Friday, 2019-08-30

« Thursday, 2019-08-29 Index Saturday, 2019-08-31 »
*** ijw has quit IRC00:01
openstackgerritJulia Kreger proposed openstack/ironic-tempest-plugin master: Make local boot explicit on software raid tests  https://review.opendev.org/67925800:16
* TheJulia creates some circular patches to see if things go *boom*00:18
TheJuliamaybe circular00:18
openstackgerritJulia Kreger proposed openstack/ironic master: Provide a more useful error message  https://review.opendev.org/67934300:26
*** rloo has quit IRC00:48
openstackgerritJulia Kreger proposed openstack/ironic-tempest-plugin master: Make local boot explicit on software raid tests  https://review.opendev.org/67925800:51
*** ricolin has joined #openstack-ironic02:16
openstackgerritKaifeng Wang proposed openstack/ironic-inspector master: Update non-standalone job to use uwsgi  https://review.opendev.org/67572402:35
*** dsneddon has quit IRC03:44
*** mkrai has joined #openstack-ironic03:52
*** mkrai has quit IRC03:55
*** mkrai has joined #openstack-ironic03:58
*** mkrai has quit IRC04:09
*** mkrai has joined #openstack-ironic04:11
*** dsneddon has joined #openstack-ironic04:18
*** jhesketh has quit IRC04:20
*** mkrai has quit IRC04:22
*** mkrai has joined #openstack-ironic04:23
*** rachit7 has joined #openstack-ironic05:13
*** hamzy has quit IRC05:16
*** e0ne has joined #openstack-ironic05:19
*** e0ne has quit IRC05:20
*** pradip_k has joined #openstack-ironic05:40
openstackgerritRachit Kapadia proposed openstack/ironic master: DRAC: Drives conversion from JBOD to RAID  https://review.opendev.org/67909305:45
*** igordc has quit IRC05:46
*** e0ne has joined #openstack-ironic05:51
*** jtomasek has joined #openstack-ironic05:54
*** jtomasek has quit IRC05:54
*** jtomasek has joined #openstack-ironic05:55
*** dsneddon has quit IRC06:06
*** e0ne has quit IRC06:24
arne_wiebalckGood morning, ironic!06:26
rachit7Good morning arne_wiebalck06:34
arne_wiebalckrachit7: o/06:34
*** dsneddon has joined #openstack-ironic06:38
*** e0ne has joined #openstack-ironic06:39
*** dsneddon has quit IRC06:43
*** rcernin has quit IRC06:48
*** dsneddon has joined #openstack-ironic07:00
*** dsneddon has quit IRC07:05
*** trident has quit IRC07:10
openstackgerritparesh sao proposed openstack/ironic master: Fixed problem with UEFI iSCSI boot for nic adapters  https://review.opendev.org/67473107:12
*** trident has joined #openstack-ironic07:18
*** trident has quit IRC07:22
*** trident has joined #openstack-ironic07:30
*** dsneddon has joined #openstack-ironic07:36
*** jhesketh has joined #openstack-ironic07:38
*** dtantsur|afk is now known as dtantsur07:46
*** pradip_k has quit IRC07:50
openstackgerritNisha Brahmankar proposed openstack/ironic-python-agent master: Adds CI job that uses DIB build images  https://review.opendev.org/67623907:57
*** lucasagomes has joined #openstack-ironic08:06
*** dougsz has joined #openstack-ironic08:07
dtantsurmorning folks08:15
dtantsurcan someone please check the outstanding IPA-builder patches? https://review.opendev.org/#/q/project:openstack/ironic-python-agent-builder+status:open08:15
dtantsurI'd really love to finish this work within the Train timeframe08:16
*** stendulker has joined #openstack-ironic08:24
*** andrein has joined #openstack-ironic08:24
*** stendulker has quit IRC08:32
*** alexmcleod has joined #openstack-ironic08:33
*** derekh has joined #openstack-ironic08:35
*** e0ne has quit IRC08:40
*** e0ne has joined #openstack-ironic08:40
openstackgerritMerged openstack/virtualbmc master: Tiny fix for documentation RST convention  https://review.opendev.org/67918708:41
*** e0ne has quit IRC08:46
openstackgerritIlya Etingof proposed openstack/sushy-tools master: Do not reuse sqlite connection  https://review.opendev.org/67879708:49
