS generation and it gets "free"-ish with a properly configured ramdisk15:18
TheJuliaiurygregory: any estimate on a CI job for dhcp-less?15:19
dtantsurwe build the ramdisk somewhere inside devstack, and it's not overly customizable. I don't remember if you can add an element.15:19
dtantsurnot a huge deal, of course15:19
iurygregoryTheJulia, no estimate atm15:19
JayFah, makes sense, shouldn't be impossible15:19
TheJuliaSeems like we're good to proceed?15:20
JayF +115:22
TheJulia#topic Priorities for the week15:22
*** openstack changes topic to "Priorities for the week (Meeting topic: ironic)"15:22
TheJulia#link https://etherpad.opendev.org/p/IronicWhiteBoard15:22
TheJuliaStarting at line 14115:22
TheJuliaI'll remove the merged items from the list (of which there are a number, one moment.15:23
TheJuliaI'm going to remove the redfish RAID WIP since it has needed rebasing/updating for a few weeks15:24
*** ianychoi has joined #openstack-ironic15:24
TheJuliaAny objections?15:24
Qianbiao<dtantsur> devstack has option to add element15:25
dtantsurcool!15:25
dtantsurTheJulia: none15:25
Qianbiaobut devstack ir really not friendly for hardware bm nodes.15:25
TheJuliaOkay, new items starting at 212.15:25
TheJuliaThe metalsmith changes are around trying to fix a bug/issue in IPA with partition images and UEFI, so I needed a compatible job to test in as well. I think it has identified another issue, but I got derailed on Friday due to the lower-constraints headaches15:26
TheJuliaEveryone good with those items?15:27
derekhIs it ok if I added one to the bug fixes: section ? https://review.opendev.org/#/c/726275/  happy to test etc.. if needed to help it along15:28
patchbotpatch 726275 - x/sushy-oem-idrac - WIP: Accommodate GET HTTP 202 status without Location - 3 patch sets15:28
dtantsurnote that most of us don't have core rights there15:28
dtantsurso it's more for rpioso, cdearborn and ajya|afk15:28
rpiosoderekh: It's a priority for us :-)15:29
derekhdtantsur: ack15:29
derekhrpioso: cool, thanks15:29
* TheJulia adds the items15:30
TheJuliaHow does that look for everyone?15:32
TheJuliaderekh: I guess it is okay to add, but again basically only dell folks have the power to merge it.15:32
derekhTheJulia: gotchya, thanks15:33
TheJuliaohh, the ipe change has changed15:34
dtantsurthe list is impressive :)15:34
TheJuliaWell, it is a week we need reviews performed _AND_ some of those items are already approved15:34
TheJuliaWhat is impressive and difficult to put into words is how awesome our leaps and bounds as a community and a project have been15:35
TheJulia(I did the cycle highlights on Friday)15:35
TheJuliadtantsur: I think I may go add ramdisk tls to that ;)15:35
JayF\o/15:35
TheJuliaNow we just need to get things out the door and released15:35
dtantsurTheJulia: I'd appreciate that ;)15:35
dtantsurwe may also add our intention to deprecate the iscsi deploy to the highlights15:36
dtantsur(I'll try to finish the patch today-tomorrow)15:36
TheJuliafyi, the Forum submission tool has been re-opened15:36
TheJuliaWhich means we likely ought to submit topics asap.15:36
TheJuliaI guess time for discussion15:37
TheJulia#topic Discussion15:37
*** openstack changes topic to "Discussion (Meeting topic: ironic)"15:37
TheJuliaI guess this is more informational.15:37
TheJuliaI held off on getting a kexec meeting scheduled because of the chance of getting some students from Boston University involved. That didn't happen and we got the sad news Friday that they couldn't form up a team to work with us for a few months. But! Ironic is going to be presented to the senior engineering students this week!15:38
TheJulia\o/15:38
TheJuliaarne_wiebalck: trandles: sorry :(15:39
TheJuliaOnward to other things at the moment!15:39
toskyfrom the zuulv3 point of view: please remember to merge the remaining zuulv3 job (it's broken, but less broken that the current one) before next week!15:39
dtantsurlet's put it on the priority list15:40
dtantsurmaybe we can fix it with a backport later on...15:40
