Thursday, 2021-02-25

« Wednesday, 2021-02-24 Index Friday, 2021-02-26 »
*** tosky has quit IRC00:00
*** k_mouza has joined #openstack-ironic00:02
*** k_mouza has quit IRC00:07
TheJuliastevebaker: is network-legacy just doing ye olde init scripts?00:15
TheJuliaand dracut is like... memories long ago00:15
stevebakerTheJulia: I think its init scripts for when the initrd needs to do network things, I don't even know if ours does00:22
stevebakeralso the failure might just be missing package dependencies00:22
TheJuliaI don't think we do00:30
TheJuliathat should be safe..... to remove... I thinik00:30
stevebakerback in a bit, I need to dad-taxi for washed out sports day00:35
*** xinliang has joined #openstack-ironic01:04
stevebakerback01:20
TheJuliawhat is sports?01:37
stevebakersomething weather dependent, apparently01:39
*** ricolin_ has joined #openstack-ironic01:55
*** cp- has quit IRC01:58
*** cp- has joined #openstack-ironic02:04
*** rcernin has quit IRC02:06
*** ricolin_ has quit IRC02:09
*** zzzeek has quit IRC02:23
*** zzzeek has joined #openstack-ironic02:24
stevebakerTheJulia: I found the problem, that was a journey https://review.opendev.org/c/openstack/tripleo-image-elements/+/77749502:34
*** rcernin has joined #openstack-ironic02:38
*** zzzeek has quit IRC02:44
*** cp- has quit IRC02:45
*** zzzeek has joined #openstack-ironic02:46
*** cp- has joined #openstack-ironic02:46
*** zzzeek has quit IRC02:51
*** zzzeek has joined #openstack-ironic02:53
*** uzumaki has joined #openstack-ironic03:10
*** zzzeek has quit IRC03:13
*** zzzeek has joined #openstack-ironic03:15
*** pmannidi_ has joined #openstack-ironic03:36
*** pmannidi has quit IRC03:38
*** SpamapS has quit IRC03:47
*** SpamapS has joined #openstack-ironic03:51
*** mkrai has joined #openstack-ironic03:53
*** mkrai has quit IRC04:17
*** mkrai has joined #openstack-ironic04:18
*** xinliang has quit IRC04:32
*** uzumaki has quit IRC04:49
*** ociuhandu has joined #openstack-ironic05:06
*** ociuhandu has quit IRC05:10
*** tzumainn has quit IRC05:17
*** uzumaki has joined #openstack-ironic05:18
*** gyee has quit IRC05:20
*** uzumaki has quit IRC05:22
*** yoctozepto0 has joined #openstack-ironic06:05
*** yoctozepto has quit IRC06:05
*** yoctozepto0 is now known as yoctozepto06:05
*** rcernin has quit IRC07:22
arne_wiebalckGood morning, ironic!07:43
jandersgood morning arne_wiebalck o/07:44
openstackgerritMerged openstack/ironic-python-agent-builder master: Make sure we can create tinyipa isos  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/77664207:45
arne_wiebalckhey janders o/07:46
*** rcernin has joined #openstack-ironic07:52
*** rpittau|afk is now known as rpittau07:57
rpittaugood morning ironic! o/07:57
jandersgood morning rpittau o/07:57
jandersHappy Thursday! :)07:57
rpittauhey janders :)07:58
rpittauThursday is the new Friday!07:58
* arne_wiebalck stops his intervention08:02
* arne_wiebalck follows "no interventions on Fridays!"08:03
janders\o/08:06
*** mkrai has quit IRC08:06
*** mkrai has joined #openstack-ironic08:10
*** ociuhandu has joined #openstack-ironic08:29
*** tosky has joined #openstack-ironic08:38
*** ociuhandu has quit IRC08:50
*** ociuhandu has joined #openstack-ironic08:52
*** ociuhandu has quit IRC08:52
*** ociuhandu has joined #openstack-ironic08:53
*** ociuhandu has quit IRC08:54
*** ociuhandu has joined #openstack-ironic08:54
*** ociuhandu has quit IRC09:05
*** ociuhandu has joined #openstack-ironic09:19
*** ociuhandu has quit IRC09:25
*** ociuhandu has joined #openstack-ironic09:31
*** ociuhandu has quit IRC09:31
*** k_mouza has joined #openstack-ironic09:31
*** ociuhandu has joined #openstack-ironic09:41
*** derekh has joined #openstack-ironic09:42
*** ociuhandu has quit IRC09:45
openstackgerritRiccardo Pittau proposed openstack/ironic master: [DNM] Debug image links  https://review.opendev.org/c/openstack/ironic/+/77752909:51
