| JudeC | johnsom: In regard to your comment about searching for the subnet using its name or UUID. Is the best way to do that through the neutron API? | 00:01 |
|---|---|---|
| *** aojea has quit IRC | 00:02 | |
| johnsom | JudeC Yes, there should be other examples. Let me see if I can find one | 00:02 |
| JudeC | sweet, just making sure I am barking up the right tree. | 00:03 |
| johnsom | This: https://github.com/openstack/python-neutronclient/blob/master/neutronclient/osc/v2/trunk/network_trunk.py#L387 and https://github.com/openstack/python-neutronclient/blob/master/neutronclient/osc/v2/trunk/network_trunk.py#L329 | 00:06 |
| johnsom | Though those might not be the best examples. Let me keep looking | 00:06 |
| johnsom | Here is another way: https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/networks.py#L35 | 00:08 |
| johnsom | Still not sure that is great eitehr | 00:08 |
| johnsom | There is this too in osc-lib: https://github.com/openstack/osc-lib/blob/master/osc_lib/api/api.py#L314 | 00:11 |
| *** bzhao has quit IRC | 00:11 | |
| JudeC | ok I think I can get the osc_lib way working pretty easily. | 00:12 |
| JudeC | thanks a bunch :) | 00:12 |
| johnsom | NP | 00:13 |
| *** JudeC has quit IRC | 00:18 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Change auth_strategy default to keystone https://review.openstack.org/459654 | 00:35 |
| rm_work | johnsom: this is pissing me off | 00:35 |
| rm_work | something about the wsgi change is broken | 00:35 |
| rm_work | see #infra, but | 00:35 |
| rm_work | [09:13:17] <rm_work>drwxr-xr-x 2 stack root 4096 Apr 30 23:48 /var/cache/octavia | 00:36 |
| rm_work | [09:13:25] <rm_work>stack 7627 0.0 0.0 13340 1100 '?' S 23:48 0:00 grep apache2 root 27414 0.0 0.0 89992 5976 '?' Ss 23:39 0:00 /usr/sbin/apache2 -k start www-data 27417 0.0 0.1 448592 15116 '?' Sl 23:39 0:00 /usr/sbin/apache2 -k start www-data 27418 0.0 0.1 514144 13852 '?' Sl 23:39 0:00 /usr/sbin/apache2 -k start | 00:36 |
| *** amotoki has joined #openstack-lbaas | 00:38 | |
| johnsom | Well, we can specify the user to run it under... | 00:40 |
| rm_work | user seems right | 00:40 |
| rm_work | but can't access | 00:40 |
| rm_work | 2017-05-01 00:05:27.362008 keystonemiddleware.exceptions.ConfigurationError: unable to access signing_dir /var/cache/octavia | 00:40 |
| rm_work | no idea why | 00:40 |
| rm_work | this is the exact config glance has | 00:40 |
| rm_work | same directory | 00:40 |
| rm_work | same owners and such | 00:41 |
| rm_work | trying to track down our wsgi script to see wtf it's trying to do | 00:41 |
| johnsom | Well, apache chowns the process. Where id you get that log? A previous run? | 00:41 |
| rm_work | so in this run | 00:42 |
| rm_work | http://logs.openstack.org/54/459654/9/check/gate-octavia-v1-dsvm-py3x-scenario-ubuntu-xenial/7f937da/logs/apache/octavia_error.txt.gz#_2017-05-01_00_05_27_362008 | 00:42 |
| *** amotoki has quit IRC | 00:42 | |
| rm_work | let me find it again | 00:42 |
| rm_work | http://logs.openstack.org/54/459654/9/check/gate-octavia-v1-dsvm-py3x-scenario-ubuntu-xenial/7f937da/logs/devstacklog.txt.gz#_2017-04-30_23_48_45_956 | 00:43 |
| rm_work | I put some echos | 00:43 |
| rm_work | for debugging | 00:43 |
| rm_work | echo -e `ls -ld $OCTAVIA_AUTH_CACHE_DIR` | 00:43 |
| rm_work | echo -e `ps aux | grep $APACHE_NAME` | 00:43 |
| rm_work | the job that's running now I just put chmod 777 to see if it's a permissions issue or like a chroot issue | 00:44 |
| johnsom | No, it's most likely running under the wrong owner | 00:45 |
| rm_work | well that grep shows `stack` | 00:45 |
| johnsom | It looks like most teams are running under a service account, "octavia" for example. | 00:47 |