*** dsneddon has quit IRC08:57
*** dougsz has quit IRC09:00
kaifengdtantsur: just curious, why the qemu is involved in building tinyipa?09:01
*** e0ne has joined #openstack-ironic09:03
dtantsurkaifeng: we need qemu-img from it09:03
dtantsurit's used by IPA09:03
kaifengdtantsur: oh, I thought we use the tool from distributions09:07
openstackgerritRachit Kapadia proposed openstack/ironic master: DRAC: Drives conversion from JBOD to RAID  https://review.opendev.org/67909309:08
dtantsurkaifeng: you mean, from instance images? no, only grub is used from them (when used).09:09
dtantsurand unfortunately, tinycorelinux has ancient (2.*?) qemu09:10
*** hamzy has joined #openstack-ironic09:10
openstackgerritShuo Liu proposed openstack/ironic master: Add ironic-specs link to readme.rst  https://review.opendev.org/67940309:20
kaifengdtantsur: got it, so we need the qemu-img in the ipa for writing images09:22
dtantsuryeah09:22
kaifengI don't remember how it costs last time I build it (maybe around ~20 minutes), qemu seems to contain quite amount of code.09:23
kaifengI just wonder it seems we maybe only use raw image streaming for the tinyipa, maybe this qemu building doesn't need at all.09:24
dtantsurwe need to cover the conversion code in the CI09:27
dtantsurit's worth investigating if we can strip down the qemu installation though09:28
*** dsneddon has joined #openstack-ironic09:30
kaifengI think we could test the conversion in the dib job, apparently it has more memory available :)09:31
dtantsurif we make it voting - sure09:33
*** dsneddon has quit IRC09:35
kaifengdtantsur: I suspect we never covered the conversion workflow, I didn't find a job has stream_raw_images set to false..09:37
dtantsurkaifeng: even if stream_raw_images is true, we only stream if glance has raw images09:38
dtantsurOR if we cache images on the conductor09:38
dtantsuri.e. if image_download_source=http (not swift)09:39
*** mkrai has quit IRC09:39
*** mkrai has joined #openstack-ironic09:39
kaifengit seems I need to pick up my lost memory, have an impression on the agent+http task, hmm..09:40
*** dsneddon has joined #openstack-ironic09:42
*** dsneddon has quit IRC09:47
*** mkrai has quit IRC09:49
kaifenggoodbye ironicers o/09:50
dtantsurbye kaifeng09:51
arne_wiebalckkaifeng: o/09:51
openstackgerritJan Horstmann proposed openstack/ironic master: Fix typo in handling of exception FailedToGetIPAddressOnPort  https://review.opendev.org/67941409:53
*** e0ne has quit IRC10:01
*** e0ne has joined #openstack-ironic10:10
*** jtwill98 has quit IRC10:16
*** dsneddon has joined #openstack-ironic10:21
*** trident has quit IRC10:23
*** mrda has quit IRC10:23
*** bnemec has quit IRC10:23
*** baha has quit IRC10:23
*** irclogbot_1 has quit IRC10:24
*** dtruong has quit IRC10:24
*** chihhsin has quit IRC10:24
*** SpamapS has quit IRC10:24
*** yoctozepto has quit IRC10:24
*** eandersson has quit IRC10:24
*** lifeless has quit IRC10:24
*** ociuhandu has joined #openstack-ironic10:24
*** trident has joined #openstack-ironic10:24
*** dtruong has joined #openstack-ironic10:27
*** yoctozepto has joined #openstack-ironic10:27
*** chihhsin has joined #openstack-ironic10:27
*** irclogbot_0 has joined #openstack-ironic10:28
*** bnemec has joined #openstack-ironic10:28
*** lifeless has joined #openstack-ironic10:29
openstackgerritMerged openstack/ironic-python-agent-builder master: Run a standalone job in the gate  https://review.opendev.org/67593910:30
*** mkrai has joined #openstack-ironic10:38
*** SpamapS has joined #openstack-ironic10:41
openstackgerritMerged openstack/ironic-python-agent-builder master: Correct the relative path calculation  https://review.opendev.org/67903310:41