openstackgerritRiccardo Pittau proposed openstack/ironic-lib master: [DNM] test unprivileged qemu convert  https://review.opendev.org/c/openstack/ironic-lib/+/77731209:53
*** yoctozepto9 has joined #openstack-ironic09:55
*** jhesketh_ has joined #openstack-ironic09:57
*** dsneddon_ has joined #openstack-ironic09:58
*** dhellmann_ has joined #openstack-ironic09:58
*** hjensas_ has joined #openstack-ironic09:59
*** ociuhandu has joined #openstack-ironic10:00
*** derekh has quit IRC10:03
*** yoctozepto has quit IRC10:03
*** zzzeek has quit IRC10:03
*** dhellmann has quit IRC10:03
*** dsneddon has quit IRC10:03
*** jhesketh has quit IRC10:03
*** hjensas has quit IRC10:03
*** irclogbot_0 has quit IRC10:03
*** mcarden has quit IRC10:03
*** jlvillal has quit IRC10:03
*** yoctozepto9 is now known as yoctozepto10:03
*** dhellmann_ is now known as dhellmann10:03
*** zzzeek has joined #openstack-ironic10:05
*** irclogbot_0 has joined #openstack-ironic10:07
*** ociuhandu has quit IRC10:11
*** k_mouza has quit IRC10:21
openstackgerritRiccardo Pittau proposed openstack/ironic-lib master: Remove default parameter from execute  https://review.opendev.org/c/openstack/ironic-lib/+/77756010:22
*** anuradha1904 has joined #openstack-ironic10:31
*** dougsz has joined #openstack-ironic10:33
*** mkrai has quit IRC10:43
*** hjensas_ is now known as hjensas10:43
*** ociuhandu has joined #openstack-ironic10:50
*** dougsz has quit IRC10:52
*** k_mouza has joined #openstack-ironic10:54
*** dtantsur|afk is now known as dtantsur10:56
dtantsurmorning ironic10:56
*** k_mouza has quit IRC10:58
*** k_mouza has joined #openstack-ironic10:58
jandersgood morning dtantsur10:59
dtantsurlooking forward towards the long weekend?11:00
dtantsurdo you plan on being on the tomorrow's SPUC?11:01
rpittaugood morning dtantsur :)11:01
jandersdtantsur yes! for the long weekend. Will probably skip the SPUC though, might out.11:01
rpittauI won't be at the SPUC tomorrow, probably be out most of the day11:02
dtantsurokay, I won't wake up for it either, I suspect :)11:03
dtantsurmmm, grenade on inspector broken too?11:04
jandershow long would you expect to take to have a node with 8x1TB disks (+ 2x1TB NVMes but that shoudln't be a problem with the NVMe patch) cleaned?11:05
jandersit's been sitting in clean_wait for a good couple hours11:05
dtantsuryou mean, shredding? many hours.11:05
openstackgerritDmitry Tantsur proposed openstack/ironic-inspector master: Enable tempurls for Swift in grenade  https://review.opendev.org/c/openstack/ironic-inspector/+/77757011:06
jandersI had to u erase_devices_priority otherwise it wouldn't run NVMe cleaning and use metadata_clean instead11:06
dtantsurright11:06
jandersprobably it would be heaps better not to have all the magnetic disks for this test11:06
jandersmaybe I will check with Derek if I can disable them on the controller or something11:07
jandersI also wonder how many operators have HDD+NVMe nodes and what would be the desired behavior in such scenarios11:08
jandersfor my test it would be nice to be able to metadata_erase the HDDs but properly clean the NVMes11:08
jandersbut other than test/demo I wonder if people would want that (if it's just the test it's probably not worth it)11:08
arne_wiebalckit may be good to have some logic saying: if it is fast, erase it, otherwise just to metadata11:09
jandersor - should metadata erase switch to NVMe clean if it finds NVMes?11:09
arne_wiebalcklike, if it is an SSD with secure erase: erase it, but if it is a HDD don/t11:09
jandersit probably makes no sense to metadata_erase NVMes with NVMe secure erase support present11:10
openstackgerritDmitry Tantsur proposed openstack/ironic master: devstack: a safeguard for disabled tempurls  https://review.opendev.org/c/openstack/ironic/+/77757211:10
