| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Updated from OpenStack Ansible Tests https://review.opendev.org/745851 | 00:04 |
|---|---|---|
| *** gyee has quit IRC | 00:56 | |
| *** spatel has joined #openstack-ansible | 01:52 | |
| spatel | good morning! | 01:52 |
| spatel | just testing my new IRC client | 01:52 |
| *** spatel has quit IRC | 01:58 | |
| *** dmsimard has quit IRC | 02:16 | |
| *** dmsimard7 has joined #openstack-ansible | 02:16 | |
| *** MickyMan77 has joined #openstack-ansible | 02:50 | |
| *** MickyMan77 has quit IRC | 02:59 | |
| *** MickyMan77 has joined #openstack-ansible | 03:15 | |
| *** idlemind has joined #openstack-ansible | 03:39 | |
| *** dmsimard7 is now known as dmsimard | 03:41 | |
| *** shyamb has joined #openstack-ansible | 04:17 | |
| *** MickyMan77 has quit IRC | 04:19 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-ansible | 04:33 | |
| *** shyam89 has joined #openstack-ansible | 05:15 | |
| *** MickyMan77 has joined #openstack-ansible | 05:15 | |
| *** shyamb has quit IRC | 05:18 | |
| *** MickyMan77 has quit IRC | 05:24 | |
| *** MickyMan77 has joined #openstack-ansible | 05:52 | |
| *** MickyMan77 has quit IRC | 06:01 | |
| *** miloa has joined #openstack-ansible | 06:12 | |
| *** d34dh0r53 has quit IRC | 06:21 | |
| *** pcaruana has joined #openstack-ansible | 06:26 | |
| *** viks____ has joined #openstack-ansible | 06:37 | |
| *** shyam89 has quit IRC | 06:58 | |
| noonedeadpunk | jrosser: it seems I need to branch asap :( I hoped I can do branching later, but got pinged about that.. | 07:33 |
| jrosser | https://review.opendev.org/#/c/753046/ that’s a bit confusing | 07:36 |
| jrosser | are we cycle-with-rc ? | 07:37 |
| noonedeadpunk | I thought it's cycle-trailing | 07:38 |
| noonedeadpunk | uh... | 07:39 |
| noonedeadpunk | I think that's the reason? https://opendev.org/openstack/releases/commit/8a9f755cb91dce22380998a28a653dd815875d15 | 07:40 |
| jrosser | that patch does not help make things clearer imho | 07:42 |
| *** tosky has joined #openstack-ansible | 07:43 | |
| noonedeadpunk | I think that 753046/ was done without look into that we have trailing type | 07:45 |
| jrosser | oh right so, i guess you just drop a -1 on https://review.opendev.org/#/c/753046/ making reference to https://github.com/openstack/releases/blob/master/deliverables/victoria/openstack-ansible.yaml#L6 | 07:45 |
| jrosser | feels like they have script or somthing to generate these patches so now we get included on main release schedule | 07:46 |
| noonedeadpunk | yeah, I think so | 07:48 |
| noonedeadpunk | but again, we should stop bumping master I guess, or switch it to victoria | 07:48 |
| jrosser | kolla doing same here https://review.opendev.org/#/c/752995/ | 07:52 |
| noonedeadpunk | hm, `Trailing deliverables trail the release, so they cannot, by definition, be independent. They need to pick between cycle-with-rc or cycle-with-intermediary models.` | 07:52 |
| noonedeadpunk | they made things more confusing imo | 07:53 |
| jrosser | agreed, seems to be word games / semantics somehow | 07:54 |
| noonedeadpunk | btw, how this has happened? https://zuul.opendev.org/t/openstack/build/279b01564a6e42f182b88e9d0bab3d55 | 07:55 |
| jrosser | i have no idea | 07:55 |
| jrosser | but in general we have huge unreliability in centos-8 and to an extent focal jobs | 07:56 |
| jrosser | epel repo is failing a lot | 07:56 |
| noonedeadpunk | but it's more about epel for centos 8 that making troubles | 07:56 |
| noonedeadpunk | yeah | 07:56 |
| noonedeadpunk | and here it was the cluster who was failing | 07:56 |
| jrosser | given that spatel is having success with the current galera version in HA i'm tempted to say we just recheck 752032 until it merges | 07:59 |
| jrosser | becasue that is blocking everything else | 08:00 |