*** ociuhandu has quit IRC10:42
*** ociuhandu has joined #openstack-ironic11:02
*** ociuhandu has quit IRC11:14
*** dsneddon has quit IRC11:28
*** dsneddon has joined #openstack-ironic11:28
*** dsneddon has quit IRC11:33
*** ricolin has quit IRC11:35
*** rh-jelabarre has joined #openstack-ironic11:39
*** tesseract has joined #openstack-ironic11:42
*** jawad_axd has joined #openstack-ironic11:48
*** ociuhandu has joined #openstack-ironic11:50
*** dsneddon has joined #openstack-ironic11:53
*** ociuhandu has quit IRC11:57
*** dsneddon has quit IRC12:02
*** yolanda has quit IRC12:12
*** dsneddon has joined #openstack-ironic12:14
*** dsneddon has quit IRC12:20
*** ociuhandu has joined #openstack-ironic12:27
*** derekh has quit IRC12:27
TheJuliaGood morning everyone12:30
*** ociuhandu has quit IRC12:33
*** sthussey has joined #openstack-ironic12:34
*** yolanda has joined #openstack-ironic12:41
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Don't try to put a bootloader in place when bootable  https://review.opendev.org/67933312:50
*** dsneddon has joined #openstack-ironic12:54
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Attempt tgtd session detach  https://review.opendev.org/67933212:54
*** derekh has joined #openstack-ironic12:59
openstackgerritJulia Kreger proposed openstack/ironic master: Add software raid release note to ironic  https://review.opendev.org/67907513:01
*** jawad_axd has quit IRC13:02
TheJuliadtantsur: I think https://review.opendev.org/#/c/679258 is going to need a updated IPA binary posted with the other fix to hopefully let the cirros images bypass trying to install a bootloader. Realistically I can't see us putting 900 MB transfers in place for raid testing and the insistance on grub bootloader on whole disk images is really a horrible bug13:05
patchbotpatch 679258 - ironic-tempest-plugin - Make local boot explicit on software raid tests - 3 patch sets13:05
*** e0ne has quit IRC13:05
*** edleafe has joined #openstack-ironic13:10
*** edleafe has left #openstack-ironic13:13
*** e0ne has joined #openstack-ironic13:14
TheJuliaIf any core reviewers could take a look at https://review.opendev.org/#/c/676889 please13:16
patchbotpatch 676889 - sushy - Change OEM extensions architecture - 4 patch sets13:16
dtantsurTheJulia: *shrugs* metalsmith CI is not less reliable than ironic CI, to be honest13:19
dtantsurand whole disk images themselves don't rely on grup. I guess it's a side effect of booting from md volumes13:20
dtantsuralso morning TheJulia13:20
TheJuliafrom a size, I'm thinking cost in terms of cpu/disk resources/ actual cost for that data to transverse wires. :(13:21
TheJuliasize standpoint13:21
TheJuliaI dunno, just rather make IPA a little bit smarter about its operation13:21
*** mkrai has quit IRC13:22
dtantsurTheJulia: also, why do we need https://review.opendev.org/#/c/679257/ if the CI currently passes?13:22
patchbotpatch 679257 - ironic - Disable software raid CI testing - 1 patch set13:22
dtantsurthe story reads wrong to me, and doesn't justify regression in testing of the key feature13:23
TheJuliadtantsur: the test is actually broken, the default is netboot, which is why the get_boot_option fix addresses, but somewhere has to give13:23
dtantsurTheJulia: the test DOES cover software RAID13:23
* dtantsur puts -2 until this discussion is over13:23
TheJuliaIPA either needs to be smarter for it to not explode, or we need to turn off software raid testing, land fixes to ensure that we can properly handle ipa command install_boot_loader command13:24
TheJuliaright now, your tests don't cause that to be fired, so they are not actually testing it as designed13:24
dtantsurTheJulia: they are testing building software RAID. if you want to remove that, let's revert all software RAID patches.13:25