jandersI don't think metadata clean would be any faster; may be the opposite11:10
arne_wiebalckI was more thinking of SSDs and HDDs, not NVMEs11:11
jandersright! I haven't written anything for SSDs yet11:11
arne_wiebalckI would guess that operators may disable the disk erasure11:11
arne_wiebalcksince cleaning can take days11:11
janders(NVMes use nvme-cli; SSDs will need blkdiscard which has much less of a guarantee of data being properly gone)11:11
janderswe were thinking of having a new clean step for blkdiscard11:12
jandersas it sits kind of in between secure erasure and metadata clean11:13
* janders is thinking what is the best way of handling this11:14
* janders realises 2115hrs before a long weekend might not be the best time for making decisions like this :)11:14
arne_wiebalckI added a point to the ops input to the ptg discussion11:15
arne_wiebalckit would be nice to have this handled automatically in a time-efficient manner (rather than skipping erase on all disks just because there are HDDs in the mix)11:15
jandersarne_wiebalck agreed11:16
jandersfast-track clean? secure when time-efficient, sacrifices security in the interest of time if needed11:16
janders(by doing metadata erase on HDDs)11:17
arne_wiebalckyeah, good code name!11:17
openstackgerritDerek Higgins proposed openstack/sushy-tools master: New option to prevent libvirt defining boot order  https://review.opendev.org/c/openstack/sushy-tools/+/77666011:18
openstackgerritDmitry Tantsur proposed openstack/ironic master: Switch iLO and iRMC to the new secure boot framework  https://review.opendev.org/c/openstack/ironic/+/77012211:19
dtantsurfolks, desperately looking for a 2nd +2 on https://review.opendev.org/c/openstack/ironic-lib/+/77435411:20
dtantsur(I'll be out the whole next week and tomorrow, and the release deadline is next Thursday)11:20
*** sshnaidm|afk is now known as sshnaidm|pto11:22
jandersarne_wiebalck do you have a link handy to the Etherpad where you made the comment you mentioned above?11:26
*** ociuhandu has quit IRC11:28
iurygregorygood morning Ironic o/11:35
iurygregoryhappy friday =P11:35
jandersHappy Friday iurygregory o/11:38
iurygregoryo/11:50
openstackgerritRiccardo Pittau proposed openstack/ironic master: [DNM] Debug image links  https://review.opendev.org/c/openstack/ironic/+/77752911:51
*** ociuhandu has joined #openstack-ironic11:58
arne_wiebalckjanders: https://etherpad.opendev.org/p/bare-metal-sig12:06
*** ociuhandu has quit IRC12:06
jandersarne_wiebalck ACK, thank you, makes sense12:07
arne_wiebalckjanders: would be good to how others clean HDD nodes ... JayF maybe?12:10
*** lucasagomes has joined #openstack-ironic12:10
*** bburns has quit IRC12:11
*** bburns has joined #openstack-ironic12:13
jandersarne_wiebalck I don't have first-hand experience with SATA secure erase - how fast is it? Is it seconds, minutes or hours to complete?12:14
jandersdo you know?12:14
arne_wiebalckSSD secure erase?12:15
arne_wiebalckseconds12:15
jandersgeneric SATA secure erase12:15
janders(SSD or SATA HDD)12:15
arne_wiebalckit only deletes the key I think12:15
jandersright!12:15
jandersso in a way NVMe cleaning, blkdiscard and SATA-secureerase fall in one basket (essentially, the non-shred basket)12:16
jandersand shred is the other basket12:16
jandersand then there is the metadata erase which would probably have to be used for HDDs that dont support SATA secure erase12:17
arne_wiebalckyeah12:18
arne_wiebalckdo we need metadata erase when we do secure erase?12:18
arne_wiebalcksorry, any disk erase12:18
jandersarne_wiebalck you mean run both one after another? No, I don't think so.12:24
jandersI will think through this in more detail on Monday, it's getting late, thoughts aren't coming to me12:26
janderswe could have some config option that makes erase_metadata use nvme/sata cleaning if available and metadata_clean everything else12:26