| jrosser | and we can maybe delay 10.5.5 galera upgrade to W release to give some time to fix it properly | 08:00 |
| noonedeadpunk | they were failing the same may tbh... | 08:00 |
| noonedeadpunk | *way | 08:00 |
| jrosser | i know, it's ridiculous :( | 08:01 |
| noonedeadpunk | so if it pass, than maybe 10.5 will also pass... | 08:01 |
| jrosser | it was good the proposal bot ran yesterday | 08:01 |
| jrosser | has shown a bunch of repos are broken | 08:01 |
| *** shyamb has joined #openstack-ansible | 08:08 | |
| noonedeadpunk | manila and magnum are for sure:( | 08:28 |
| *** shyamb has quit IRC | 08:41 | |
| *** shyamb has joined #openstack-ansible | 09:19 | |
| *** MickyMan77 has joined #openstack-ansible | 09:31 | |
| *** shyam89 has joined #openstack-ansible | 09:37 | |
| *** MickyMan77 has quit IRC | 09:39 | |
| *** shyamb has quit IRC | 09:39 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova master: Make possible to set nova_libvirt_images_rbd_pool to empty string https://review.opendev.org/754147 | 09:51 |
| *** MickyMan77 has joined #openstack-ansible | 09:58 | |
| MickyMan77 | Hi all, I get this error when I run this scripts/bootstrap-ansible.sh || http://paste.openstack.org/show/798386/ | 10:06 |
| MickyMan77 | Failed to download remote objects and refs: fatal: error in object: unshallow 5dd4abe61adcdcc3950afd3950619c4c98b8b809\nfatal: The remote end hung up unexpectedly\n" | 10:06 |
| MickyMan77 | Failed to download remote objects and refs: fatal: error in object: unshallow 115e2897b51a632728d17c540f844a434d835227\nfatal: The remote end hung up unexpectedly\n" | 10:07 |
| noonedeadpunk | MickyMan77: I think there might be issue with opendev infra at moment | 10:12 |
| noonedeadpunk | so worth just trying again a bit later | 10:13 |
| *** SecOpsNinja has joined #openstack-ansible | 10:28 | |
| *** noonedeadpunk has quit IRC | 10:33 | |
| itandops | hi guys, I have an issue with galera after deploying openstack 20.1.6. I get this message "haproxy[2535]: backend galera-back has no server available!" which cause loosing API access. | 10:35 |
| itandops | I can see logs and my conf there: http://paste.openstack.org/show/798388/ I use the same conf to deploy 19.1.5 whithout any issue. | 10:36 |
| *** shyam89 has quit IRC | 10:38 | |
| SecOpsNinja | hi everyone. im having a strange error in one of my neutron routers. i have 2 interfaces (1 internal 10.0.0.0/24 and 1 external with 10.10.10.0/24) and have configured this router as the next hop (in the private network) for the external network. th eproblem seems that despite the router is receiving packages for the external network trought its private network, its not forwarding it to exte | 10:39 |
| SecOpsNinja | rnal interface. I have make a tcpdump in both tap interfaces and only get information of the ping requests in internal interfac. on the external network nothing but the icmp gets host unreachable. im using linux bridges so where is the problem? iptables? how dos neutron-l3-agent and neutron-linuxbrige-agent makes this l3 routes? | 10:39 |
| *** shyamb has joined #openstack-ansible | 10:52 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Updated from OpenStack Ansible Tests https://review.opendev.org/754164 | 11:18 |
| CeeMac | quick question, which libvirt version comes down with OSA stable/train? | 11:27 |
| CeeMac | i'm hitting a bug in libvirt v4 which was evidently resolved in v5 | 11:28 |
| CeeMac | and i'm presuming doing an inplace upgrade of libvirt is going to break some things | 11:28 |
| *** lkoranda has joined #openstack-ansible | 11:35 | |
| admin0 | jrosser, looks like i can't build up a hypervisor with ceph support and later change the nova.conf to have images_type = default and expect it to work :( | 11:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Switch tempestconf job from centos-7 to centos-8 https://review.opendev.org/754076 | 12:05 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Updated from OpenStack Ansible Tests https://review.opendev.org/754183 | 12:05 |