TheJuliabecause of our existing netboot default and that logic has not been added to state that only whole disk images are built13:25
openstackgerritIlya Etingof proposed openstack/ironic-specs master: Update L3 based deployment spec  https://review.opendev.org/67278013:25
TheJuliayour going to the extreme though, I'm hoping we can fix the test and IPA, I've got all the patches up to actually do that13:25
dtantsurMe? :)13:25
TheJuliaI don't want to remove it, but there is an underlying bug in ironic that we need to fix if we're ever going to get away from the default_boot_option always being netboot13:26
dtantsurSure, let's fix the bug13:26
TheJuliawell, patches are up for that...13:26
dtantsurI don't see a reason to reduce software RAID coverage to 0% now that we've discovered it's "just" 80% instead of 100%13:26
dtantsur(roughly speaking)13:26
TheJuliadtantsur: I was going to immediatley put a revert up, but this again, something needs to move even temporarily to fix it.13:27
dtantsurThis is why I'm asking whether CI is broken by this13:27
dtantsurIf it's not, we don't need a temporary fix. So, is it?13:28
TheJuliait was only temorarily to allow other patches to continue to move forward13:28
TheJuliaand ultimately, the last patch I posted yesterday to change IPA negates the need13:28
dtantsurgreat, let's abandon it and fix a thing at a time13:28
dtantsurreduce the CI coverage to let a few things in sounds a bit too nuclear for non-critical fixes13:29
TheJuliaso I'll rebase the patch after it, it is goin gto fail until IPA patch has merged and is in the images.13:29
openstackgerritRachit Kapadia proposed openstack/ironic master: DRAC: Drives conversion from JBOD to RAID  https://review.opendev.org/67909313:29
openstackgerritJulia Kreger proposed openstack/ironic master: Fix get_boot_option logic  https://review.opendev.org/67913213:29
TheJuliaand done13:29
dtantsurthx!13:30
TheJuliano problem13:30
TheJuliajust maybe add more logic to IPA beyond what I've done in the patch to do additional checks. :)13:30
dtantsursomething inside me is uneasy about https://review.opendev.org/#/c/679333/ but I cannot express it13:30
patchbotpatch 679333 - ironic-python-agent - Don't try to put a bootloader in place when bootable - 2 patch sets13:30
TheJuliaI know what it is, it is only reliant upon the boot flag13:31
dtantsurplease watch out for the metalsmith job failures since it's the only job that tests partitions + local boot13:31
dtantsurI suspect it shouldn't have hit that code path in the previous run. If it does, this approach may be wrong.13:31
TheJuliafinally always gets called though, so it was stupid on my part to not put it before the try13:32
dtantsurTheJulia: something I cannot quite get is why on Earth we're trying grub-install for whole disk images13:32
TheJuliadtantsur: uefi nvram update13:32
dtantsurTheJulia: right, but hitting "return" means that the bootloader wouldn't be installed13:32
dtantsurand the bootloader has to be installed in that jobg13:32
dtantsurah, uefi. happiness.13:32
TheJuliadtantsur: yes...13:32
TheJuliauefi13:33
TheJuliaI think the refactoring ?Nitesh? started on is actually heading in the right-ish direction13:33
dtantsurTheJulia: could you add a debug logging to that patch? something like "Skipping installing bootloader on %s because it's marked as bootable"?13:33
TheJuliabut as a separate command seems... wrong workflow wise :\13:33
dtantsur".. because partition %s is marked as bootable."13:33
TheJulia++13:33
TheJuliasure13:33
dtantsurmm, yeah, the code path was hit for a partition image.13:35
* dtantsur dives into ramdisk logs13:35
dtantsurTheJulia: yep, we end up with a boot partition: http://paste.openstack.org/show/767639/13:36