TheJuliaThere is the ongoing topic of the raid_config behavior contract and the desire to change it.  We briefly brought this up last week with less than our normal number of attendees last week and the consensus was that we should have it as a Forum discussion topic. Are there any other feelings on this topic? Does everyone that missed last week's meeting concur that the forum is the right place at this point?15:40
TheJuliatosky: do you have the link handy?15:41
toskyTheJulia: https://review.opendev.org/#/c/705030/ - iurygregory is the expert there, I don't remember if the other failures are unrelate15:42
patchbotpatch 705030 - ironic - Native zuulv3 grenade multinode multitenant - 33 patch sets15:42
toskyunrelated*15:42
iurygregoryI will update removing the Depends-On so we can merge since it needs to be max till 20 Sep15:43
* TheJulia takes silence on discussion topics as agreement15:44
TheJuliaI guess Baremetal SIG is nexg15:44
TheJulianext15:44
TheJulia#topic Baremetal SIG15:44
*** openstack changes topic to "Baremetal SIG (Meeting topic: ironic)"15:44
TheJuliaarne_wiebalck: ?15:44
arne_wiebalckTwo little things.15:44
arne_wiebalckThe doodle for the regular slot does only have replies from Ironic cores.15:45
arne_wiebalckI was hoping for some wider interest, shall we resend the mail, or just start?15:45
TheJuliaarne_wiebalck: The mailing list is likely a good idea15:46
iurygregorymaybe resend email and tell it will start on *day time*15:46
arne_wiebalckOk, I will resend.15:46
arne_wiebalck2nd item: Should we ask for a forum slot for the SIG? Or is one at the PTG enough?15:46
TheJuliaarne_wiebalck: I think that would be a good idea15:46
TheJuliaWill you make the submission?15:46
arne_wiebalckAt both, or only one?15:47
TheJuliasorry, the forum segment for the SIG15:47
arne_wiebalckIMO, the forum would be sufficient.15:47
arne_wiebalckright15:47
TheJuliaI think it is the right venue for the SIG to do requirements gathering to feed into the ptg15:47
arne_wiebalckyes, my thinking as well15:47
arne_wiebalck ok, I will submit15:47
arne_wiebalckThat's it from me.15:48
TheJuliak, I'll submit the raid_config contract one after the meeting. I don't think anyone has raised any other topics fo rit15:48
TheJuliaOkay!15:48
TheJulia#topic RFE Review15:48
*** openstack changes topic to "RFE Review (Meeting topic: ironic)"15:48
TheJuliaWe had one item from last week, which got a little further discussion before the meeting, so I'm not sure we need to rehash it during this meeting and can remove it form the agenda15:48
TheJuliaAny obejctions?15:48
dtantsurnot from me15:49
* dtantsur is still thinking about forum topics15:49
TheJulia#topic Open Discussion15:49
dtantsurwell, operator feedback can always work15:49
*** openstack changes topic to "Open Discussion (Meeting topic: ironic)"15:49
arne_wiebalckdtantsur: ++15:49
dtantsurmaybe also solving support matrix around 2 types of images vs 2 boot modes vs software RAID15:50
TheJuliaIt might be a worthwhile forum topic because there is a delineation in between "Ironic as a interface layer for multitenant cloud computing" and "ironic as a general purpose tool in an enterprise"15:50
TheJuliadtantsur: sounds like something to submit!15:50
dtantsurTheJulia: are you going to bulk-submit everything we come up with?15:51
TheJuliaI was thinking individually because there is the internal limit15:52
TheJuliaof 3 submissions per person15:52
TheJulia(unless you kindly ask jimmy and make a good case)15:52
TheJuliaI've already got one for the project update, so I can submit two more15:52
dtantsurI can submit something for image types/boot modes, especially if I find a submission link :)15:53
TheJuliacfp.openstack.org15:53
TheJuliaLink is also on the agenda I believe15:53
dtantsurokie, will do after the meeting15:53
*** zzzeek has quit IRC15:54
TheJuliaarne_wiebalck: would it make sense to combine operator feedback in with the conundrum I mentioned earlier?15:54
arne_wiebalckwhich conundrum is that?15:55
arne_wiebalckah, yes15:55
TheJuliayeah15:55
