Monday, 2021-03-22

*** tosky has quit IRC00:26
openstackgerritMerged zuul/zuul-jobs master: Revert "Pin to npm4 until npm 5.6.0 comes out"
*** stevebaker has quit IRC02:15
openstackgerritMerged opendev/glean master: Reduce metadata read/parsing overhead
openstackgerritMerged opendev/glean master: Run glean fewer times in
openstackgerritMerged zuul/zuul-jobs master: Clone nvm from official source
*** JayF has quit IRC02:23
openstackgerritMerged opendev/glean master: Add Python 3.6 to supported versions
openstackgerritMerged opendev/glean master: Override NetworkManager to wait for udev-settle
*** JayF has joined #opendev02:39
*** hemanth_n has joined #opendev03:27
openstackgerritIan Wienand proposed opendev/glean master: Stop requiring /usr/local/bin links for
*** raukadah is now known as chkumar|ruck05:02
*** ykarel has joined #opendev05:28
*** ykarel has quit IRC05:29
*** ykarel has joined #opendev05:29
*** whoami-rajat_ has joined #opendev05:51
*** marios has joined #opendev06:03
*** ysandeep|away is now known as ysandeep06:17
openstackgerritIan Wienand proposed opendev/glean master: Create "legacy" script path
openstackgerritIan Wienand proposed opendev/glean master: Run a glean-early service to mount configdrive
*** ralonsoh has joined #opendev06:43
*** eolivare has joined #opendev07:33
*** amoralej|off is now known as amoralej07:35
*** amoralej is now known as amoralej|off07:35
*** slaweq has quit IRC07:40
*** ysandeep is now known as ysandeep|lunch07:43
*** slaweq has joined #opendev07:46
*** hashar has joined #opendev07:54
*** fressi has joined #opendev08:05
*** andrewbonney has joined #opendev08:13
*** rpittau|afk is now known as rpittau08:19
*** ysandeep|lunch is now known as ysandeep08:28
*** tosky has joined #opendev08:49
*** jpena|off is now known as jpena08:57
*** amoralej|off is now known as amoralej09:04
openstackgerritMerged opendev/yaml2ical master: Adds second- and fourth- week recurring meetings
*** ykarel is now known as ykael|lunch09:21
*** sboyron has joined #opendev09:28
*** moshiur has joined #opendev09:49
moshiurGood Morning. please review: , I have another PR in ironic-python-agent-builder : which is waiting to get merge and depends on the PR in diskimage-builder.09:50
*** ykael|lunch is now known as ykael10:03
*** dtantsur|afk is now known as dtantsur10:06
dtantsurianw: hi! in did you mean using the same flag or adding a new similar flag? the existing flag does the opposite thing.10:24
*** stephenfin has joined #opendev10:33
*** stephenfin has quit IRC11:04
*** stephenfin has joined #opendev11:05
*** dirk has quit IRC11:18
*** dirk has joined #opendev11:23
*** ykael is now known as ykarel11:25
*** ykarel_ has joined #opendev11:58
*** ykarel has quit IRC12:00
*** whoami-rajat_ is now known as whoami-rajat12:04
*** artom has joined #opendev12:05
*** ykarel_ is now known as ykarel12:09
*** hashar is now known as hasharLunch12:17
openstackgerritMerged openstack/project-config master: Add Backport-Candidate label to Ironic project
openstackgerritMerged openstack/project-config master: Mirror new Magnum and Manila charms to GitHub
*** ykarel_ has joined #opendev12:50
*** jpena is now known as jpena|lunch12:50
*** ykarel has quit IRC12:50
*** hasharLunch is now known as hashar12:55
*** brinzhang has joined #opendev13:05
*** hemanth_n has quit IRC13:20
*** amoralej is now known as amoralej|lunch13:21
*** ykarel_ has quit IRC13:37
*** ykarel_ has joined #opendev13:37
*** lpetrut has joined #opendev13:41
*** jpena|lunch is now known as jpen13:51
*** jpen is now known as jpena13:51
*** ysandeep is now known as ysandeep|afk14:00
*** jhesketh has quit IRC14:02
*** jhesketh has joined #opendev14:04
*** fressi has quit IRC14:13
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: DNM: test role job triggering
*** amoralej|lunch is now known as amoralej14:25
corvusi'd like to restart zuul now  (sorry i didn't get to that on friday)14:26
corvusthis will put some of the new zk changes into production14:27
openstackgerritBrian Rosmaita proposed opendev/irc-meetings master: Set cinder-bs meeting duration
corvusi have run the image pull playbook, so they should be up to date :)14:29
openstackgerritJeremy Stanley proposed zuul/zuul-jobs master: Revert "Temporarily stop running Gentoo base role tests"
*** lbragstad__ is now known as lbragstad14:36
openstackgerritBrian Rosmaita proposed opendev/irc-meetings master: Add Cinder Festival of Reviews meeting
corvus#status log restarted zuul at 92f43d874ae8cc3e39d6455e3c8d9f8d0ca13eb7  (event queues are in zookeeper)14:38
openstackstatuscorvus: finished logging14:38
fungithanks! sorry, was looking at another window or i'd have offered to help14:38
corvusfungi: well, depending on how things go, help may still be needed, but i hope not!  :)14:39