dtantsurI'm afraid this approach won't fly. Maybe we should skip grub installation only for whole disk images?13:36
TheJuliato do that we would need to make a breaking change and pass another argument along13:38
TheJuliahmm13:38
TheJuliaso we set bootable flag13:38
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Don't try to put a bootloader in place when bootable  https://review.opendev.org/67933313:38
dtantsuryep. or it's somehow on the image13:38
* dtantsur is not sure where these flags are stored13:38
TheJuliaso... in that case, ^^^ + actually looking at the mbr pointer13:38
TheJuliaI can hack on that after meetings today13:38
dtantsurTheJulia: yes, we do https://github.com/openstack/ironic-lib/blob/master/ironic_lib/disk_utils.py#L320-L32513:39
TheJuliadtantsur: flags are in the partition table, the mbr record needs to be decoded to see if it is present and if there is a decoder13:39
TheJuliayeah, I just pulled the same file and saw that, so I need to revise that patch a little more to make it smarter13:39
dtantsurwe may need the check on whether the image is whole disk one13:40
* TheJulia wishes file would return "x86 boot sector" when reading a device13:40
TheJuliawe would need to make lots of changes to do that including passing a new argument into IPA13:40
TheJuliaor the install bootloader command13:40
dtantsurindeed13:40
TheJuliaI think that, and solely trusting it for the smaller and theoretically diminishing case of mbr booting may not be worth that headache.13:41
TheJuliabecause, no matter what, we likely ought to check if things actually look valid enough to boot13:41
dtantsurso what's the suggestion?13:41
TheJuliamake the last patch I just updated smarter13:42
TheJuliaonly the boot loader installation adds the mbr record13:42
TheJuliaso if it is already present, it should be there with the pointer and should at least boot13:42
TheJuliaor throw some awesome error13:42
dtantsuroh, so check if  the bootloader is already installed? might work13:43
TheJuliayeah13:43
dtantsurack, cool13:45
dtantsurthis is even more robust, I guess13:45
dtantsuralthough I still don't fully understand on what code path we hit that in our CI13:45
dtantsurwe don't test RAID + UEFI, do we?13:45
TheJuliadtantsur: no, it is presently mbr only. I can't think of his name has patches up to start working towards enabling that13:45
dtantsurthen isn't the issue that we even try to install bootloader for a whole disk image?13:46
TheJuliawe hit that deep in the prepare_instance code path in the deploy interfaces13:46
dtantsurI think for long time we have been saying that whole disk images default to local boot, and the boot loader has to be installed13:46
dtantsurif we try grub-install, we break windows support at least13:46
TheJuliaTechnically yes, except we need to ensure the nvram is up to date because some vendors decided not to look for defaults with their hardware13:46
dtantsurnvram is not needed without uefi, no?13:47
TheJuliayeah, we likely should do some exaustive checking13:47
dtantsurand does it mean that we don't support windows with uefi?13:47
TheJuliapresent, not required by spec, required by some vendors13:47
TheJuliadtantsur: technically right now we wouldn't be able to with the way the code is written13:47
dtantsurugh13:47
TheJulianow, if we make it smarter, then yes, we should be able to just fall through and windows should deploy13:48
* dtantsur wants a CI job with windows13:48
TheJuliathis is where I was starting to think about ?Nitesh?'s uefibootmgr patches is heading in the right direction13:48
TheJuliame too13:48
*** EmilienM is now known as EvilienM13:49
*** Goneri has joined #openstack-ironic13:49
dtantsurand, back to the "deprecate partition images" discussion, we need to reduce the supported matrix13:50