| SecOpsNinja | yep, from what im seeing in ip netns qrouter it seams to be a arp problem? but its strange because it works for other machine in same network... | 12:21 |
| SecOpsNinja | and that other machine can ping the machine that im having trouble getting the arp response... wtf? | 12:22 |
| mgariepy | my first PR on ceph got merged :D | 12:31 |
| *** shyamb has quit IRC | 12:31 | |
| jrosser | SecOpsNinja: is this vxlan? | 12:31 |
| SecOpsNinja | jrosser, nope the problem seems to be in the external provider vlan network. but only for some machines in that network... im trying to troubleshoot with arp and tcpdump do try to understand what is happening | 12:33 |
| SecOpsNinja | jrosser, from what im able to see the the problem seems to be in router. im making a tcpdump icmp and arp capture in pysical host network and in the tap router interface and it seems that the arp responsed arend't getting to the router | 12:47 |
| SecOpsNinja | so in bond interface i get the request and replys of arp but in the tap interface only get the arp requests. this seems a iptables problem... | 12:49 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_keystone master: Fix keystone nginx behaviour https://review.opendev.org/754382 | 12:54 |
| *** noonedeadpunk has joined #openstack-ansible | 12:56 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: Updated from OpenStack Ansible Tests https://review.opendev.org/754169 | 12:57 |
| *** spatel has joined #openstack-ansible | 13:06 | |
| *** mbuil has quit IRC | 13:07 | |
| spatel | good morning! finally i got sometime to come online | 13:07 |
| *** mbuil has joined #openstack-ansible | 13:07 | |
| noonedeadpunk | dmsimard: does ara now missing handlers execution for some reason? | 13:10 |
| noonedeadpunk | see no handlers here https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_b66/742105/5/check/openstack-ansible-functional-centos-8/b66b199/logs/ara-report/playbooks/1.html#results | 13:10 |
| noonedeadpunk | but task has eventually failed on handlers https://zuul.opendev.org/t/openstack/build/b66b199b9f7145d2b3025686f98b5ac5/log/job-output.txt#5216 | 13:11 |
| *** spatel has quit IRC | 13:12 | |
| *** itandops has quit IRC | 13:13 | |
| *** itandops has joined #openstack-ansible | 13:14 | |
| *** wpp has quit IRC | 13:16 | |
| *** sum12 has quit IRC | 13:17 | |
| *** sum12 has joined #openstack-ansible | 13:18 | |
| *** wpp has joined #openstack-ansible | 13:21 | |
| *** spatel has joined #openstack-ansible | 13:23 | |
| *** sum12 has quit IRC | 13:26 | |
| *** sum12 has joined #openstack-ansible | 13:30 | |
| CeeMac | noonedeadpunk: jrosser where would I go to see what versions of "things" are installed in a given release? | 13:43 |
| CeeMac | im sure there is a text file that is used for the versions | 13:44 |
| noonedeadpunk | https://opendev.org/openstack/openstack-ansible/src/branch/master/playbooks/defaults/repo_packages/openstack_services.yml ? | 13:44 |
| CeeMac | cant remember what its called or where | 13:44 |
| jrosser | CeeMac: you mean libvirt? or a service? | 13:44 |
| CeeMac | well, either, but in this instance, libvirt/qemu yes | 13:44 |
| CeeMac | wasnt sure if that would be in openstack-hosts role or nova role | 13:45 |
| jrosser | you will have the train version of the Ubuntu cloud repo (you’re on Ubuntu?) | 13:45 |
| CeeMac | ubuntu, yes | 13:45 |
| noonedeadpunk | yeah, I don't think we bump distro packages | 13:45 |
| CeeMac | oh, so its o/s dependant | 13:45 |
| CeeMac | from the distro | 13:45 |
| CeeMac | that makes sense actually | 13:45 |
| jrosser | from UCa | 13:45 |