jandersbut maybe it's better to have a clean step that is middle ground between metadata_erase and erase_devices12:27
jandersthat would do "deep" cleaning if hardware assist is available (NVMe or SATA secure erase) or just metadata wipe if not12:27
*** ociuhandu has joined #openstack-ironic12:29
jandersit seems to me that, as per your comment, in the current state of things, if an operator is running a hybrid HDD-NVMe environment NVMe cleaning isn't all that helpful12:29
jandersif it's NVMe only nodes it should work a treat12:29
jandersand I have to admit I always ran like this when NVMes were available so I haven't given much thought to HDD NVMe hybrid scenario12:30
janderstill it bit me today in testing (cause our lab boxes are like that! :) )12:30
jandersdtantsur TheJulia rpittau what do you think?12:31
dtantsura new clean step may be a good idea12:31
janderswe could also plug blkdiscard SSD cleaning into that...12:31
dtantsurquick-and-maybe-secure12:32
janders"best effort"12:32
jandersbut given what we're talking about it should be a reasonably good effort12:32
dtantsuryeah12:32
jandersshall we throw in blkdiscard based mechanism for SSDs?12:32
janders(if they dont support secure erase)?12:32
dtantsuryep, in this context it's appropriate12:32
jandersdo you know from experience if SATA SSDs tend to support SATA secure erase?12:33
jandersno first hand experience, hence the question12:33
dtantsurno clue either12:33
jandersit probably won't hurt to code it up such that it tests for SATA secure erase support12:33
jandersand should it fail, checks if it is an SSD and blkdiscards prior to metadata erase12:33
dtantsuryep. let's create an RFE?12:34
* dtantsur brb, lunch12:34
jandersOK!12:34
jandersthank you dtantsur12:34
jandersI'm wrapping up so have a good break next week dtantsur well deserved! :)12:34
janderssee you next week Ironic o/12:35
iurygregorybye janders enjoy the long weekend12:37
iurygregory:D12:37
*** uzumaki has joined #openstack-ironic12:48
iurygregorydtantsur, since the ManagedBy is an array, I presume I would need to check all Managers available in ManagedBy to check the VirtualMedia, right?12:48
dtantsuriurygregory: correct. that's what ironic does at least.13:00
dtantsurjanders: you too!13:00
iurygregorydtantsur, ack o/13:03
arne_wiebalckdtantsur: for https://review.opendev.org/c/openstack/ironic/+/776930, is what rloo proposes what you had in mind for lazy loading (i.e. make the ports, portgroups, ... an @property of the class and get them from the DB when the property is accessed) ?13:12
*** ociuhandu has quit IRC13:25
*** ociuhandu has joined #openstack-ironic13:26
*** ociuhandu has quit IRC13:30
*** ociuhandu has joined #openstack-ironic13:31
*** ociuhandu has quit IRC13:34
*** ociuhandu has joined #openstack-ironic13:34
*** ociuhandu has quit IRC13:34
*** ociuhandu has joined #openstack-ironic13:35
*** tkajinam has quit IRC13:36
*** ociuhandu has quit IRC13:37
*** ociuhandu has joined #openstack-ironic13:37
*** jamesdenton has joined #openstack-ironic13:40
*** rloo has joined #openstack-ironic13:43
*** iurygregory has quit IRC13:43
*** iurygregory has joined #openstack-ironic13:44
openstackgerritRiccardo Pittau proposed openstack/ironic-python-agent master: Use try_execute from ironic-lib  https://review.opendev.org/c/openstack/ironic-python-agent/+/77759113:46
*** k_mouza has quit IRC13:54
*** k_mouza has joined #openstack-ironic13:55
*** k_mouza has quit IRC14:00
openstackgerritRiccardo Pittau proposed openstack/ironic-lib master: Explicit execute arguments  https://review.opendev.org/c/openstack/ironic-lib/+/77759414:00
*** k_mouza has joined #openstack-ironic14:00
*** yoctozepto has quit IRC14:01
*** yoctozepto has joined #openstack-ironic14:01
rpittauiurygregory: what I found out about privsep is that it takes the privileged thing very seriously, that means that all the commands are effectively run as root, in the case of a file output, that will have user/group also assigned to root14:05