TheJulia++13:53
dtantsurTheJulia: are you fine with me doing the actual deprecation instead of (or rather, replacing) https://review.opendev.org/#/c/677558/ ?13:53
patchbotpatch 677558 - ironic-lib - Add todo note for is_block_device - 2 patch sets13:53
TheJuliadtantsur: go right ahead13:54
dtantsurokay, will get some tea and jump on it13:54
TheJuliadtantsur: I have administrative things to do later today, and I'm likely not going to have time to next week.13:54
*** dabadaba has quit IRC13:58
*** dabadaba has joined #openstack-ironic13:59
*** dsneddon has quit IRC14:00
TheJuliayay, looks like no first meeting \o/14:00
*** ociuhandu has joined #openstack-ironic14:01
*** ociuhandu has quit IRC14:02
openstackgerritMerged openstack/sushy-tools master: Do not reuse sqlite connection  https://review.opendev.org/67879714:07
TheJuliawell, found an easy way14:12
*** dougsz has joined #openstack-ironic14:15
*** dougsz has quit IRC14:17
openstackgerritIlya Etingof proposed openstack/sushy-tools master: Fix non-functional driver configuration on Py2  https://review.opendev.org/67945614:17
etingofif rpioso could pass this ^ to jtwill98, that might be helpful..14:18
*** baha has joined #openstack-ironic14:18
*** dsneddon has joined #openstack-ironic14:18
*** rachit7 has quit IRC14:20
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Don't try to put a bootloader in place when bootable  https://review.opendev.org/67933314:21
TheJuliadtantsur: ^^^14:21
*** dsneddon has quit IRC14:23
dtantsurTheJulia: makes sense, let's see what the CI says14:24
*** dsneddon has joined #openstack-ironic14:24
TheJuliayup, that will be the true test14:26
*** dsneddon has quit IRC14:29
dtantsurugh, iscsi_verify_attempts is also used by ironic14:29
*** dsneddon has joined #openstack-ironic14:30
*** ociuhandu has joined #openstack-ironic14:34
*** e0ne has quit IRC14:35
*** mkrai has joined #openstack-ironic14:39
*** ociuhandu has quit IRC14:39
*** e0ne has joined #openstack-ironic14:39
rpiosoetingof: Done14:40
rpiosoetingof: Would you like his contact info?14:41
etingofrpioso, I just recalled I have his e-mail!14:42
etingofrpioso does not like to IRC proxy it seems ;)14:42
* rpioso doesn't mind, but is lossy14:43
rpiosoetingof: I suggested he hangout with us here :-)14:44
etingofdid we share him off?14:44
etingof*scare14:44
rpiosoTexans aren't so easily scared.14:45
etingofsure, Chuck Norris can divide by zero14:47
*** trandles has joined #openstack-ironic14:47
TheJuliawell, I thought one could now divide by zero in theoretical terms14:48
TheJuliao/ trandles14:48
TheJuliatrandles: something occured to me last night, you said you were trying to use the iscsi deploy interface, except if memory serves you don't have local storage so I think you'd want the ramdisk interface14:49
openstackgerritDmitry Tantsur proposed openstack/ironic master: Prepare for deprecation of iscsi_verify_attempts in ironic-lib  https://review.opendev.org/67946314:49
dtantsurTheJulia: step one ^^^14:49
TheJuliahopefully only a two step process... and not like sw raid :(14:49
dtantsurshould be only two :)14:52
dtantsuryeah, RAID is a fun task.. but a great achievement as well14:52
TheJuliayeah14:52
TheJuliait is an awesome gain for ironic14:53
openstackgerritDmitry Tantsur proposed openstack/ironic-lib master: Deprecate [disk_utils]iscsi_verify_attempts  https://review.opendev.org/67755814:56
dtantsurTheJulia: step 2 ^^14:56
TheJuliachanging default boot_option would also be epic for this cycle14:57
* TheJulia hopes14:57
dtantsurEPIC14:58
TheJuliaCI failure rates should also go down... at least some14:59
dtantsurdo we have a reason to not deprecate netboot? BFV?15:00