| rm_work | ok | 00:48 |
| rm_work | you want to fix it inside that patch? | 00:48 |
| johnsom | Hmm. My suspicion is we just need to add "user=" to the WSGIDaemonProcess in the template. | 00:48 |
| rm_work | ah k | 00:48 |
| rm_work | give me the location and i can do it | 00:48 |
| rm_work | or you can do it | 00:48 |
| johnsom | Well, I'm not sure "octavia" is the right answer, as I'm not sure we set that up | 00:48 |
| rm_work | remove the 777 then | 00:48 |
| rm_work | k | 00:48 |
| johnsom | It lives here: https://github.com/openstack/octavia/blob/master/devstack/files/wsgi/octavia-api.template#L21 | 00:49 |
| rm_work | k | 00:50 |
| johnsom | Ok, I think two things might need to happen. | 00:51 |
| johnsom | 1. change the template to have user octavia, probably should be a sed from the settings file | 00:51 |
| johnsom | 2. change the create_cache_dir in the devstack plugin to create the directory with octavia as the owenr | 00:52 |
| johnsom | Want me to do it? | 00:52 |
| rm_work | yeah IMO | 00:54 |
| rm_work | i mean | 00:54 |
| rm_work | i'm THERE | 00:54 |
| rm_work | but kinda distracted | 00:54 |
| rm_work | if you want to just do it | 00:54 |
| johnsom | Ok | 00:54 |
| johnsom | Just not sure what else I will break doing this, but I guess we will see | 00:55 |
| johnsom | Why did you take this out? | 00:55 |
| johnsom | iniset $OCTAVIA_CONF service_auth signing_dir $signing_dir | 00:55 |
| rm_work | it's deprecated in our config | 00:56 |
| rm_work | and doing the other keystone middleware line DOES IT | 00:56 |
| rm_work | internally | 00:56 |
| rm_work | so ours is pointless | 00:56 |
| rm_work | configure_auth_token_middleware $OCTAVIA_CONF octavia $OCTAVIA_AUTH_CACHE_DIR | 00:57 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Change auth_strategy default to keystone https://review.openstack.org/459654 | 01:03 |
| johnsom | Let's try that | 01:03 |
| rm_work | $OCTAVIA is a username? | 01:08 |
| rm_work | hmm | 01:16 |
| rm_work | something didn't work... | 01:16 |
| rm_work | ah yeah that doesn't create a SYSTEM user | 01:43 |
| rm_work | just the keystone user | 01:43 |
| rm_work | johnsom: ^^ | 01:45 |
| rm_work | don't know if we want to do that | 01:45 |
| rm_work | or just... | 01:45 |
| rm_work | actually use the STACK user | 01:45 |
| rm_work | it seems others just do that maybe... | 01:45 |
| johnsom | Ah, ok, my quick glance at devstack scripts directed me wrong. | 01:46 |
| johnsom | It just seems like everyone is using their own username for the apache wsgi | 01:47 |
| johnsom | hmmm | 01:47 |
| *** JudeC has joined #openstack-lbaas | 01:47 | |
| rm_work | hold on | 01:48 |
| rm_work | gonna put it back to stack user | 01:48 |
| rm_work | i think it's fine | 01:48 |
| rm_work | with the other changes you made | 01:48 |
| rm_work | 1 sec | 01:48 |
| rm_work | there IS an $APACHE_USER | 01:49 |
| rm_work | but | 01:49 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Change auth_strategy default to keystone https://review.openstack.org/459654 | 01:49 |
| rm_work | it defaults to same as $STACK_USER | 01:49 |
| rm_work | but probably safest to use it anyway | 01:49 |
| rm_work | I believe with your user thing set it might work | 01:49 |
| rm_work | see that change | 01:49 |
| rm_work | (also moved the service account create back to the old spot) | 01:49 |
| rm_work | (and put my debugs in again temporarily) | 01:50 |
| rm_work | johnsom: ^^ | 01:52 |
| *** aojea has joined #openstack-lbaas | 01:58 | |
| *** aojea has quit IRC | 02:04 | |
| rm_work | johnsom: did you happen to catch the logs from my `chmod 777` attempt? | 02:19 |
| rm_work | i never actually resolved 100% that the issue was permissions | 02:19 |
