Tuesday, 2019-06-18

« Monday, 2019-06-17 Index Wednesday, 2019-06-19 »
*** gthiemon1e has joined #openstack-lbaas00:06
lxkonghi, i am running devstack now but it always fails when building amphora image with the error `/tmp/dib_build.Pl90ccgm/hooks/extra-data.d/98-source-repositories: line 257: /home/lingxiankong/.cache/image-create/source-repositories/repositories_flock: Permission denied`00:07
*** gthiemonge has quit IRC00:07
johnsomYou probably ran it once as root and are now running it as a user00:08
johnsomYou can remove the whole .cache directory if you want, it will rebuild it.00:08
lxkonghmm...yes, you are righ00:08
lxkongright00:08
johnsomThat is a diskimage-builder directory and file00:08
lxkongjohnsom: btw, did you see the logs i sent yesterday?00:09
johnsomlxkong Yes! Thank you, I have been trying things based on what I saw all day long.00:10
lxkongjohnsom: cool, no worries, just double check00:10
johnsomCan't say I have found the solution yet, but it was super helpful00:10
lxkongglad to hear it's helpful to you00:10
*** mithilarun has quit IRC00:25
*** lemko has quit IRC00:35
rm_workugh, running into issues with our noop tests because the new network-delta calculation does some actual logic about what networks/subnets are plugged, so the configuration has to match something sane00:41
rm_workbut the noop drivers just can't be that intelligent00:41
rm_worki've tried to make them *much more* intelligent00:41
rm_workbut there's a cap to that I think00:41
rm_workI just can't see a way to get them to be smart enough to return data that's valid enough to not break the calculate flow00:42
*** yamamoto has joined #openstack-lbaas00:45
johnsomHmm, is it because the dict it's storing is to much info?00:46
johnsomPSA: The Linux kernel SACK issues public today do impact the amphora images, so you all may want to build new images once your Amphora OS distribution of choice releases the patched kernels.00:52
johnsomhttps://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2019-1147700:53
*** sapd1_x has quit IRC00:55
*** ricolin has joined #openstack-lbaas01:05
rm_workwoo more hax?01:14
johnsomhttps://images.app.goo.gl/Wcv3jmkAyrkXzHna801:16
rm_workwhat is the .io page?01:18
rm_work:D01:19
rm_workif there's no clever name and .io website, it's not a real vulnerability ^_^01:19
johnsomIt is "SACK" I guess01:21
johnsomhttps://thumbs.gfycat.com/PalatableFlawedCockatiel-size_restricted.gif01:22
*** hongbin has joined #openstack-lbaas01:32
johnsomAnd the neutron cleanup: https://review.opendev.org/66583801:38
*** yamamoto has quit IRC01:42
*** yamamoto has joined #openstack-lbaas01:42
*** goldyfruit has quit IRC01:57
rm_workugh, pyroute2 broke something01:58
rm_workfiguring out which commit it was now01:58
*** bzhao__ has joined #openstack-lbaas02:03
rm_workbelgh02:05
rm_workhttps://github.com/svinota/pyroute2/commit/b69999ec82667828816fa2a7b19d0a4f2fe06a8302:05
rm_workbreaks OSX testing again02:05
*** ramishra has joined #openstack-lbaas03:06
*** hongbin has quit IRC03:47
*** psachin has joined #openstack-lbaas03:50
johnsomAll of this gate instability is super annoying: "'Failed to run /opt/stack/new/grenade/projects/70_cinder/resources.sh create '"04:05
johnsomI don't care about cinder!04:05
*** threestrands has joined #openstack-lbaas04:05
johnsomIt's a legacy job or I would go remove cinder from the picture. It's harder to do on the legacy jobs.04:06
johnsom /opt/stack/new/grenade/projects/70_cinder/resources.sh:create:198 :   die 198 'SSH to the client did not work, something very wrong'04:08
johnsomTruth04:08
openstackgerritAdam Harwell proposed openstack/octavia master: Allow multiple VIPs per LB  https://review.opendev.org/66023904:10
openstackgerritAdam Harwell proposed openstack/octavia master: WIP: fix plugging member subnets on existing networks  https://review.opendev.org/66540204:10