*** ijw has joined #openstack-ironic15:05
rpiosoetingof: lol15:06
*** ociuhandu has joined #openstack-ironic15:07
*** ijw has quit IRC15:11
*** ijw has joined #openstack-ironic15:11
*** ijw has quit IRC15:13
*** ijw has joined #openstack-ironic15:14
*** eandersson has joined #openstack-ironic15:30
*** gyee has joined #openstack-ironic15:33
*** mkowalski has joined #openstack-ironic15:43
*** mkowalski has quit IRC15:46
NobodyCamGood Morning Ironic'er and ...15:47
NobodyCamofc...15:47
NobodyCamTGIF!15:47
*** alexmcleod has quit IRC15:50
*** tesseract has quit IRC15:52
*** jtwill98 has joined #openstack-ironic15:57
dtantsur\o/15:58
dtantsurmorning NobodyCam15:58
*** lucasagomes has quit IRC15:59
NobodyCamMorning dtantsur :)16:03
*** e0ne has quit IRC16:04
*** e0ne has joined #openstack-ironic16:09
TheJuliadtantsur: bfv and ramdisk interface..16:12
dtantsurTheJulia: I'm wondering if we should leave netboot as a special case for these two16:12
dtantsur.. which probably won't reduce the amount of code too much :(16:12
*** dsneddon has quit IRC16:14
TheJuliaNot really, partition will ease a bit if there is nobody screaming for need16:14
*** ijw_ has joined #openstack-ironic16:15
*** ijw_ has quit IRC16:15
*** ijw_ has joined #openstack-ironic16:16
*** ijw_ has quit IRC16:17
*** dsneddon has joined #openstack-ironic16:17
*** ijw_ has joined #openstack-ironic16:17
*** ijw has quit IRC16:18
openstackgerritJulia Kreger proposed openstack/ironic-python-agent-builder master: Add coreutils to tinybuild's IPA image  https://review.opendev.org/67947816:29
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Don't try to put a bootloader in place when bootable  https://review.opendev.org/67933316:30
*** e0ne has quit IRC16:30
dtantsurhave a great weekend16:36
*** dtantsur is now known as dtantsur|afk16:36
*** igordc has joined #openstack-ironic16:39
*** dsneddon has quit IRC16:40
*** rachit7 has joined #openstack-ironic16:46
arne_wiebalckHave a good week-end everyone o/16:46
rachit7(y)16:46
*** rachit7 has quit IRC16:51
TheJuliao/16:53
*** bnemec is now known as beekneemech16:53
*** derekh has quit IRC17:00
*** andrein has quit IRC17:01
*** mkrai has quit IRC17:08
*** dsneddon has joined #openstack-ironic17:09
openstackgerritMerged openstack/ironic-python-agent-builder master: tinyipa: update qemu and tgt dependencies  https://review.opendev.org/67841517:15
*** ricolin has joined #openstack-ironic17:28
*** rmcall has quit IRC17:36
*** andrein has joined #openstack-ironic17:45
*** ociuhandu has quit IRC17:51
*** ociuhandu has joined #openstack-ironic17:57
*** ricolin has quit IRC18:00
*** ociuhandu has quit IRC18:02
openstackgerritJulia Kreger proposed openstack/ironic-python-agent-builder master: Add file to tinybuild's IPA image  https://review.opendev.org/67947818:02
*** jtwill98 has quit IRC18:04
*** jtomasek has quit IRC18:07
*** jtomasek has joined #openstack-ironic18:09
openstackgerritJulia Kreger proposed openstack/ironic-python-agent master: Don't try to put a bootloader in place when bootable  https://review.opendev.org/67933318:17
*** dsneddon has quit IRC18:18
trandlesTheJulia, yes that's correct about iscsi deploy but right now I'm configuring ironic to provision my cloud servers. Then I'll move on to the diskless HPC compute nodes.18:33
trandlesIronic: One Service To Provision Them All18:33
TheJuliatrandles: ahh!18:45
TheJuliatrandles: Excellent18:45
* TheJulia ponders stickers....18:45
*** dsneddon has joined #openstack-ironic18:50
*** dsneddon has quit IRC18:55
*** andrein has quit IRC18:58
openstackgerritDoug Hellmann proposed openstack/ironic master: add table of available cleaning steps to documentation  https://review.opendev.org/67949919:25