| rm_work | but hoping to see some results from this maybe... if we're lucky | 02:19 |
| rm_work | oh | 02:21 |
| rm_work | it looks like it worked | 02:21 |
| rm_work | i see one ubuntu dsvm success | 02:21 |
| rm_work | and two octavia dsvm | 02:22 |
| rm_work | woooo | 02:22 |
| rm_work | ok should i remove those debugs? | 02:22 |
| rm_work | part of me just wants to +A this rofl | 02:22 |
| *** yamamoto has joined #openstack-lbaas | 02:24 | |
| rm_work | ah damn there's one failure, wtf | 02:30 |
| rm_work | ah, nova servers aren't building | 02:31 |
| rm_work | k | 02:31 |
| rm_work | well then i'll submit the patch to remove those debugs | 02:31 |
| rm_work | right, the OVH problem is back | 02:32 |
| rm_work | fff | 02:32 |
| rm_work | with the number of gates we have voting now that require VMs, this is kinda shitty, we have like a 50% or higher chance to get one voting failure, i think | 02:32 |
| rm_work | based on rough head-maths | 02:32 |
| rm_work | oh right nm the neutron-lbaas ones aren't voting yet, yay me | 02:33 |
| rm_work | johnsom: i'm assuming you're in and out | 02:33 |
| rm_work | i'll post this once the last test finishes there (I want it to actually record that run) | 02:34 |
| rm_work | and we can +2/+A :) | 02:34 |
| *** amotoki has joined #openstack-lbaas | 02:39 | |
| *** amotoki has quit IRC | 02:43 | |
| johnsom | Yeah, on and off | 02:44 |
| johnsom | Yeah, I'm not sure what is up with OVH, some pass, some don't | 02:44 |
| *** JudeC has quit IRC | 02:44 | |
| rm_work | ah two fails | 02:46 |
| rm_work | >_> | 02:46 |
| rm_work | .... want this to finish so i can kick the next | 02:51 |
| rm_work | but it's taking forever | 02:51 |
| rm_work | and i need to go to lunch now... so maybe I will just give up waiting even though it's close | 02:51 |
| rm_work | log for a fail: http://logs.openstack.org/54/459654/12/check/gate-octavia-v1-dsvm-scenario-ubuntu-xenial/dff38c1/ | 02:52 |
| rm_work | log for a success: http://logs.openstack.org/54/459654/12/check/gate-octavia-v1-dsvm-py3x-scenario-ubuntu-xenial/302dbf8/ | 02:52 |
| rm_work | hmm | 02:52 |
| rm_work | interestingly it was both py2 that failed and both py3 that succeeded >_> | 02:52 |
| rm_work | probably coincidence | 02:52 |
| rm_work | ... and py3 on n-lbaas that are succeeded | 02:52 |
| rm_work | and py2 that are still going... | 02:52 |
| rm_work | umm | 02:52 |
| rm_work | maybe not coincidence? | 02:53 |
| rm_work | >_> | 02:53 |
| rm_work | ah ok nope, there we go | 02:53 |
| rm_work | success on a py2 | 02:53 |
| rm_work | so it is just coincidence <_< weird | 02:54 |
| rm_work | wish this last one would finish | 02:54 |
| rm_work | k don't care | 02:55 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Change auth_strategy default to keystone https://review.openstack.org/459654 | 02:55 |
| *** armax has joined #openstack-lbaas | 03:06 | |
| *** fnaval has quit IRC | 03:10 | |
| *** fnaval has joined #openstack-lbaas | 03:13 | |
| *** yamamoto has quit IRC | 03:27 | |
| *** aojea has joined #openstack-lbaas | 03:35 | |
| *** aojea has quit IRC | 03:40 | |
| *** armax has quit IRC | 03:41 | |
| *** aojea has joined #openstack-lbaas | 03:45 | |
| *** aojea has quit IRC | 03:46 | |
| *** aojea has joined #openstack-lbaas | 03:46 | |
| *** aojea has quit IRC | 03:51 | |
| *** armax has joined #openstack-lbaas | 04:00 | |
| *** armax has quit IRC | 04:09 | |
| *** yamamoto has joined #openstack-lbaas | 04:09 | |
| *** links has joined #openstack-lbaas | 04:12 | |
| *** armax has joined #openstack-lbaas | 04:14 | |
| *** gcheresh has joined #openstack-lbaas | 04:22 | |
| *** armax has quit IRC | 04:46 | |