*** zzzeek has joined #openstack-ironic15:55
rpiosoTheJulia: Regarding the ongoing topic of the raid_config behavior contract and desire to change it ^^^, I and others would like to better understand the basis for the concerns within our project. Could we discuss at both the Forum and PTG?15:56
dtantsurI suspect the PTG is a better place even15:56
rpioso+115:56
TheJuliaPTG is a far better place for the raid behavior contract change proposed15:56
dtantsurnot sure how many non-contributors will understand what we'll be talking about15:56
rpiosoI for one don't :-)15:57
TheJuliaI think operator feedback on "will this be useful" is vital to come into that discussion with15:57
rpiosoAt least the concerns.15:57
TheJuliaarne_wiebalck: are you going to submit operator feedback session or do you want me to?15:57
arne_wiebalckI can do it.15:57
TheJuliak15:57
rpioso+1 re: vote for both15:57
TheJuliaokay then15:58
TheJuliaWell, Unless there is something else, I think we're done for our meeting this week.15:58
TheJuliaThanks everyone!15:59
arne_wiebalckThanks, TheJulia !15:59
rpiosoTheJulia: Thank you!15:59
TheJulia#endmeeting15:59
*** openstack changes topic to "Bare Metal Provisioning | Status: http://bit.ly/ironic-whiteboard | Docs: http://docs.openstack.org/ironic/ | Bugs: https://storyboard.openstack.org/#!/project_group/75 | Contributors are generally present between 6 AM and 12 AM UTC, If we do not answer, please feel free to pose questions to openstack-discuss mailing list."15:59
openstackMeeting ended Mon Sep 14 15:59:33 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:59
openstackMinutes:        http://eavesdrop.openstack.org/meetings/ironic/2020/ironic.2020-09-14-15.00.html15:59
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/ironic/2020/ironic.2020-09-14-15.00.txt15:59
openstackLog:            http://eavesdrop.openstack.org/meetings/ironic/2020/ironic.2020-09-14-15.00.log.html15:59
QianbiaoThanks, TheJulia16:00
Qianbiaolol, I have no idea about most of the discussion  ：）16:00
JayFThat's true of almost all of us Qianbiao, don't feel alone :)16:00
Qianbiao;) o/16:01
JayFWe all have our areas of specialty, except TheJulia and dtantsur who seem to know everything sometimes :D16:01
*** lucasagomes has quit IRC16:01
TheJuliaI know nothing!16:01
dtantsurif only :D16:01
*** zzzeek has quit IRC16:02
TheJuliadtantsur: if only.... we would rule the world!16:03
dtantsur\o/16:03
QianbiaoMy question is coming. someone familiar with infiniband port? I can get IB IF info from inspector now.16:03
TheJuliaI guess that means I need to make a roux for lunch.... so I can roux the day!16:04
Qianbiaobut it is filtered by standard processor.16:04
QianbiaoIs there a method that can add IB IF port to ironic port.16:04
* TheJulia escorts herself out for the bad pun16:04
*** zzzeek has joined #openstack-ironic16:04
dtantsur:D16:04
*** priteau has quit IRC16:04
dtantsurQianbiao: did you figure out why it's ruled out?16:05
Qianbiaoyes16:05
TheJuliaYeah, that is the biggest question I would have, why is it excluded16:05
QianbiaoIn "ValidateInterfacesHook"16:05
iurygregorydtantsur, for bifrost 8.3.1 on bugfix/8.3 the commit I've found is 681f054d07cd21404525330fe15c86f781094aa916:05
dtantsurQianbiao: maybe you need to change [processing]add_ports?16:05
QianbiaoI have set it to all for now16:05
dtantsuriurygregory: looks about right16:06
iurygregoryso I only need to ask in the #openstack-release with this information?16:06
QianbiaoIn "class ValidateInterfacesHook(base.ProcessingHook):", it validate the mac address, with "netutils.is_valid_mac(mac):"16:06
Qianbiaoall ib mac does not match the rule.16:06
Qianbiaoinspector data http://paste.openstack.org/show/797846/16:06
dtantsurQianbiao: the mellanox hardware manager is supposed to provide MACs in the right format..16:07
Qianbiaodtantsur hmmm, there must be something wrong. i found this doc: https://docs.openstack.org/ironic-inspector/latest/user/usage.html#infiniband-support16:09
Qianbiaodoes it means I need to pxe through IB port?16:09
Qianbiaocurrently, we are pxe on eth port.16:09
dtantsurQianbiao: I think this applies to PXE through an IB port16:09