rpittauone thing that we could try is run the qemu commands unpribileged14:05
rpittauunprivileged*14:05
rpittauin a normal run with rootwrap the resulting image belongs to the ironic user14:06
iurygregoryrpittau, ack I will give a try14:08
openstackgerritRiccardo Pittau proposed openstack/ironic-lib master: [DNM] test unprivileged qemu convert  https://review.opendev.org/c/openstack/ironic-lib/+/77731214:13
iurygregoryoh =)14:23
*** uzumaki has quit IRC14:33
openstackgerritYogesh proposed openstack/ironic master: Add idrac HW type IPMI interface support  https://review.opendev.org/c/openstack/ironic/+/77186214:34
dtantsurarne_wiebalck: yep, exactly this14:36
arne_wiebalckdtantsur: ok, thanks, I will see if I can update the patch accordingly, thanks!14:37
dtantsurnp, it shouldn't be very complex14:37
dtantsurbut the impact should be much larger since it will affect most API calls14:37
arne_wiebalckright14:38
dtantsurarne_wiebalck: one note: cache the retrieved objects so that two accessing don't result in two loads14:38
arne_wiebalckdtantsur: how do I do this?14:39
*** uzumaki has joined #openstack-ironic14:41
dtantsurarne_wiebalck: something like http://paste.openstack.org/show/803007/14:42
arne_wiebalckdtantsur: ah, ok, thanks! this is what rloo suggested I think14:43
dtantsurI suspect so14:43
dtantsurI don't think we share tasks between threads, so locking is probably not necessary here14:43
arne_wiebalckok14:44
arne_wiebalckI won't get to this before tomorrow at the earliest14:44
* arne_wiebalck is still adopting nodes into Ironic ...14:44
*** zzzeek has quit IRC14:51
*** tzumainn has joined #openstack-ironic14:52
*** zzzeek has joined #openstack-ironic14:53
*** uzumaki has quit IRC15:05
JayFjanders: arne_wiebalck: all questions about mechanisms of secure erase depend on what drives you buy... I've seen drives that do a DBAN under the hood when you pass an ATA secure erase, I've seen enterprise-quality drives that just swap out a key it was using to transparently encrypt data15:22
JayFjanders: arne_wiebalck: So basically, as with all ironic things, if you care deeply about it works test hardware before purchase and lean on vendors to implement things like ATA secure erase sanely15:23
arne_wiebalckJayF: is that what you do for your hardware?15:23
arne_wiebalckJayF: or do you have nodes where you rely on shred?15:23
JayFI don't wanna speak to any of the security configuration of our cluster.15:24
JayFI will say that at Rackspace, I had issues with a drive, as mentioned above, which did a DBAN on being told to secure erase15:24
JayFand those disks ran out of write cycles in a hurry15:25
arne_wiebalck:)15:25
arne_wiebalck ok, thanks!15:25
JayFthere, for OnMetal, we had basically two types of disks: ones that supported ATA secure erase, and ones that needed a vendor utility15:25
JayFthe  vendor utility was a nonpublic thing our purchasing group got from the vendor in order to meet requirements15:25
arne_wiebalckI see15:26
JayFI think the exceptional job we did cleaning things in OnMetal was guided by that process -- we had the interfaces needed to clean nodes as a first line requirement for hardware we procured for the project15:26
JayFAnd not everyone gets the chance to buy hardware specifically for Ironic, which makes it harder15:26
*** rh-jelabarre has joined #openstack-ironic15:29
*** rh-jelabarre has quit IRC15:29
*** rh-jelabarre has joined #openstack-ironic15:30
arne_wiebalckright15:35
arne_wiebalckthanks JayF !15:36
JayFnot a problem, good luck!15:42
openstackgerritRiccardo Pittau proposed openstack/ironic-python-agent-builder master: Pass rescue user password during user creation  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/77761115:50