*** yboaron_ has joined #openstack-lbaas04:21
*** yboaron_ has quit IRC04:25
*** pcaruana has joined #openstack-lbaas04:56
openstackgerritMichael Johnson proposed openstack/octavia master: Specify the linux-image-kvm kernel for ubuntu  https://review.opendev.org/66586105:01
*** rtjure has joined #openstack-lbaas05:24
*** ivve has quit IRC05:29
*** AlexStaf has joined #openstack-lbaas05:32
*** NobodyCam has quit IRC05:35
*** johnsom has quit IRC05:36
*** NobodyCam has joined #openstack-lbaas05:36
*** rpittau|afk has quit IRC05:36
*** hogepodge has quit IRC05:36
*** rpittau|afk has joined #openstack-lbaas05:37
*** hogepodge has joined #openstack-lbaas05:38
*** johnsom has joined #openstack-lbaas05:38
*** vishalmanchanda has joined #openstack-lbaas06:21
*** rpittau|afk is now known as rpittau06:28
*** ivve has joined #openstack-lbaas06:40
*** gcheresh has joined #openstack-lbaas06:47
*** lemko has joined #openstack-lbaas06:49
lemkoHi. I'm getting many amphora in error state (openstack loadbalancer amphora list), what's the best way to delete them?06:52
*** ccamposr has joined #openstack-lbaas06:54
*** trident has quit IRC06:57
*** trident has joined #openstack-lbaas06:59
cgoncalveslemko, users cannot delete amphorae. I'd suggest you to check what went wrong on your amphorae, fix the problem, and then trigger amphora failover ($ openstack loadbalancer amphora failover $AMP_ID)06:59
lemkoI will try, thanks07:00
openstackgerritAdam Harwell proposed openstack/octavia master: WIP: fix plugging member subnets on existing networks  https://review.opendev.org/66540207:07
*** luksky has joined #openstack-lbaas07:08
*** yetiszaf has quit IRC07:09
*** tesseract has joined #openstack-lbaas07:24
*** threestrands has quit IRC07:36
openstackgerritMerged openstack/octavia master: Fix a python3 issue in the amphora-agent  https://review.opendev.org/66549807:42
nmagnezi^^ took something around 700 rechecks.. but at least we have it merged now \o/07:43
*** h3rbz has quit IRC08:20
rm_workeugh yeah the amount of rechecks recently has been RIDICULOUS08:22
rm_worki'm about to start setting a lot more of the test suites as non-voting, or else just adding @skiptest to the specific tests that aren't working reliably08:23
rm_workand it's really depressing that the cent7 test is STILL flaky, though i think i saw that you had a patch up for that now? almost fixed?08:24
rm_workwe should not be having gate reliability problems still at this point >_<08:24
*** psachin has quit IRC08:24
*** gcheresh has quit IRC08:29
*** gcheresh has joined #openstack-lbaas08:30
*** ltomasbo has joined #openstack-lbaas08:33
*** ltomasbo has left #openstack-lbaas08:33
openstackgerritAdam Harwell proposed openstack/octavia master: Workaround pyroute2 platform-dependent imports  https://review.opendev.org/66591508:48
openstackgerritMerged openstack/octavia stable/stein: Add Stein octavia-v2-dsvm-scenario-ubuntu-xenial  https://review.opendev.org/65268709:01
openstackgerritAdam Harwell proposed openstack/octavia-tempest-plugin master: Add amphora update service client and API test  https://review.opendev.org/63329509:05
rm_workahh just rebased that but now i see why it's in merge conflict -- it's at the end of a chain lol09:06
*** yamamoto has quit IRC09:23
*** ataraday has joined #openstack-lbaas09:24
*** yamamoto has joined #openstack-lbaas09:24
*** yamamoto has quit IRC09:29
openstackgerritAdam Harwell proposed openstack/octavia-lib master: Additional VIPs is also relevant in provider_base  https://review.opendev.org/66451009:31
openstackgerritAdam Harwell proposed openstack/octavia-tempest-plugin master: WIP: Test Additional VIPs  https://review.opendev.org/66446209:32
*** yboaron_ has joined #openstack-lbaas09:43
openstackgerritAdam Harwell proposed openstack/octavia-tempest-plugin master: Add octavia-v2-dsvm-scenario-ipv6 to check queue  https://review.opendev.org/59407809:56