*** zzzeek has quit IRC16:09
dtantsurif you don't need it, I think it's worth trying to understand why the mlnx hardware manager doesn't work the way you expect16:10
*** zzzeek has joined #openstack-ironic16:10
TheJuliaDid the mac filtering logic get changed anytime recently becasue IB macs are longer if memory serves16:10
QianbiaoYes TheJulia, I am pretty sure with this. IB mac is different with normal mac16:11
QianbiaoBut whether inspector transform it to normal MAC regex, i am not sure.16:11
Qianbiaodtantsur do you mean mlnx code in IPA will transform the IB MAC16:12
dtantsurQianbiao: that's how I read it, I'm completely unfamiliar with it however16:12
Qianbiaook I will check source code.16:12
dtantsurTheJulia: Ocata is no longer recently :) but that applies only to PXE via IB, Qianbiao just needs the ports discovered16:13
dtantsurthe mlnx hardware manager seems to do what is needed. the question is why it does not kick in. ramdisk logs may provide a clue.16:13
QianbiaoTheJulia dtantsur yes, we are pxe over eth port. Just want to discovery the IB port, and setup IB partition(vlan) for IB networking.16:15
*** zzzeek has quit IRC16:15
TheJuliahardware manager logs do seem needed :\16:15
QianbiaoTheJulia dtantsur https://github.com/IamFive/gist/tree/master/ib-inspect-ramdisk-log16:17
dtantsur"No Mellanox devices found" ehmmm16:17
Qianbiaooh ....16:17
QianbiaoI see that too..at the moment.16:18
dtantsurI wonder if they have different vendor IDs. or the devices are not detected by the kernel.16:18
*** zzzeek has joined #openstack-ironic16:19
*** stendulker has quit IRC16:19
dtantsurit feels that IPA recognizes only lo and eno216:20
dtantsurQianbiao: my wild guess at this point that the operating system does not recognize or does not properly initialize the IB interfaces16:20
Qianbiaodtantsur hmmm. it seems all interface is ready16:20
QianbiaoI see that in log at somewhere16:20
dtantsurdo you see any mentions of the IB interfaces?16:21
dtantsurnot sure if `ip addr` is supposed to show IB interfaces, but they're not there: https://github.com/IamFive/gist/blob/master/ib-inspect-ramdisk-log/ip_addr16:21
*** derekh has quit IRC16:22
Qianbiao<dtantsur> than why inspect data contains the ib port16:22
TheJuliaSo rebuilds/ephemeral preservation and opt-in/out cleaning.  Are there any other sort of identity crisis sort of issues that anyone can think of?16:22
Qianbiaohttp://paste.openstack.org/show/797846/ line 5316:23
dtantsurQianbiao: mmmmmmmmmmmmm, okay, you got me :)16:23
*** derekh has joined #openstack-ironic16:23
TheJulia(identify crsis is likely a little strong of a statement, but hopefully conveys the conundrums)16:23
Qianbiaodtantsur TheJulia thanks, this issue is pretty closed to fixed. I will check the log again.16:24
dtantsurI don't see the devices in lshw either O_o16:25
*** mkrai has quit IRC16:25
dtantsurI don't know how IPA even learns about them, but at least we know that the mlnx hardware manager does not recognize them for some reason16:25
Qianbiao<dtantsur> .... I copied a wrong ramdisk log16:25
dtantsurheh, good, because I already started thinking I was insane :D16:26
dtantsur(well, less sane than usual)16:26
TheJuliacould the issue be that the hardware is not enabled because of the boot mode?16:26
TheJulialike... maybe it doesn't have uefi firmware or bios firmware?16:26
TheJuliawe've seen that before with some of the early 10GB nics16:26
TheJuliaarne_wiebalck: I was thinking the conundrum should really be a separate session. Thoughts?16:27
Qianbiaodtantsur log updated16:27
arne_wiebalckAh, I understood you wanted to combine these into one.16:27
TheJuliaI'm really unsure16:27
TheJuliaso if anyone has strong feelings either way?16:28
arne_wiebalckI think it fits well with the ops session: hopefully there will be people using it either way.16:28
TheJuliaokay, then mind if I add some words to that submission16:29
arne_wiebalckTheJulia: Or did you want to make this a more general overall direction discussion?16:29
TheJuliaWell, I was thinking general discussion as a larger community would be good16:29
TheJuliaSo maybe we need both?16:29