| jrosser | everything we care about is in UCA rather than the actual Ubuntu repo | 13:46 |
| CeeMac | uca being? | 13:46 |
| CeeMac | not familiar with that | 13:46 |
| noonedeadpunk | ubuntu cloud archive | 13:47 |
| CeeMac | ohhhh | 13:47 |
| CeeMac | ok | 13:47 |
| noonedeadpunk | https://wiki.ubuntu.com/OpenStack/CloudArchive | 13:47 |
| CeeMac | thanks :) | 13:47 |
| jrosser | that contains the Ubuntu openstack packages but also all the important dependencies | 13:47 |
| jrosser | so a coherent set for bionic/train or whatever release you need | 13:48 |
| CeeMac | great | 13:48 |
| jrosser | just watch out for libvirt python bindings if you try to wedge v5 in somehow | 13:48 |
| CeeMac | yeah, not sure i want to do that tbh | 13:49 |
| CeeMac | but i might have to if i keep having these issues | 13:49 |
| CeeMac | i'm hoping its in the right version for train then i'm golden | 13:49 |
| *** lkoranda has quit IRC | 13:49 | |
| CeeMac | looks promising: libvirt5.4.0-0ubuntu5.4 | 13:50 |
| CeeMac | actually, even stein would do at 5.0.0-1 | 13:51 |
| CeeMac | looks like I know what I'm doing next week! | 13:51 |
| CeeMac | thanks for the pointers as always :) | 13:51 |
| CeeMac | oh, did the release notes get updated in stein for the known issues around repo sync, and was it rabbit or galera cluster that required some TLC? | 13:52 |
| CeeMac | R > S | 13:52 |
| noonedeadpunk | iirc we got type in lsyncd config which should be fixed nowadays | 13:53 |
| noonedeadpunk | *typo | 13:53 |
| CeeMac | ok thanks | 13:55 |
| dmsimard | noonedeadpunk: not sure tbh | 14:02 |
| dmsimard | noonedeadpunk: not going to have the time to troubleshoot soon but if you think handlers are really missing that's a bug | 14:03 |
| noonedeadpunk | yeah, don't see handlers anywhere :( | 14:04 |
| dmsimard | could you create an issue on github ? would be very appreciated <3 | 14:04 |
| *** shyamb has joined #openstack-ansible | 14:07 | |
| noonedeadpunk | dmsimard: https://github.com/ansible-community/ara/issues/178 is it ok? | 14:13 |
| dmsimard | very much so | 14:14 |
| dmsimard | thanks! | 14:14 |
| dmsimard | do you have jobs running ansible 2.7 or 2.8 ? curious if it reproduces there too | 14:14 |
| noonedeadpunk | we're running on 2.9.0 | 14:15 |
| noonedeadpunk | *2.9.9 only | 14:15 |
| noonedeadpunk | we have older ansible for older branches only | 14:16 |
| noonedeadpunk | but we don't have modern ara there... | 14:16 |
| dmsimard | ok np it's something I can find out | 14:16 |
| *** d34dh0r53 has joined #openstack-ansible | 14:16 | |
| dmsimard | I guess you're right, there's no tasks with handler:true https://api.demo.recordsansible.org/api/v1/tasks?playbook=1007&handler=true | 14:21 |
| dmsimard | there's none at all :O https://api.demo.recordsansible.org/api/v1/tasks?handler=true | 14:22 |
| dmsimard | good catch | 14:22 |
| SecOpsNinja | yep i do have a strange error (in network interfaces, debian 10 kernel seetings or neutron configuration) because i haven't changed any thing and now the arp request passed from the bond1 vlan interface > tap -> internal router interface (a few hours agor it was only passing the arp requests from the router to the bond1 and bloquink the arp replys at tap interface). do you have any idea wat | 14:29 |
| SecOpsNinja | h maybe causing this? this is a every intermitent problem and im not able to find what is causing this... | 14:29 |
| *** d34dh0r53 has quit IRC | 14:31 | |
| *** nurdie has quit IRC | 14:33 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Revert "Improve compatibility with connection delegation" https://review.opendev.org/752032 | 14:35 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Update galera to 10.5.5 https://review.opendev.org/742105 | 14:38 |
| *** nurdie has joined #openstack-ansible | 14:41 | |