| *** gcheresh has quit IRC | 04:46 | |
| *** armax has joined #openstack-lbaas | 05:00 | |
| rm_work | johnsom: feel free to +A :P | 05:01 |
| *** armax has quit IRC | 05:04 | |
| *** gcheresh has joined #openstack-lbaas | 05:05 | |
| *** krypto has joined #openstack-lbaas | 05:15 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Fix a bug where ports may not be deleted https://review.openstack.org/431324 | 05:24 |
| rm_work | johnsom: got bored waiting for you to fix this one: | 05:24 |
| rm_work | so I fixed it :P | 05:24 |
| rm_work | brb | 05:24 |
| *** JudeC has joined #openstack-lbaas | 05:43 | |
| *** aojea has joined #openstack-lbaas | 05:47 | |
| *** aojea has quit IRC | 05:52 | |
| *** aojea has joined #openstack-lbaas | 06:19 | |
| *** krypto has quit IRC | 06:22 | |
| rm_work | AUGH I MISSED A PEP8 | 06:23 |
| rm_work | I am a shame to my family | 06:24 |
| rm_work | oh didn't mean to leave that in anyway | 06:26 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Fix a bug where ports may not be deleted https://review.openstack.org/431324 | 06:26 |
| nmagnezi | rm_work, LOL | 06:37 |
| rm_work | o/ | 06:37 |
| nmagnezi | o/ | 06:37 |
| nmagnezi | rm_work, thanks for helping with the noauth/keystone patch. I planned to continue with it next week but i guess the sooner the better :) | 06:39 |
| rm_work | johnsom helped as well :) | 06:40 |
| rm_work | yeah this particular one was annoying me | 06:40 |
| rm_work | and I had started on it before | 06:40 |
| *** amotoki has joined #openstack-lbaas | 06:40 | |
| rm_work | and was about to pick it back up when you submitted one | 06:40 |
| nmagnezi | oh :D | 06:40 |
| rm_work | it would have been a lot easier if it'd merged BEFORE the wsgi-server patch | 06:40 |
| * rm_work sighs | 06:40 | |
| rm_work | mostly it was an issue with that thing's config | 06:41 |
| nmagnezi | and yeah I saw Michael helped but he is prolly sleeping now so i wouldn't ping him :D | 06:41 |
| rm_work | lol | 06:41 |
| rm_work | my theory is, if people are available to be pinged, they'll be pinged, and if not, it's their own fault if they don't mute their phones <_< | 06:41 |
| rm_work | but i leave the decision about availability in their hands | 06:41 |
| rm_work | no assumptions on my part :P | 06:41 |
| *** amotoki has quit IRC | 06:45 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Vote: Should project_id remain in the response for all objects? https://review.openstack.org/461344 | 06:46 |
| rm_work | nmagnezi: ^^ what do you think | 06:46 |
| rm_work | JudeC: ^^ | 06:46 |
| * nmagnezi looks | 06:47 | |
| rm_work | just throwing it out there, easier to just do the code and put it up for review, than to try to explain it | 06:47 |
| JudeC | looking | 06:48 |
| JudeC | LGTM lol but thats not saying a whole lot :) | 06:50 |
| rm_work | it's kinda an opinion piece :P | 06:53 |
| *** tesseract has joined #openstack-lbaas | 07:06 | |
| *** rcernin has joined #openstack-lbaas | 07:12 | |
| nmagnezi | rm_work, I think it's a good idea overall, but i'd like to check something in the v2 api ref | 07:13 |
| nmagnezi | rm_work, but looks like https://developer.openstack.org/api-ref/load-balancing/v2/index.html is broken | 07:13 |
| nmagnezi | rm_work, or did I get the wrong url? | 07:13 |
| nmagnezi | rm_work, ah, it's https://developer.openstack.org/api-ref/load-balancer/v2/index.html | 07:13 |
| *** rcernin has quit IRC | 07:32 | |
| *** JudeC has quit IRC | 07:48 | |
| *** JudeC has joined #openstack-lbaas | 08:17 | |
| *** aojea has quit IRC | 08:50 | |
| *** aojea has joined #openstack-lbaas | 08:50 | |
| *** voelzmo has joined #openstack-lbaas | 08:53 | |
| *** aojea has quit IRC | 08:54 | |
| *** JudeC has quit IRC | 09:11 | |