arne_wiebalckInput for this would mostly come from operators, no?16:30
TheJuliaYeah16:30
arne_wiebalckI mean, everyone will/would have an opinion, but in the end it is what people operating the physical infrastructure would like to see.16:31
arne_wiebalckAnd where the emphasis should go.16:31
TheJulia++16:31
arne_wiebalckAnyway, does not mean I have a strong opinion, happy to take this out again from the ops session.16:32
openstackgerritVerification of a change to openstack/ironic failed: Accept and use a TLS certificate from the agent  https://review.opendev.org/74713616:32
openstackgerritVerification of a change to openstack/ironic-python-agent failed: Generate a TLS certificate and send it to ironic  https://review.opendev.org/74993016:32
dtantsurQianbiao: the journal is truncated for some reason16:32
TheJuliaLets have it in the ops session then16:32
Qianbiao<dtantsur> checking16:33
arne_wiebalckTheJulia: Ok, apologies for the misunderstanding!16:33
TheJuliaarne_wiebalck: made a minor addition16:34
Qianbiaodtantsur not sure why, but that is all in ramdisk log tar.gz16:34
arne_wiebalckTheJulia: feel free to edit, I was merely the submitter :)16:34
dtantsurQianbiao: bad luck :(16:34
TheJuliaI do like that I could edit it... thanks jimmy!16:35
arne_wiebalckThat's indeed handy!16:35
arne_wiebalckWhat are multiple levels of RAID?16:35
dtantsurTheJulia: will many people be upset if I cope-paste the same thing into all 3 fields the submission requires?16:35
* dtantsur hates too many free-form fields16:36
TheJuliadtantsur: not really16:36
arne_wiebalckTheJulia: multiple levels of RAID is RAID on top of RAID?16:36
TheJuliaYeah, basic input into the raid_config contract change16:37
arne_wiebalckAh, ok.16:37
arne_wiebalckBye everyone o/16:38
TheJuliaonboarding session submitted16:39
Qianbiaoarne_wiebalck raid is pretty complex. I code almost 3000 line to finish ibmc raid config logics.16:40
trandlesTheJulia: yay!16:43
TheJuliaraid is one of the most complex and opinionated topics I've ever encountered in this community16:43
JayFI suspect if we ever get in the business of doing freeform partitions/LVM it'll get a run for its money in complexity16:43
TheJuliaWell yes16:43
JayF(obviously KS driver will do this, but is doing it by punting all the complexity to external installers)16:43
* TheJulia remembers the discussion that was had in the very very very cold room in Dublin16:44
* dtantsur only remembers the cold room16:44
dtantsurkidding, I remember it as well. it's an awesome feature, but the required heavy lifting is.. enourmous16:44
JayFIt's nice to have a bridge in the external installer driver zer0c00l is working on, but I don't know if we'll be able to avoid it in Ironic-proper installs forever16:45
Qianbiaodtantsur thanks for your help. the mlnx mac issue, i will figure it out tomorrow. (everything wrong must happen in ipa/mlnx.py)16:48
Qianbiaoneed to go, see you all.16:49
dtantsursee you Qianbiao16:49
*** Qianbiao is now known as Qianbiao|afk16:49
*** Qianbiao|afk has quit IRC16:49
*** derekh has quit IRC17:00
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.3: Fix backup node lookup  https://review.opendev.org/75186417:03
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.3: Update the cache if we don't have a root device hint  https://review.opendev.org/75186517:04
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.2: Update the cache if we don't have a root device hint  https://review.opendev.org/75186617:08
openstackgerritJulia Kreger proposed openstack/ironic-python-agent stable/ussuri: Update the cache if we don't have a root device hint  https://review.opendev.org/75186817:12
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.3: Fix backup node lookup  https://review.opendev.org/75186417:15
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.3: Fix backup node lookup  https://review.opendev.org/75186417:16
openstackgerritJulia Kreger proposed openstack/ironic-python-agent bugfix/6.2: Fix backup node lookup  https://review.opendev.org/75187017:17
openstackgerritJulia Kreger proposed openstack/ironic-python-agent stable/ussuri: Fix backup node lookup  https://review.opendev.org/75187117:17