fungiindeed, i'm around14:40
corvusi think the 'zuul enqueue' commands that are running now are using the new code; they seem to be behaving as normal14:40
corvushere's something interesting: since i rolled the zk cluster last week, we ended up with all our connections on 2 zk nodes; now that i've rolled zuul, it's spreading back out to all 3  (nodepool is still only going to be talking to 2)14:41
fungikinda makes sense... zuul is connecting to all three cluster nodes, then you reboot one and the connections it had to that go to one of the other nodes, repeat two more times, you're left with zuul connected to the first two cluster nodes you restarted. now zuul restarts and connects to all three again14:44
fungibasically zuul never had a need to connect to the third cluster node after it was restarted because it was connected to the first two and those remained up14:46
corvusyep.  it may make system load comparisons a little tricky -- right now it actually looks like zk server network traffic and cpu load has decreased -- so the effect of spreading the load is outweighing the additional load of storing the event queues in zk.  good news so far i guess.  :)14:46
corvusbut we're also not done enqueuing yet14:47
fungii suppose we could create a summary graph in grafana with the values for all cluster members added together14:47
corvusyep; though i was looking at cacti for cpu load14:48
fungioh, that got it14:48
*** ykarel_ is now known as ykarel14:49
corvusre-enqueue complete14:52
*** ysandeep|afk is now known as ysandeep15:00
*** whoami-rajat has quit IRC15:00
*** moshiur has quit IRC15:05
*** lpetrut has quit IRC15:18
*** ralonsoh has quit IRC15:21
*** ykarel has quit IRC15:26
*** ralonsoh has joined #opendev15:27
*** radez has joined #opendev15:45
radezclarkb: ping, I'm trying to force my project to use version 2.7.0 of neutron lib in ci, but when I set my constraints ci says it can't find the package. Do you know if I can get 2.7 instead of the latest 2.10?15:47
radezI have a local upper constrains set for neutron-lib===2.7.015:48
*** mlavalle has joined #opendev15:48
*** ysandeep is now known as ysandeep|dinner15:55
fungiradez: have you checked whether that version of neutron-lib is compatible with neutron 17.1.1?15:55
fungithe tox log says the job used pip 21.0.1 which means you're going to have to satisfy the new dependency solver15:58
fungi says neutron-lib>=2.6.0 so that's probably not the cause16:00
fricklertox still uses reqs u-c in addition to your local ones, if one says ==2.7.0 and the other ==2.10.0, that's not solvable16:01
frickleryou likely need to modify the global reqs and drop neutron-lib there16:01
fungithe tox log also says tat pip install command has multiple constraints files applied, yeah: pip install -c/home/zuul/src/ -c/home/zuul/src/ -r/home/zuul/src/ -r/home/zuul/src/
fungii think openstack/requirements/upper-constraints.txt is going to overwrite any conflicting values from x/networking-ansible/local-upper-constraints.txt due to the ordering there16:02
fricklerwith old pip (9.0.1 from bionic), the first constraint wins and installation succeeds. with latest pip (21.0.1) an error occurs16:08
fricklerto reproduce: echo neutron-lib===2.7.0 > x;echo neutron-lib===2.10.0 > y;pip install -cx -cy neutron-lib16:09
fungioh neat, yeah16:10
fungiso it really wants it to be the union of constraints16:10
fungiand the union of 2.7.0 and 2.10.0 is null, so the union of those and >=2.4.0 is also null16:12
radezfungi: would I be better to do this in the reqs file instead of upper constraints?16:12
fungiradez: as frickler suggested, better to only use one constraints file16:12
radezack, lemme see if I can figure it out with out the local upper16:13
fungiif you want different versions than what's in the upper-constraints.txt in openstack/requirements, then don't use that file16:13
fungiyou could periodically sync from openstack's upper-constraints.txt but edit the entries16:13
fungithere is even an edit-constraints utility in the requirements repo meant for programatically altering constraints file entries16:14
fungithough calling it on the fly from tox is nontrivial/messy (generally involves replacing the install_command with a wrapper script)16:14
fungibut you could just merge the relevant contents from the constraints in the requirements repo into your local constraints file and drop the central one from your tox config16:15
radezgot it, thx for the help fungi frickler16:16
*** lpetrut has joined #opendev16:37
openstackgerritDmitry Tantsur proposed opendev/glean master: Do not require external mock on Python 3