*** rcernin has quit IRC09:57
*** yamamoto has joined #openstack-lbaas10:41
*** psachin has joined #openstack-lbaas10:49
*** yamamoto has quit IRC10:50
*** yamamoto has joined #openstack-lbaas10:59
openstackgerritAdam Harwell proposed openstack/octavia master: WIP: Allow IPv6 health network in devstack  https://review.opendev.org/66510311:02
*** gcheresh has quit IRC11:09
rm_workaugh seriously gates wtf11:27
*** luksky has quit IRC11:30
openstackgerritAnn Taraday proposed openstack/octavia master: Use retry for AmphoraComputeConnectivityWait  https://review.opendev.org/66279111:41
*** yamamoto has quit IRC11:48
*** yamamoto has joined #openstack-lbaas11:49
*** boden has joined #openstack-lbaas11:51
*** yamamoto has quit IRC11:53
*** luksky has joined #openstack-lbaas12:02
*** goldyfruit has joined #openstack-lbaas12:15
rm_workataraday: you here?12:31
ataradayrm_work, hi12:31
rm_worksomehow I just noticed, those are *single task* flows12:32
rm_workis there a reason they even need to be a subflow?12:32
rm_workdoes the "retry flow" thing start the entire flow over and not a single task?12:32
ataradayrm_work, retry can be applyed only on flow, not a task12:33
rm_workerr12:33
ataradayso I have a to intoduce separate flow to retry single task12:33
rm_workhmmm ok12:33
rm_workI think I get it12:33
rm_workoh yeah, nm derp12:34
rm_workI see it, k looks good then :D12:34
*** yamamoto has joined #openstack-lbaas12:35
ataradayrm_work, https://wiki.openstack.org/wiki/TaskFlow/Retry12:36
rm_workyep, +2 :)12:36
ataradaythese are some examples12:36
ataradaythanks!12:37
rm_workI just missed where the actual retry was being inserted, kept only seeing the inject thing12:37
rm_worki blame it on 5:30am12:37
rm_worki'm calling it a night, catch ya'll in the ... evening12:37
*** goldyfruit has quit IRC12:38
*** goldyfruit has joined #openstack-lbaas13:00
openstackgerritAnn Taraday proposed openstack/octavia master: [WIP] Transition l7policy flows to dicts  https://review.opendev.org/66597713:16
*** ataraday has quit IRC13:18
*** mkuf has joined #openstack-lbaas13:29
*** pcaruana has quit IRC13:37
*** yamamoto has quit IRC13:41
*** bonguardo has joined #openstack-lbaas13:41
*** yamamoto has joined #openstack-lbaas13:42
*** yamamoto has quit IRC13:42
*** yamamoto has joined #openstack-lbaas13:43
*** psachin has quit IRC13:46
*** vishalmanchanda has quit IRC13:48
*** pcaruana has joined #openstack-lbaas13:54
*** ianychoi_ is now known as ianychoi14:26
*** Vorrtex has joined #openstack-lbaas14:32
*** bonguardo has quit IRC14:35
*** fnaval has joined #openstack-lbaas14:37
*** ccamposr__ has joined #openstack-lbaas14:50
*** ccamposr has quit IRC14:53
*** ivve has quit IRC15:33
*** yboaron_ has quit IRC15:38
*** luksky has quit IRC15:39
*** lemko has quit IRC15:39
*** NobodyCam has quit IRC15:39
*** kklimonda has quit IRC15:39
*** lemko has joined #openstack-lbaas15:40
*** NobodyCam has joined #openstack-lbaas15:40
*** kklimonda has joined #openstack-lbaas15:41
*** goldyfruit has quit IRC15:43
*** goldyfruit has joined #openstack-lbaas15:55
*** pcaruana has quit IRC16:01
*** fnaval has quit IRC16:10
*** fnaval_ has joined #openstack-lbaas16:10
*** Vorrtex has quit IRC16:13
*** mithilarun has joined #openstack-lbaas16:14
*** rpittau is now known as rpittau|afk16:14
openstackgerritMerged openstack/octavia master: Add a note about nova hardware architectures  https://review.opendev.org/66573216:36
openstackgerritMerged openstack/octavia master: Clarify that the certificate guide should be used  https://review.opendev.org/66251216:36
*** pcaruana has joined #openstack-lbaas16:45
*** ricolin has quit IRC16:46
*** ramishra has quit IRC16:55
openstackgerritjacky06 proposed openstack/octavia-dashboard master: Sync Sphinx requirement  https://review.opendev.org/66605416:56
*** altlogbot_3 has quit IRC17:00
*** ccamposr__ has quit IRC17:01
*** altlogbot_3 has joined #openstack-lbaas17:01
*** irclogbot_2 has quit IRC17:02
*** irclogbot_2 has joined #openstack-lbaas17:03
openstackgerritMichael Johnson proposed openstack/octavia master: Specify the linux-image-kvm kernel for ubuntu  https://review.opendev.org/66586117:12