| *** voelzmo has quit IRC | 09:30 | |
| *** voelzmo has joined #openstack-lbaas | 09:31 | |
| *** aojea has joined #openstack-lbaas | 09:32 | |
| *** voelzmo has quit IRC | 09:45 | |
| *** chlong has joined #openstack-lbaas | 11:07 | |
| *** gcheresh has quit IRC | 11:36 | |
| *** gcheresh has joined #openstack-lbaas | 12:11 | |
| *** amotoki has joined #openstack-lbaas | 12:27 | |
| *** links has quit IRC | 13:29 | |
| *** cpuga has joined #openstack-lbaas | 13:39 | |
| *** cpuga_ has joined #openstack-lbaas | 13:41 | |
| rm_work | xgerman: when you get on, this please: https://review.openstack.org/#/c/459654/ :) | 13:43 |
| *** cpuga_ has quit IRC | 13:43 | |
| *** cpuga_ has joined #openstack-lbaas | 13:44 | |
| *** cpuga has quit IRC | 13:44 | |
| xgerman | k | 13:45 |
| rm_work | ah you are already on :P | 13:46 |
| nmagnezi | Three +2, must be my lucky day :D | 13:46 |
| rm_work | and I think https://review.openstack.org/#/c/431324/ is ready as well | 13:49 |
| rm_work | I added the test for johnsom | 13:49 |
| rm_work | xgerman: also I want to know what you think on a couple patches | 13:49 |
| xgerman | sure | 13:49 |
| rm_work | 1) cascade delete, implemented slightly differently: https://review.openstack.org/#/c/461182/ | 13:49 |
| *** gcheresh has quit IRC | 13:50 | |
| rm_work | 2) project_id on responses: https://review.openstack.org/#/c/461344/ | 13:50 |
| rm_work | (I realized I THINK it is a little weird to remove it even from responses, though definitely we should not store it on every object) | 13:50 |
| rm_work | ah, also single create: https://review.openstack.org/459032 | 13:51 |
| rm_work | I will try to poke at status-tree today | 13:51 |
| rm_work | but right now I need to sleep for a few hours >_> | 13:51 |
| johnsom | Thanks Adam | 14:01 |
| xgerman | +1 | 14:03 |
| xgerman | johnsom you have a decent single create set-up? | 14:04 |
| xgerman | or should I do some in-depth testing? | 14:04 |
| johnsom | There is an example in the patch docs. But I had problems with the patch last week. | 14:05 |
| xgerman | ok, I guess then I need to stack it — | 14:07 |
| *** gcheresh has joined #openstack-lbaas | 14:12 | |
| *** krypto has joined #openstack-lbaas | 14:25 | |
| *** krypto has joined #openstack-lbaas | 14:25 | |
| *** other_krypt0 has joined #openstack-lbaas | 14:28 | |
| *** krypto has quit IRC | 14:29 | |
| *** catintheroof has joined #openstack-lbaas | 14:30 | |
| *** krypto has joined #openstack-lbaas | 14:35 | |
| *** other_krypt0 has quit IRC | 14:35 | |
| *** catintheroof has quit IRC | 14:36 | |
| *** catintheroof has joined #openstack-lbaas | 14:40 | |
| *** other_Krypt0 has joined #openstack-lbaas | 14:40 | |
| *** krypto has quit IRC | 14:41 | |
| *** catintheroof has quit IRC | 14:42 | |
| *** other_Krypt0 is now known as krypto | 14:43 | |
| *** krypto is now known as Guest60753 | 14:43 | |
| *** fnaval has quit IRC | 14:46 | |
| openstackgerrit | Merged openstack/octavia master: Change auth_strategy default to keystone https://review.openstack.org/459654 | 14:48 |
| *** gcheresh has quit IRC | 14:57 | |
| *** fnaval has joined #openstack-lbaas | 15:00 | |
| xgerman | mmh, our single create example is bogus | 15:04 |
| openstackgerrit | German Eichberger proposed openstack/octavia master: Fixes small issues https://review.openstack.org/461425 | 15:08 |
| *** aojea has quit IRC | 15:11 | |
| *** links has joined #openstack-lbaas | 15:13 | |
| *** links has quit IRC | 15:17 | |
| xgerman | ok, and when I send it against Adam’s patch it causes an error (might be me but we really should in that case review our validation logic) | 15:17 |
| *** gcheresh has joined #openstack-lbaas | 15:18 | |