dhellmannTheJulia, JayF: ^ based on our conversation earlier this week19:26
*** dsneddon has joined #openstack-ironic19:29
*** dsneddon has quit IRC19:35
*** mbeierl has quit IRC19:37
*** andrein has joined #openstack-ironic19:41
*** Goneri has quit IRC19:45
*** andrein has quit IRC19:55
*** andrein has joined #openstack-ironic19:56
donnyddoes anyone know if cern is using magnum and ironic together?20:05
donnydwow i should have googled that before i asked20:06
donnydLOL20:06
donnydAlso has anyone gotten idrac and serial console to work20:06
*** dsneddon has joined #openstack-ironic20:09
*** dsneddon has quit IRC20:14
*** Goneri has joined #openstack-ironic20:23
*** ociuhandu has joined #openstack-ironic20:25
*** ociuhandu has quit IRC20:30
openstackgerritJulia Kreger proposed openstack/ironic-python-agent-builder master: Add file to tinybuild's IPA image  https://review.opendev.org/67947820:34
TheJuliadonnyd: i don't think the idrac hardware type has the ability to load an ipmi socat console20:35
donnydit has a serial over lan console20:35
TheJuliaI mean their hardware type, give me a moment20:37
donnydwas hoping someone had done it before with a dell server20:37
donnydkk20:37
TheJuliahttps://opendev.org/openstack/ironic/src/branch/master/ironic/drivers/drac.py#L33 doesn't list a console interface, it is just not a part of the driver in ironic, but if you use ipmi, you should be able to use the ipmi hardware type/driver20:40
*** dsneddon has joined #openstack-ironic20:47
*** rmcall has joined #openstack-ironic20:50
*** dsneddon has quit IRC20:52
donnydThanks TheJulia20:59
donnydIs there someone using not the ipmi driver?20:59
donnydI use ipmi for everything20:59
donnydmainly because I have a mishmash of HW vendors in the rack21:00
*** dsneddon has joined #openstack-ironic21:19
*** dsneddon has quit IRC21:24
*** hwoarang has quit IRC21:25
*** hwoarang has joined #openstack-ironic21:28
*** ijw_ has quit IRC21:50
*** dsneddon has joined #openstack-ironic21:52
*** trandles has left #openstack-ironic21:55
*** dsneddon has quit IRC21:57
TheJuliadonnyd: oh, well, that is a decent reason to. So what sort errors are you seeing? I take it you have the console_interface set in ironic?22:07
*** Goneri has quit IRC22:08
donnydInterestingly enough I get no errors, on the ironic side the console starts and socat is listening22:09
donnydbut when i pull up the console in horizon I just get a blank screen. And yes I did configure nova-compute and nova-serialproxy22:09
donnydserialproxy is listening, but nothing seems to happen22:10
donnydmaybe better to debug another day22:10
donnydHope everyone has a great weekend :)22:10
*** Goneri has joined #openstack-ironic22:13
TheJuliadonnyd: if you telnet to the port setup for the socat console, do you end up connected to the the console?22:15
donnydOo that is a good idea22:16
donnydI will give that a whirl22:16
TheJuliaokY22:16
TheJuliaerr, okay22:16
donnydokY is pretty good too22:17
*** jtomasek has quit IRC22:19
*** dsneddon has joined #openstack-ironic22:25
*** dsneddon has quit IRC22:31
openstackgerritJulia Kreger proposed openstack/ironic-python-agent-builder master: Add file to tinybuild's IPA image  https://review.opendev.org/67947822:43
*** vesper has quit IRC22:46
*** vesper11 has joined #openstack-ironic22:47
*** rh-jelabarre has quit IRC22:53
*** dsneddon has joined #openstack-ironic22:56
*** yaawang has quit IRC23:00
*** yaawang has joined #openstack-ironic23:01
*** dsneddon has quit IRC23:01
*** jhesketh has quit IRC23:18
*** dsneddon has joined #openstack-ironic23:31
*** dsneddon has quit IRC23:35
*** gyee has quit IRC23:37
*** Goneri has quit IRC23:50
« Thursday, 2019-08-29 Index Saturday, 2019-08-31 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!