openstackgerritMerged openstack/octavia stable/stein: Fix a python3 issue in the amphora-agent  https://review.opendev.org/66569817:46
*** gcheresh has joined #openstack-lbaas18:00
*** lemko has quit IRC18:18
*** mithilarun has quit IRC18:20
*** mithilarun has joined #openstack-lbaas18:21
*** mithilarun has quit IRC18:25
*** mithilarun has joined #openstack-lbaas18:30
*** pcaruana has quit IRC18:45
*** mithilarun has quit IRC19:01
*** mithilarun has joined #openstack-lbaas19:02
colin-saw this and thought it might be worth sharing https://www.haproxy.com/blog/haproxy-2-0-and-beyond/19:02
*** pcaruana has joined #openstack-lbaas19:05
*** luksky has joined #openstack-lbaas19:05
johnsomYeah, all good stuff19:06
*** mithilarun has quit IRC19:06
openstackgerritMichael Johnson proposed openstack/octavia master: Specify the linux-image-kvm kernel for ubuntu  https://review.opendev.org/66586119:21
*** gcheresh has quit IRC19:35
*** ivve has joined #openstack-lbaas19:55
*** gcheresh has joined #openstack-lbaas20:14
*** mithilarun has joined #openstack-lbaas20:14
*** gcheresh has quit IRC20:35
*** boden has quit IRC21:25
*** tesseract has quit IRC21:47
*** ivve has quit IRC21:48
*** fnaval_ has quit IRC21:50
rm_worksweet, 2.0!21:55
rm_workwe can get that in CENT9 hopefully21:55
rm_workmaybe 202421:55
*** fnaval has joined #openstack-lbaas21:55
*** ccamposr has joined #openstack-lbaas21:57
rm_workjohnsom: cool so why -W?22:06
rm_workon the kvm thing22:06
rm_workHAProxy Data Plane API will be coolbeans to integrate :D22:07
johnsomWell, I want to test it and I'm getting push back on how I did the DIB side....22:07
johnsomrm_work Yeah, well, it requires another process22:07
rm_workthat's maybe ... good?22:07
rm_workbecause it can run outside the netns22:08
johnsomWell, they added "native" netns a release or two ago as well. If we trust that....22:08
johnsomBut, yeah, maybe it can run outside. I think so as they talk about side-car22:09
johnsomCool beans, the kvm kernel works, even on the xen hosts.22:09
johnsomAs I would expect given it's just removing a bunch of useless kernel modules22:10
rm_worklol22:12
rm_workyeah i did wonder about that22:12
rm_workbut ... lol ok22:12
rm_workso now centos will be even more comparatively bloated :D22:12
rm_workwoo HAProxy 2.1 Preview: UDP Support22:22
rm_worklol22:22
*** mithilarun has quit IRC22:23
johnsomSo comparing the old kernel to the new, size: 605397504 vs 393244160 according to glance22:25
johnsomWorth the effort.22:25
rm_workyes22:25
rm_workdefinitely22:26
rm_worki am still looking into switching us to ubuntu22:26
rm_worki got the OK to do it22:26
rm_workbut all our build systems only have RHEL images available right now and cross-building isn't easy22:26
johnsomNow I just need to negotiate with the DIB folks on which of the three ways to implement this we should go with.22:27
rm_workneed to figure out all the shit to install to get debootstrap and such on rhel22:27
johnsomThe diskimage-builder bin-dep should do that for you right?22:27
rm_workhmmm22:27
rm_workdoes it?22:27
rm_workmaybe i just suck at reading instructions22:27
rm_worki'll try presently22:27
*** luksky has quit IRC22:47
openstackgerritMichael Johnson proposed openstack/octavia master: Add get method support to the driver-agent  https://review.opendev.org/66502922:48
*** mithilarun has joined #openstack-lbaas22:55
*** ccamposr has quit IRC22:56
*** rtjure has quit IRC23:07
*** rcernin has joined #openstack-lbaas23:16
*** rcernin has quit IRC23:17
*** rcernin has joined #openstack-lbaas23:18
openstackgerritMichael Johnson proposed openstack/octavia master: Add get method support to the driver-agent  https://review.opendev.org/66502923:30
*** mithilarun has quit IRC23:54
*** mithilarun has joined #openstack-lbaas23:54
*** mithilarun has quit IRC23:58
« Monday, 2019-06-17 Index Wednesday, 2019-06-19 »

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