openstackgerritRiccardo Pittau proposed openstack/ironic-python-agent-builder master: Pass rescue user password during user creation  https://review.opendev.org/c/openstack/ironic-python-agent-builder/+/77761115:51
*** diurnalist has joined #openstack-ironic16:03
openstackgerritMerged openstack/ironic master: [trivial] fix typos in conductor  https://review.opendev.org/c/openstack/ironic/+/77716216:06
JayFdtantsur: you still working? Curious if you had any objection to me changing https://storyboard.openstack.org/#!/story/2007765 to represent my new no-conductor-to-ipa comms spec16:07
rpittauiurygregory: https://review.opendev.org/c/openstack/ironic-lib/+/777312 I think we're on the right path using the unprivileged qemu convert command16:19
iurygregory\o/16:20
iurygregoryrpittau, ack16:20
iurygregoryI will push in the main patch16:20
* iurygregory brb time for lunch16:23
*** ociuhandu_ has joined #openstack-ironic16:29
*** ociuhandu has quit IRC16:32
openstackgerritJay Faulkner proposed openstack/ironic-specs master: No Conductor to IPA Communication spec  https://review.opendev.org/c/openstack/ironic-specs/+/77717216:32
*** ociuhandu_ has quit IRC16:33
dtantsurJayF: I'm still here, just forgot to turn IRC back on after rebooting :)16:40
dtantsurno objections, go ahead16:41
JayFack thanks!16:41
openstackgerritDmitry Tantsur proposed openstack/ironic master: Make fast-track work with redfish-virtual-media  https://review.opendev.org/c/openstack/ironic/+/77762116:50
openstackgerritMerged openstack/ironic-inspector master: Enable tempurls for Swift in grenade  https://review.opendev.org/c/openstack/ironic-inspector/+/77757016:51
*** lucasagomes has quit IRC17:02
*** bdodd has quit IRC17:04
dtantsuroh ffs, sushy-emulator stores its database in a temporary directory17:04
dtantsurwhy wouldn't it indeed....17:04
*** uzumaki has joined #openstack-ironic17:05
openstackgerritDmitry Tantsur proposed openstack/ironic-inspector stable/victoria: Enable tempurls for Swift in grenade  https://review.opendev.org/c/openstack/ironic-inspector/+/77755117:26
dtantsurfolks, I'll be on PTO tomorrow and next week17:27
dtantsurif somebody reviews https://review.opendev.org/c/openstack/ironic-lib/+/774354 I will address any comments nonetheless17:27
rpittauenjoy dtantsur :)17:28
dtantsurThanks! I'm going to try replicate https://blog.eldruin.com/ccs811-indoor-air-quality-sensor-driver-in-rust/ on my hifive1 board :)17:30
rpittaummm that looks fun :)17:30
dtantsuryeah. I've already got all the bits + the funny traffic lights from https://pi-buch.info/co2-ampel-mit-dem-raspberry-pico-und-circuitpython/17:30
rpittau:D17:31
rpittauI'm going to leave too now, long weekend awaits17:32
rpittausee you on monday! o/17:32
dtantsurany cool plans? :)17:32
rpittauheh fixing stuff at home, building furnitures, hopefully building my retro game joystick at some point!17:32
rpittauI ahve all the pieces, missing the time17:33
dtantsurI see :) good luck!17:33
rpittauthanks!17:34
*** rpittau is now known as rpittau|afk17:34
arne_wiebalckbye everyone o/17:38
dtantsuro/17:39
*** dtantsur is now known as dtantsur|afk17:39
*** bdodd has joined #openstack-ironic17:39
*** diurnalist has quit IRC17:43
*** jlvillal_ has joined #openstack-ironic18:01
*** k_mouza has quit IRC18:05
*** jlvillal_ is now known as jlvillal18:17
*** uzumaki has quit IRC18:35
*** uzumaki has joined #openstack-ironic18:36
*** rloo has quit IRC19:15
*** rloo has joined #openstack-ironic19:15
*** zzzeek has quit IRC19:29
*** zzzeek has joined #openstack-ironic19:29
*** uzumaki has quit IRC19:33
*** uzumaki has joined #openstack-ironic19:36
*** uzumaki has quit IRC19:37
*** stevebaker has quit IRC20:02
*** stevebaker has joined #openstack-ironic20:02
*** adrianc_ has joined #openstack-ironic20:35