TheJuliayay for backports17:18
* TheJulia takes a short break for a snack17:24
*** dtantsur is now known as dtantsur|afk17:35
dtantsur|afko/17:35
*** Lucas_Gray has quit IRC17:41
openstackgerritVerification of a change to openstack/ironic-python-agent failed: Remove unused jobs  https://review.opendev.org/74972518:20
*** dsneddon has joined #openstack-ironic18:23
*** mgoddard has quit IRC18:34
*** mgoddard has joined #openstack-ironic18:41
*** jtomasek has joined #openstack-ironic18:55
openstackgerritJulia Kreger proposed openstack/ironic master: Clarify HPE Edgeline support  https://review.opendev.org/75188119:01
*** jtomasek has quit IRC19:08
*** zzzeek has quit IRC19:14
*** zzzeek has joined #openstack-ironic19:16
*** zzzeek has quit IRC19:20
*** zzzeek has joined #openstack-ironic19:23
*** JamesBenson has quit IRC19:24
*** JamesBenson has joined #openstack-ironic19:28
*** zzzeek has quit IRC19:28
*** zzzeek has joined #openstack-ironic19:29
*** zzzeek has quit IRC19:33
*** zzzeek has joined #openstack-ironic19:35
*** uzumaki has quit IRC19:41
*** zzzeek has quit IRC19:54
openstackgerritMerged openstack/bifrost master: Remove unnecessary overrides of ansible_python_interpreter  https://review.opendev.org/74926219:56
*** zzzeek has joined #openstack-ironic19:57
openstackgerritVerification of a change to openstack/ironic failed: Accept and use a TLS certificate from the agent  https://review.opendev.org/74713620:16
openstackgerritBill Dodd proposed openstack/ironic master: WIP: Add Redfish RAID management to Ironic  https://review.opendev.org/73000020:22
*** hjensas is now known as hjensas|afk20:44
openstackgerritBill Dodd proposed openstack/sushy master: Make Actions field in Volume resource optional  https://review.opendev.org/75190021:01
openstackgerritJulia Kreger proposed openstack/ironic master: Minor agent version code cleanup  https://review.opendev.org/74955221:10
openstackgerritBill Dodd proposed openstack/sushy-tools master: Add Volumes nav property to Storage template  https://review.opendev.org/75190321:22
TheJuliaerbarr: hey, are you guys carrying a downstream patch on the devstack plugin? I see jobs failed after the ibmc change was proposed and that is the only thing that made sense.21:29
openstackgerritJulia Kreger proposed openstack/metalsmith master: Replace the Centos7 image job with a Centos8 UEFI based job  https://review.opendev.org/74831521:58
erbarrTheJulia, which patch?22:02
TheJuliahttps://review.opendev.org/#/c/748841/22:06
patchbotpatch 748841 - ironic - Feat: add ibmc hardware info support for devstack - 3 patch sets22:06
erbarrthat looks to be a setuptools failure on the one job i checked, I'll do more digging in a little bit, need to walk my cat around the house22:23
erbarryard*22:23
*** tosky has quit IRC22:33
openstackgerritVerification of a change to openstack/bifrost failed: Always setup environment in bifrost-cli and make it less verbose  https://review.opendev.org/75031522:34
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Add basic retries for inspection  https://review.opendev.org/74559822:38
TheJuliaerbarr: You mean your cat doesn't walk  you?22:41
TheJulia:)22:41
*** tkajinam has joined #openstack-ironic22:52
*** dtruong has quit IRC22:54
*** dtruong has joined #openstack-ironic22:54
erbarrTheJulia, lol, yea he goes where he pleases, there's no fence and some wildlife around and he's elderly in a new place so, just keeping him from getting in trouble22:59
*** portdirect has quit IRC23:13
*** portdirect has joined #openstack-ironic23:13
openstackgerritVerification of a change to openstack/bifrost failed: Force updating repos with copy_from_local_path  https://review.opendev.org/75061723:17
openstackgerritVerification of a change to openstack/bifrost failed: Re-download checkums on upgrades and updates  https://review.opendev.org/75066523:17
jandersgood morning Ironic o/23:34
*** Goneri has quit IRC23:53
« Sunday, 2020-09-13 Index Tuesday, 2020-09-15 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!