*** ysandeep|dinner is now known as ysandeep16:44
*** amoralej is now known as amoralej|off16:46
*** lpetrut has quit IRC16:48
openstackgerritDmitry Tantsur proposed opendev/glean master: Update hacking and fix pep8 violations
*** marios is now known as marios|out16:54
*** marios|out has quit IRC16:58
openstackgerritDmitry Tantsur proposed openstack/diskimage-builder master: simple-init: support installing Glean from packages
*** dtantsur is now known as dtantsur|afk17:09
*** whoami-rajat_ has joined #opendev17:10
*** rpittau is now known as rpittau|afk17:11
*** auristor has quit IRC17:21
*** ysandeep is now known as ysandeep|away17:40
*** eolivare has quit IRC17:41
*** auristor has joined #opendev17:45
*** jpena is now known as jpena|off18:04
*** hashar has quit IRC18:05
*** whoami-rajat_ is now known as whoami-rajat18:06
*** andrewbonney has quit IRC18:09
*** dtantsur has joined #opendev18:34
*** slaweq_ has joined #opendev18:37
*** jpenag has joined #opendev18:39
*** dtantsur|afk has quit IRC18:41
*** jpena|off has quit IRC18:41
*** slaweq has quit IRC18:41
*** sboyron has quit IRC19:25
*** hamalq has joined #opendev19:27
*** ralonsoh has quit IRC19:28
*** whoami-rajat has quit IRC20:19
openstackgerritJames E. Blair proposed opendev/system-config master: Downgrade scheduler to Zuul 4.1.0
corvusfungi: ^20:44
corvuser, we may need to manually deploy that20:48
corvusi'll work on that20:48
fungiokay, back, sorry disappeared for a few20:49
corvusi'm putting zuul01 into emergency disable and will manually apply that change (and try to save queues across the restart)20:49
fungii approved the downgrade, but yeah, there's a chance it won't even be tested20:50
corvusyeah i didn't see it in the status20:50
fungia very good chance, in that case20:51
corvusoh derp, i looked at the wrong tenant (opendev)20:51
corvusit is being tested in the openstack tenant right now20:52
fungiahh, yeah, system-config is still in openstack20:52
corvusso we could just let it run20:52
fungii'm good with the manual restart if you're wanting to be able to push fixes in the zuul tenant20:52
fungier, manual deploy i mean20:53
corvusokay, i'll manual now and we'll let the system catch up20:53
corvus#status log added zuul01 to emergency and restarted scheduler on 4.1.0 due to event queue bugs20:55
openstackstatuscorvus: finished logging20:55
fungii also a few minutes ago rechecked my earlier zuul-jobs change which wasn't being tested under the other zuul version, and it's been running and completing jobs since the restart21:09
ianwo/ lmn if i can help21:27
corvusfungi, ianw: i think we're good and stable; if you want to give everything a quick once over to double check that, that would be helpful; i'll continue working on tests/fixes21:32
*** stevebaker has joined #opendev21:42
*** artom has quit IRC21:46
*** artom has joined #opendev22:00
*** artom has quit IRC22:06
*** slaweq_ has quit IRC22:30
fungicorvus: thinks look generally good, i went through the scheduler debug logs some more and didn't see anything new. also no more of the suspect exception after the downgrade22:37
fungier, things22:37
corvusfungi: cool, i just finished up patches for the errors you observed; i'm trying to see if i can repro avass's now22:38
fungiianw: thanks for sending out a meeting agenda!22:40
openstackgerritMerged opendev/system-config master: Downgrade scheduler to Zuul 4.1.0
openstackgerritMerged opendev/system-config master: kerberos-kdc: quote some integers to avoid string/int confusion
fungiprometheanfire: does this look to you like the configure-mirrors role has bitrotted for gentoo support?
fungithat's the result of trying to reenable gentoo testing for the zuul-jobs repo23:04
prometheanfirefungi: no, that to me sounds like portage didn't sync or something23:10
fungiprometheanfire: that change failed basically ever gentoo job it was trying to add back23:11
prometheanfirethe emerge-webrsync is what syncs the package tree23:12
prometheanfire2021-03-22 21:06:46.499572 | gentoo-17-0-systemd | !!! Invalid Repository Location (not a dir): '/var/db/repos/gentoo'23:14
prometheanfirethat's the root error, maybe there needs to be a mkdir -p first somewhere23:14
prometheanfireok, location is defaulted (not set by DIB), because it's not running on gentoo it's not available by default, I think23:18
fungiso that happened during image creation?23:18
prometheanfireiirc the it's bind-mounted in (/var/db/repos/)23:19
prometheanfirehmm, in root.d I mkdir ${PORTDIR}23:21
prometheanfireI don't see anything wrong with it outright...23:22
prometheanfirebut that's not in the dib part, that's in the ansible part before dib I think23:25
prometheanfirefungi: this may help, not sure....
fungithanks, i'll try wedging that in23:39
*** mlavalle has quit IRC23:57

Generated by 2.17.2 by Marius Gedminas - find it at!