| *** gcheresh has quit IRC | 15:19 | |
| *** voelzmo has joined #openstack-lbaas | 15:33 | |
| *** amotoki has quit IRC | 15:55 | |
| *** aojea has joined #openstack-lbaas | 16:11 | |
| *** aojea has quit IRC | 16:16 | |
| *** tesseract has quit IRC | 16:29 | |
| *** openstack has joined #openstack-lbaas | 17:02 | |
| *** adams.freenode.net sets mode: +ns | 17:02 | |
| *** openstack has joined #openstack-lbaas | 17:25 | |
| *** bzhao has joined #openstack-lbaas | 17:41 | |
| *** voelzmo has joined #openstack-lbaas | 18:00 | |
| *** voelzmo has quit IRC | 18:01 | |
| *** voelzmo has joined #openstack-lbaas | 18:02 | |
| *** Guest60753 has quit IRC | 18:08 | |
| xgerman | afk for a bit | 18:10 |
| *** voelzmo has quit IRC | 18:12 | |
| *** aojea has joined #openstack-lbaas | 18:13 | |
| *** openstackstatus has joined #openstack-lbaas | 18:14 | |
| *** ChanServ sets mode: +v openstackstatus | 18:14 | |
| *** aojea has quit IRC | 18:18 | |
| *** blogan_ has joined #openstack-lbaas | 18:37 | |
| *** blogan has quit IRC | 18:39 | |
| *** JudeC has joined #openstack-lbaas | 18:45 | |
| johnsom | Oh! found my problem with devstack. We introduced a bug recently with the new threading stuffs. | 18:54 |
| johnsom | ./clean.sh isn't shutting down the octavia-worker | 18:55 |
| johnsom | https://www.irccloud.com/pastebin/GjKrW27s/ | 18:55 |
| johnsom | Sounds like an afternoon project | 18:55 |
| johnsom | I'm going to get lunch, back in a few | 18:56 |
| *** voelzmo has joined #openstack-lbaas | 19:01 | |
| *** aojea has joined #openstack-lbaas | 19:04 | |
| *** dougwig has joined #openstack-lbaas | 19:15 | |
| xgerman | oh, yeah, noticed that as well - thought it was my un-even devstack | 19:16 |
| *** blogan_ is now known as blogan | 19:26 | |
| *** gcheresh has joined #openstack-lbaas | 19:41 | |
| *** KeithMnemonic has joined #openstack-lbaas | 19:41 | |
| johnsom | Nope, we broke something... | 19:54 |
| *** JudeC has quit IRC | 20:00 | |
| *** KeithMnemonic has quit IRC | 20:09 | |
| *** kishore448 has joined #openstack-lbaas | 20:12 | |
| johnsom | Yeah, neat. kill -15 has no effect.... | 20:18 |
| *** blogan has quit IRC | 20:22 | |
| xgerman | kill -9 worked for me | 20:24 |
| kishore448 | hi all, need assistance in installing octavia on ubunut16.04 | 20:25 |
| johnsom | xgerman Yes, but.... | 20:25 |
| johnsom | kishore448 Hi, what issue are you seeing? | 20:26 |
| *** voelzmo has quit IRC | 20:27 | |
| kishore448 | I installed octavia using "pip install octavia" .. all the octavia-*.py files in /usr/local/bin .. i also performed all the pre-req steps like creating lb-mgmt-net, aphora image, nova flavor etc... | 20:27 |
| johnsom | Ok, sounds good so far | 20:28 |
| kishore448 | after creating octavia.conf file with appropriate parameters, I am not sure how to start the octavia services | 20:28 |
| *** cpuga_ has quit IRC | 20:28 | |
| *** blogan has joined #openstack-lbaas | 20:28 | |
| johnsom | kishore448 We do this in dev environments: | 20:30 |
| johnsom | /usr/local/bin/octavia-api --config-file /etc/octavia/octavia.conf & echo $! >/opt/stack/status/stack/o-api.pid; fg | 20:30 |
| johnsom | /usr/local/bin/octavia-worker --config-file /etc/octavia/octavia.conf & echo $! >/opt/stack/status/stack/o-cw.pid; fg | 20:30 |
| johnsom | ... | 20:30 |
| johnsom | All four processes work the same | 20:31 |
| kishore448 | Are there any written start up scripts for these services or I need to do the same...BTW we are trying setup in production ( replace existing haproxy plugin and use octavia) | 20:33 |
| johnsom | Ah, sorry no. We haven't developed any startup scripts for the processes yet. So much to do, not enough time in a day | 20:34 |