| *** shyamb has quit IRC | 14:45 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_octavia master: Define condition for the first play host one time https://review.opendev.org/754411 | 14:47 |
| *** d34dh0r53 has joined #openstack-ansible | 14:48 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_panko master: Define condition for the first play host one time https://review.opendev.org/754414 | 14:50 |
| *** pcaruana has quit IRC | 14:51 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_placement master: Define condition for the first play host one time https://review.opendev.org/754419 | 14:54 |
| noonedeadpunk | jrosser: fwiw, alera 10.5.5 works perfectly on 2.9.13 ansible with https://review.opendev.org/#/c/752032/ applied. and errors I saw previously were super weird, like ansible did bootstrap on wrong host or smth like that... | 14:55 |
| noonedeadpunk | *galera | 14:55 |
| noonedeadpunk | it was reproduced pretty easily, and now 3 runs on a row in sandbox without errors... | 14:56 |
| jrosser | but still in CI we see trouble? | 14:57 |
| noonedeadpunk | I guess we see it because removed ansible_host and ansible is still not in fixed state | 14:57 |
| noonedeadpunk | (ie 2.9.9) | 14:57 |
| noonedeadpunk | so we might be lucky and somehow pass with n of rechecks | 14:58 |
| jrosser | and we can’t merge 2.9.13 without first fixing galena role | 14:58 |
| noonedeadpunk | we can set centos to nv if got tired of rechecks | 14:59 |
| *** miloa has quit IRC | 15:00 | |
| *** pcaruana has joined #openstack-ansible | 15:03 | |
| *** shyamb has joined #openstack-ansible | 15:10 | |
| *** d34dh0r53 has quit IRC | 15:11 | |
| *** d34dh0r53 has joined #openstack-ansible | 15:12 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_sahara master: Define condition for the first play host one time https://review.opendev.org/754425 | 15:16 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_swift master: Define condition for the first play host one time https://review.opendev.org/754428 | 15:23 |
| *** djhankb has quit IRC | 15:27 | |
| *** djhankb has joined #openstack-ansible | 15:27 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_tacker master: Define condition for the first play host one time https://review.opendev.org/754430 | 15:29 |
| *** MickyMan77 has quit IRC | 15:33 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_trove master: Define condition for the first play host one time https://review.opendev.org/754431 | 15:35 |
| *** shyamb has quit IRC | 15:42 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-galera_server master: Revert "Improve compatibility with connection delegation" https://review.opendev.org/752032 | 15:49 |
| *** gyee has joined #openstack-ansible | 15:55 | |
| *** MickyMan77 has joined #openstack-ansible | 16:13 | |
| *** watersj has quit IRC | 16:18 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-rsyslog_client master: Updated from OpenStack Ansible Tests https://review.opendev.org/754189 | 16:18 |
| *** MickyMan77 has quit IRC | 16:22 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: Add option to override horizon policies https://review.opendev.org/748264 | 16:23 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: Do not manage /etc/ssl or /etc/pki directories or symlinks https://review.opendev.org/754093 | 16:24 |
| fridtjof[m] | I feel like the deployment target host setup docs could use some love. | 16:29 |
| fridtjof[m] | They have not changed much over recent releases, and a lot of it is unnecessary/does not work on recent ubuntu versions for example | 16:30 |
| fridtjof[m] | for example, the linux-image-extra packages don't seem to exist anymore beginning with ubuntu 18.04 | 16:31 |
| fridtjof[m] | also, ensuring a kernel version > 3.13 is completely pointless if OSA only supports 18.04 which initially shipped with 4.13 (i think?) | 16:34 |