*** logan- has quit IRC20:35
*** dtantsur has joined #openstack-ironic20:35
*** lbragstad_ has joined #openstack-ironic20:35
*** beekneemech has joined #openstack-ironic20:35
*** mugsie_ has joined #openstack-ironic20:36
*** persia_ has joined #openstack-ironic20:36
*** benj_- has joined #openstack-ironic20:36
*** early has quit IRC20:36
*** fdegir has quit IRC20:36
*** benj_ has quit IRC20:36
*** zigo has quit IRC20:36
*** trandles has quit IRC20:36
*** persia has quit IRC20:36
*** melwitt has quit IRC20:36
*** gregwork has quit IRC20:36
*** mugsie has quit IRC20:36
*** buhman has quit IRC20:36
*** early has joined #openstack-ironic20:36
*** melwitt has joined #openstack-ironic20:36
*** rh-jlabarre has joined #openstack-ironic20:37
*** benj_- is now known as benj_20:37
*** jmccrory_ has joined #openstack-ironic20:37
*** dhellmann_ has joined #openstack-ironic20:37
*** tosky_ has joined #openstack-ironic20:37
*** rh-jlabarre has quit IRC20:37
*** buhman has joined #openstack-ironic20:37
*** dtantsur has quit IRC20:37
*** dtantsur has joined #openstack-ironic20:37
*** buhman has quit IRC20:37
*** buhman has joined #openstack-ironic20:37
*** rh-jlabarre has joined #openstack-ironic20:37
*** jmccrory has quit IRC20:37
*** jmccrory_ is now known as jmccrory20:38
*** logan- has joined #openstack-ironic20:38
*** strigazi has quit IRC20:38
*** adrianc has quit IRC20:38
*** cp- has quit IRC20:38
*** parallax has quit IRC20:38
*** parallax has joined #openstack-ironic20:40
*** zzzeek has quit IRC20:40
*** rh-jelabarre has quit IRC20:40
*** tosky has quit IRC20:40
*** trident has quit IRC20:40
*** ricolin has quit IRC20:40
*** bnemec has quit IRC20:40
*** bburns has quit IRC20:40
*** dhellmann has quit IRC20:40
*** dhellmann_ is now known as dhellmann20:40
openstackgerritMerged openstack/ironic stable/victoria: Enable swift temporary URLs in grenade and provide a good error message  https://review.opendev.org/c/openstack/ironic/+/77742920:41
openstackgerritMerged openstack/ironic-inspector stable/victoria: Enable tempurls for Swift in grenade  https://review.opendev.org/c/openstack/ironic-inspector/+/77755120:41
*** mnasiadka has quit IRC20:41
*** zzzeek has joined #openstack-ironic20:41
*** tosky_ is now known as tosky20:42
*** bburns has joined #openstack-ironic20:42
*** lifeless_ has joined #openstack-ironic20:42
*** mkowalski has quit IRC20:42
*** mnasiadka has joined #openstack-ironic20:42
*** dtantsur|afk has quit IRC20:42
*** trident has joined #openstack-ironic20:42
*** seba has quit IRC20:43
*** lifeless has quit IRC20:43
*** cp- has joined #openstack-ironic20:43
*** lbragstad has quit IRC20:43
*** bdodd has quit IRC20:43
*** bdodd has joined #openstack-ironic20:46
mnaserTheJulia: i love finding a nicely placed TODO that already mention the issue i ran into :P https://github.com/openstack/ironic/blob/master/ironic/dhcp/neutron.py#L71-L7920:50
mnaserport.get('fixed_ips')[0] = ipv6 address in my case, im doing all my stuff over ipv4, i end up with this in my extra_dhcp_opts: ip_version='6', opt_name='59', opt_value='tftp://1.2.3.4/ipxe.efi'20:51
mnaser(the other ip addres is ipv4)20:51
iurygregoryhehehe20:54
mnaserhttps://github.com/openstack/ironic/blob/8604f84fd7bda4e30d3f07005c4901f3662303a7/ironic/common/pxe_utils.py#L430-L43320:55
mnaseri guess whatever is calling this is calling it with ip_version=620:55
mnaseri don't really have any good choices right now with this i think20:57
mnaseri'd be happy to pick up the work to clean this up but i'm unsure on what would be the best way to increase the chance of success of this20:59
*** mkowalski has joined #openstack-ironic21:04
openstackgerritJay Faulkner proposed openstack/ironic-specs master: No Conductor to IPA Communication spec  https://review.opendev.org/c/openstack/ironic-specs/+/77717221:11