| xgerman | you can always look at waht we did for openstack ansible | 20:36 |
| kishore448 | Ah ok..I will try use your commands...Thanks a lot Johnsom ... Thanks xgerman | 20:38 |
| *** gcheresh has quit IRC | 20:54 | |
| *** cpuga has joined #openstack-lbaas | 20:57 | |
| *** JudeC has joined #openstack-lbaas | 20:58 | |
| *** JudeC has quit IRC | 21:00 | |
| *** sanfern has joined #openstack-lbaas | 21:16 | |
| *** aojea has quit IRC | 21:18 | |
| *** sanfern has quit IRC | 21:29 | |
| *** sanfern has joined #openstack-lbaas | 21:30 | |
| *** catintheroof has joined #openstack-lbaas | 21:50 | |
| *** KeithMnemonic has joined #openstack-lbaas | 21:52 | |
| *** JudeC has joined #openstack-lbaas | 21:59 | |
| rm_work | xgerman: i fixed the api-ref *in* that patch :) | 22:02 |
| rm_work | hmmm interesting error | 22:03 |
| xgerman | yep, noticed | 22:04 |
| JudeC | johnsom: I got the searching for subnet by name figured out, should I do the same for network? | 22:06 |
| *** kishore448 has quit IRC | 22:09 | |
| *** KeithMnemonic has quit IRC | 22:10 | |
| rm_work | taking a look at your example now | 22:10 |
| rm_work | trying to reproduce | 22:10 |
| rm_work | but gotta restack i think :/ | 22:10 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Add support for single-create for APIv2 https://review.openstack.org/459032 | 22:11 |
| rm_work | ^^rebaswe | 22:11 |
| johnsom | JudeC Yes | 22:12 |
| JudeC | figured :) | 22:12 |
| JudeC | ty | 22:12 |
| johnsom | As for stacking, if you reclone nova dumped n-crt (nova-cert) so pull it from localrc or stack bombs.... | 22:12 |
| rm_work | johnsom: does harlowja's patch fix your thing? https://review.openstack.org/#/c/460823/ | 22:13 |
| johnsom | I have been feeling the devstack love this afternoon... Sigh | 22:13 |
| rm_work | i've never tried restacking | 22:13 |
| rm_work | so | 22:13 |
| rm_work | or rather | 22:13 |
| rm_work | i never use clean.sh | 22:13 |
| rm_work | i just blow the VM away every time | 22:13 |
| johnsom | Darn, forgot about that patch | 22:14 |
| johnsom | It might | 22:14 |
| johnsom | JudeC the bugs should say "name or ID" on the params that need to be looked up | 22:27 |
| JudeC | kk | 22:35 |
| *** catintheroof has quit IRC | 22:39 | |
| *** JudeC has quit IRC | 22:42 | |
| *** JudeC has joined #openstack-lbaas | 23:07 | |
| *** fnaval has quit IRC | 23:09 | |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 23:25 |
| rm_work | augh | 23:30 |
| rm_work | something else in the gate broke now | 23:30 |
| * rm_work sighs | 23:30 | |
| rm_work | http://logs.openstack.org/24/431324/4/check/gate-octavia-v1-dsvm-scenario-ubuntu-xenial/c365b5d/logs/devstacklog.txt.gz#_2017-05-01_22_48_05_593 | 23:30 |
| rm_work | Barbican not starting | 23:30 |
| johnsom | Hmmmm systemd.... | 23:31 |
| johnsom | I bet they switched over to systemd default | 23:35 |
| johnsom | I'm not sure if those "newfangled" systemd logs are even captured anywhere.... | 23:36 |
| johnsom | It is this line: http://logs.openstack.org/24/431324/4/check/gate-octavia-v1-dsvm-scenario-ubuntu-xenial/c365b5d/logs/devstacklog.txt.gz#_2017-05-01_22_48_05_271 | 23:37 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 23:37 |
| johnsom | systemd requires the full path, no idea why barbican is different than the others though | 23:37 |
| openstackgerrit | Joshua Harlow proposed openstack/octavia master: Shutdown a endpoint workers executor on termination https://review.openstack.org/460823 | 23:39 |
| openstackgerrit | Jude Cross proposed openstack/python-octaviaclient master: Add loadbalancer commands to client https://review.openstack.org/454516 | 23:51 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!