| fridtjof[m] | maybe i'll propose a patch later, if I have time to figure out your patch process :) | 16:35 |
| *** MickyMan77 has joined #openstack-ansible | 16:51 | |
| SecOpsNinja | one quick questions shouldn't the tap interface have the same mac as the internal interface in neutron router? | 16:51 |
| noonedeadpunk | I don't think it should | 16:55 |
| noonedeadpunk | as tap is outside of ns, and it maps to qr interface or qg | 16:56 |
| noonedeadpunk | https://docs.openstack.org/openstack-ansible/latest/ru/reference/architecture/container-networking.html#neutron-traffic | 16:56 |
| jrosser | just ignore the container on that diagram for the neutron agent | 16:56 |
| noonedeadpunk | yeah... | 16:57 |
| SecOpsNinja | noonedeadpunk, ok thanks. im trying to understand a problem that i have that some times the arp replys dont go forwarder to inside the tap / router interface | 16:57 |
| SecOpsNinja | despite shouwing on bond interface of the same bridge | 16:57 |
| fridtjof[m] | Do you have any tips on putting /etc/openstack_deploy in a git repo? | 16:57 |
| *** MickyMan77 has quit IRC | 16:57 | |
| *** MickyMan77 has joined #openstack-ansible | 16:58 | |
| jrosser | fridtjof[m]: it works just fine :) | 16:58 |
| jrosser | and is also an *excellent* plan | 16:58 |
| *** MickyMan77 has quit IRC | 16:58 | |
| *** MickyMan77 has joined #openstack-ansible | 16:59 | |
| fridtjof[m] | What do I do about the files that are created at runtime (eg the inventory files, and user_secrets? I'm leaning towards ignoring them | 16:59 |
| jrosser | you can make your deploy host however you like | 16:59 |
| jrosser | mine is deployed itself with ansible, and /etc/openstack_deploy is cloned into place | 16:59 |
| jrosser | then user_secrets.yml is read from a encrypted key/value store | 17:00 |
| fridtjof[m] | Ah, like Vault? | 17:00 |
| jrosser | hashi vault yes | 17:00 |
| SecOpsNinja | i will confirme next time that this fails again, but it seems that the mac address do te qg-ID interface wasn't in linux bridge forwarding table that was the reason to be droped but its sstrange that started to work without changes in configurations.... | 17:00 |
| jrosser | fridtjof[m]: i would tend to put all of the files in git, especially the inventory files. those are the ones you want to be able to back out changes from if something goes bad | 17:01 |
| jrosser | you protect yourself against the dynamic invetory doing something you then don't like | 17:02 |
| *** MickyMan77 has quit IRC | 17:03 | |
| fridtjof[m] | hm, good point. | 17:06 |
| fridtjof[m] | Importing changes seems like it'll be painful though, I'm trying to keep the deploy server's git access readonly | 17:07 |
| fridtjof[m] | Is 19.1.3 actually a released version? There are no release notes for it, but there is a git tag. I'm unsure now | 17:16 |
| noonedeadpunk | we don't have relese note for each version | 17:17 |
| noonedeadpunk | as we write notes only for some significat things only | 17:17 |
| *** SecOpsNinja has left #openstack-ansible | 17:19 | |
| fridtjof[m] | ah, i see it's basically just a sha bump | 17:20 |
| *** MickyMan77 has joined #openstack-ansible | 17:33 | |
| *** MickyMan77 has quit IRC | 17:41 | |
| fridtjof[m] | I'm interested in using tempest. Is it just meant for OSA CI? Or can I use it to test my deployment too? | 17:48 |
| itandops | noonedeadpunk: any suggestion about my error ? | 17:51 |
| noonedeadpunk | which one? | 17:51 |
| itandops | I have an issue with galera after deploying openstack 20.1.6. I get this message "haproxy[2535]: backend galera-back has no server available!" which cause loosing API access. I can see logs and my conf there: http://paste.openstack.org/show/798388/ I use the same conf to deploy 19.1.5 whithout any issue. | 17:52 |