*** rcernin has quit IRC21:12
mnaseri have a clean fix, running unit tests locally21:16
*** k_mouza has joined #openstack-ironic21:25
*** k_mouza has quit IRC21:31
openstackgerritTzu-Mainn Chen proposed openstack/ironic master: Allow instance_info to override node interface  https://review.opendev.org/c/openstack/ironic/+/77743421:42
iurygregorymnaser, feel free to push the patch and we will be happy to take a look =)21:53
mnaseriurygregory: i just finally got the unit tests to pass, will be pushing in a few seconds21:53
iurygregorytomorrow most of the cores will be off (Red Hat recharge day), but we will look at it next week =)21:54
openstackgerritMohammed Naser proposed openstack/ironic master: Add both IPv4 and IPv6 DHCP options if interface has both  https://review.opendev.org/c/openstack/ironic/+/77769121:57
mnaserTheJulia, iurygregory ^ :)21:57
iurygregorymnaser, small thing can you add a releasenote ? =)21:59
mnasersure22:00
iurygregorywould be good to have a story, but releasenote would be more important since we may want to backport =)22:00
mnaseriurygregory: i can try my way with storyboard, i kinda suck at it :P22:00
iurygregorymnaser, you don't need =)22:01
iurygregoryI know you are quite busy at vexxhost =)22:01
mnaserhttps://storyboard.openstack.org/#!/story/200866022:02
iurygregorywow that was fast22:02
*** mkowalski has quit IRC22:05
openstackgerritMohammed Naser proposed openstack/ironic master: Add both IPv4 and IPv6 DHCP options if interface has both  https://review.opendev.org/c/openstack/ironic/+/77769222:05
mnaseriurygregory: added release note and reference to story/task there22:05
iurygregorymnaser, thanks! I've added to my list to look after dinner :D22:06
mnaserthank you, i will cherry pick it here locally22:06
*** mkowalski_ has quit IRC22:06
mnaserand see if it fixes my stuff22:06
iurygregorymnaser, ack, let us know the result22:06
*** mkowalski has joined #openstack-ironic22:07
iurygregorytell guilherme to do it =P22:07
mnaserhaha for once i'm doing the ironic stuff ;)22:07
iurygregorypoor guilhermesp XD22:07
*** rcernin has joined #openstack-ironic22:09
*** rcernin has quit IRC22:15
*** rcernin has joined #openstack-ironic22:15
guilhermespiurygregory: lol  i can always do things but let give mnaser  a chance to have lots of fun :P22:18
iurygregoryguilhermesp, Ironic is always a lot of fun :D22:19
*** lbragstad_ is now known as lbragstad22:26
*** tkajinam has joined #openstack-ironic22:58
JayFfor IPA compatability, I know we ensure new-IPA works on old-Ironic, do we worry about compatability in the other direction?23:07
JayFI guess we have to?23:07
JayFat least IPA(N) has to work on Ironic(N-1)... although I guess we could document that you have to do IPA first23:08
* JayF goes looking, remembering something vaguely about an agent version23:08
JayFUsing this as an example https://opendev.org/openstack/ironic-python-agent/commit/af5f05a0ee6125d9f3c2a541ba164f768b47d2ea it looks like we may not have to do anything on the ironic side for a new feature/config support, cool23:12
JayFif that's wrong someone please tell me (preferably on  https://review.opendev.org/c/openstack/ironic-specs/+/777172 )23:12
openstackgerritJay Faulkner proposed openstack/ironic-specs master: No Conductor to IPA Communication spec  https://review.opendev.org/c/openstack/ironic-specs/+/77717223:14
JayF^ the above spec has gone through a little review, but would appreciate an opinion from someone else as to the direction. Doesn't have to be a deep review, just want to make sure it's headed in the right direction.23:26
*** rloo has quit IRC23:33
*** k_mouza has joined #openstack-ironic23:33
*** ociuhandu has joined #openstack-ironic23:34
*** k_mouza has quit IRC23:34
*** ociuhandu has quit IRC23:38
« Wednesday, 2021-02-24 Index Friday, 2021-02-26 »

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