| noonedeadpunk | fridtjof[m]: um, we want to make it usable for testing deployments by users, but I don't think we used/tested it outside of CI | 17:52 |
| noonedeadpunk | So you should try and tell us if smth goes wrong so we could fix that) | 17:52 |
| fridtjof[m] | will do! | 17:53 |
| itandops | noonedeadpunk: since the version of galera has changed since 20.0.0 release. I wonder if a miss some config ? | 17:53 |
| fridtjof[m] | last question before I do though - how does tempest behave with e.g. public/physical networks? Currently I only have one flat network, which exists already (because it's being used of course) | 17:54 |
| noonedeadpunk | fridtjof[m]: so, you should provide public network to the tempest role. and private ones may be either created for you or you may create it as well and provide to tempest role | 18:01 |
| fridtjof[m] | ah, okay. So if the network already exists, tempest won't fail | 18:01 |
| noonedeadpunk | also you would probably to keep `tempest_run` as falso to avoid running these tasks https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_resources.yml | 18:02 |
| fridtjof[m] | for normal deployment runs? | 18:05 |
| *** ChiTo has joined #openstack-ansible | 18:06 | |
| fridtjof[m] | os-tempest-install doesn't do anything if tempest_run is false, so I have to set it for tempest to do anything | 18:06 |
| noonedeadpunk | ah, damn it, yeah, https://review.opendev.org/#/c/739809/ just landed... | 18:07 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/ussuri: Run tempest install when tempest_run is False https://review.opendev.org/754465 | 18:07 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Run tempest install when tempest_run is False https://review.opendev.org/754466 | 18:07 |
| ChiTo | Hi everybody, I am starting to configure my networking components based on openvswitch with DVR, and I noticing that the br-vlan / br-vxlan or any br-provider network is being created by the playbook, but it is struggling because I don't want that the playbooks creates them, is there any possibility to tell that don't create them due I have already create them? Also I am installing the neutron on bare metal servers, not sure if | 18:07 |
| ChiTo | any variable container_* is neccesary on that sense? | 18:07 |
| fridtjof[m] | ah, still on stein | 18:08 |
| noonedeadpunk | ChiTo: I'm pretty sure we don not do network configuration (especially bridges creation) except aio builds | 18:08 |
| fridtjof[m] | okay anyway, if I execute os-tempest-install with tempest_run, I get this: | 18:09 |
| fridtjof[m] | ConflictException: 409: Client Error for url: http://10.1.60.11:9696/v2.0/networks.json, Unable to create the flat network. Physical network prod-app is in use. | 18:09 |
| noonedeadpunk | so if you don;t run bootstrap-aio, none networking should be done by OSA | 18:09 |
| fridtjof[m] | So it seems it always tries to create the network, even if it already exists | 18:09 |
| noonedeadpunk | I guess this might be also fixed with collections or more modern ansible:( | 18:12 |
| noonedeadpunk | I think what you can do is to jsut comment out stuff from tempest_resources.yml or add when: false.... | 18:12 |
| fridtjof[m] | Would you say Train or Ussuri are more stable than Stein? | 18:13 |
| fridtjof[m] | I might just start migrating | 18:13 |
| fridtjof[m] | (or at least, not less stable lol) | 18:14 |
| noonedeadpunk | not stable for sure. we deploy things and rarely backport them, because features shouldn't be backported actually.. | 18:14 |
| fridtjof[m] | *not less stable, i assume? :D | 18:15 |
| noonedeadpunk | but in terms of tempest things are actually not so straightforward unfortunatelly | 18:15 |
| ChiTo | noonedeadpunk: I noticed that the ovs-vsctl showed the creation of br-vxlan and br-vlan but since I have already create them it struggled showing: "error: "could not add network device br-vxlan to ofproto (File exists)" | 18:15 |
| noonedeadpunk | *not less stable, sorry) | 18:15 |
| noonedeadpunk | oh, ovs should create all bridges iirc | 18:16 |
| noonedeadpunk | and it's point of ovs, that it handles networking, and you say it they way it should do it | 18:16 |
| noonedeadpunk | but I think you may ask jamesdenton - he's great expert in networking! | 18:17 |
| ChiTo | noonedeadpunk: I am going to try to remove them on my metal servers to let the OVS just to create them | 18:18 |
| noonedeadpunk | yep, I think it's the way | 18:18 |
| jamesdenton | ChiTo i would update your openstack_user_config provider network definitions to use something other than br-vlan as the name. I like to use br-provider or something like that | 18:19 |
| jamesdenton | and the neutron playbook will try to create the OVS bridge for you, which is what i suspect is happening here. there's already a linuxbridge with thesame name | 18:20 |
| fridtjof[m] | btw, is there a script to show all git logs for all roles/projects between two OSA releases? | 18:23 |
| jamesdenton | have to step away for about 45 min. if you want to post your yml config files and the output of brctl show and ovs-vsctl on paste.openstack.org show i can try to take a look | 18:24 |
| *** MickyMan77 has joined #openstack-ansible | 18:24 | |
| noonedeadpunk | fridtjof[m]: I think we don't have any out of the box | 18:25 |
| fridtjof[m] | Okay, I guess it would be relatively easy, at least for the ansible roles | 18:28 |
| noonedeadpunk | sorry, I also need to go off -pretty late here | 18:28 |
| fridtjof[m] | bye! | 18:29 |
| *** MickyMan77 has quit IRC | 18:33 | |
| *** pfsmorigo has quit IRC | 18:36 | |
| *** spatel has quit IRC | 18:38 | |
| ChiTo | jamesdenton: sounds great, thankss for the hint, same for vxlan right? | 18:42 |
| ChiTo | jamesdenton: I am redeploying trying to change the bridges names to let the OVS to create them and I will keep you updated. Thanks a lot! | 18:42 |
| *** MickyMan77 has joined #openstack-ansible | 19:10 | |
| *** MickyMan77 has quit IRC | 19:18 | |
| jamesdenton | ordinarily there wouldn't be an OVS bridge for br-vxlan. | 19:51 |
| jamesdenton | in fact, i think we're in the process of updating docs to help clarify that | 19:51 |
| *** ChiTo has quit IRC | 19:59 | |
| openstackgerrit | Merged openstack/ansible-role-systemd_service master: Updated from OpenStack Ansible Tests https://review.opendev.org/754196 | 20:01 |
| *** spatel has joined #openstack-ansible | 20:09 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Revert "Improve compatibility with connection delegation" https://review.opendev.org/752032 | 20:10 |
| *** spatel has quit IRC | 20:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_blazar master: Updated from OpenStack Ansible Tests https://review.opendev.org/754163 | 20:30 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift master: Updated from OpenStack Ansible Tests https://review.opendev.org/754181 | 20:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_mistral master: Updated from OpenStack Ansible Tests https://review.opendev.org/754174 | 20:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara master: Updated from OpenStack Ansible Tests https://review.opendev.org/754179 | 20:50 |
| *** nurdie_ has joined #openstack-ansible | 22:13 | |
| *** nurdie has quit IRC | 22:16 | |
| *** nurdie_ has quit IRC | 22:19 | |
| *** nurdie has joined #openstack-ansible | 22:19 | |
| *** nurdie has quit IRC | 22:23 | |
| *** tosky has quit IRC | 22:58 | |
| *** MickyMan77 has joined #openstack-ansible | 23:00 | |
| *** MickyMan77 has quit IRC | 23:09 | |
| *** ioni has quit IRC | 23:13 | |
| *** ChosSimbaOne_ has quit IRC | 23:27 | |
| *** ChosSimbaOne has joined #openstack-ansible | 23:28 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!