| *** Swami has quit IRC | 00:04 | |
| abaindur | johnsom: or rm_work: is there a way to scale up/down a loadbalancer? | 00:05 |
|---|---|---|
| abaindur | https://developer.openstack.org/api-ref/load-balancer/v2/#update-a-load-balancer | 00:05 |
| rm_work | Not yet | 00:05 |
| abaindur | I dont see a flavor ID parameter for update | 00:05 |
| abaindur | but i only see it for create | 00:06 |
| johnsom | Not at this point. | 00:06 |
| johnsom | Do you need to? | 00:06 |
| abaindur | No, we were just wondering, a nice to have | 00:06 |
| abaindur | so i guess right now only solution is to delete it, then create a new one? | 00:06 |
| johnsom | Yeah, with flavors we will allow different sized amphroa and likely enable threading. Beyond that is the ActiveActive work | 00:06 |
| abaindur | meaning we would also need to delete the members, pools, listeners... and recreate all those resources again from scartch | 00:07 |
| johnsom | Correct. or change the flavor setting in the cfg and do a failover. | 00:07 |
| abaindur | the flavor in the cfg is the default one, if flavor_id not specified in the create POST call, right? | 00:08 |
| abaindur | otherwise it creates using flavor in the request | 00:08 |
| abaindur | https://developer.openstack.org/api-ref/load-balancer/v2/#create-a-load-balancer | 00:08 |
| johnsom | Technically you can do a nova resize, but it will reboot the instance, which will delete the TLS certs. | 00:08 |
| johnsom | No, the flavor_id in the API is for Octavia flavors, not nova flavors. Octavia flavors aren't fully implemented yet. | 00:09 |
| johnsom | I am planning to finish that for Stein | 00:09 |
| colin- | was setting up some benchmark tests for my amusement and ran iperf on the mgmt NIC of my amphora, was surprised to get 2x the throughput i did to a similar backend when using the VIP through the same amphora. does that make sense and is netns expected to suffer to a large degree? | 00:09 |
| abaindur | not sure I follow then. if its defined in the cfg file, what is purpose of the flavor_id in POST call? | 00:09 |
| johnsom | No. I have got 14Gbps with iperf3 thorugh our amp | 00:10 |
| colin- | ok | 00:10 |
| abaindur | amp_flavor_id is "Nova instance flavor id for the Amphora"... so what is the purpose of the flavor_id in current create call? | 00:10 |
| johnsom | abaindur the flavor in the config file is the nova flavor. The flavor in the Octavia API is for Octavia flavors (not nova) | 00:10 |
| colin- | and you expect the performance characteristics to be similar for traffic using the namespaced NIC as it would be for traffic on the native one? | 00:11 |
| abaindur | What is an Octavia flavor then :) | 00:11 |
| abaindur | ah i see.. you said its not implemented yet, in stein | 00:12 |
| johnsom | Octavia flavors will implement this spec: https://docs.openstack.org/octavia/latest/contributor/specs/version1.0/flavors.html | 00:12 |
| abaindur | so currently, even as of rocky, the flavor_id in create is basically ignored | 00:12 |
| johnsom | colin- Well, native will always be faster, but the way we use the namespace it doesn't have much performance penalty. We don't use bridges and veth pairs and all that muck that takes the perf down. | 00:13 |
| johnsom | abaindur Correct | 00:13 |
| johnsom | abaindur We lost a lot of developers off the project all of a sudden and it just didn't get finished per the original plan. | 00:14 |
| johnsom | Well, OpenStack lost a lot of devs, and our project was one that suffered. | 00:14 |
| johnsom | colin- http://paste.openstack.org/show/731546/ | 00:16 |
| johnsom | 1 vCPU, 1GB RAM, Rocky codebase | 00:17 |
| johnsom | Of course a million variables in the clouds will change those numbers, so your mileage will vary. | 00:17 |
| colin- | yeah | 00:20 |
| colin- | what is your line speed? | 00:20 |
| colin- | i'm not approaching even 50% line speed on my tests run similarly so definitely some key differences | 00:20 |
| johnsom | The bypass was 15.4 Gbits/sec | 00:21 |
| johnsom | colin- Oh, my bad, it had this patch in it, so I guess it was a Stein amp image: https://review.openstack.org/#/c/598379/ | 00:24 |
| johnsom | I did those changes internal before I upstreamed them. | 00:24 |
| colin- | understood, thanks for the additional details. between two nova vms in the same network i can get almost double what i can through an octavia vip so i must have something that is off | 00:25 |
| colin- | i'll take a look at those changes you linked specifically about tcp kernel splicing maybe they could hepl | 00:25 |
| johnsom | Yeah, the splicing is helpful to bring the CPU load down. Also, make sure your haproxy package from your OS vendor isn't forcing it off on the command line for the systemd wrapper. I saw that happening somewhere too. | 00:26 |
| *** rcernin has quit IRC | 00:27 | |
| johnsom | Ok, off to dinner. catch you all tomorrow | 00:28 |
| *** rcernin has joined #openstack-lbaas | 00:29 | |
| colin- | ttyl, enjoy | 00:30 |
| *** dlundquist has joined #openstack-lbaas | 00:38 | |
| *** dlundquist has quit IRC | 00:43 | |
| *** dlundquist has joined #openstack-lbaas | 00:44 | |
| *** hongbin has joined #openstack-lbaas | 01:57 | |
| *** JudeCross has quit IRC | 01:57 | |
| *** ramishra has joined #openstack-lbaas | 02:09 | |
| *** JudeCross has joined #openstack-lbaas | 02:17 | |
| *** yamamoto has quit IRC | 02:23 | |
| *** yamamoto has joined #openstack-lbaas | 02:23 | |
| *** abaindur has quit IRC | 02:23 | |
| *** dosaboy has quit IRC | 02:35 | |
| *** annp has joined #openstack-lbaas | 03:01 | |
| *** JudeCross has quit IRC | 03:04 | |
| *** JudeCross has joined #openstack-lbaas | 03:05 | |
| *** sapd1 has quit IRC | 03:10 | |
| *** sapd1 has joined #openstack-lbaas | 03:10 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix IPv6 in Active/Standby topology https://review.openstack.org/589292 | 03:27 |
| *** hongbin has quit IRC | 03:42 | |
| *** rcernin has quit IRC | 04:42 | |
| *** rcernin has joined #openstack-lbaas | 04:46 | |
| *** openstackgerrit has quit IRC | 04:52 | |
| *** openstackgerrit has joined #openstack-lbaas | 04:57 | |
| *** aojea has joined #openstack-lbaas | 05:34 | |
| *** celebdor has joined #openstack-lbaas | 05:39 | |
| *** aojea has quit IRC | 05:51 | |
| *** Emine has quit IRC | 05:56 | |
| *** phuoc has joined #openstack-lbaas | 06:33 | |
| *** phuoc_ has quit IRC | 06:36 | |
| *** ccamposr has joined #openstack-lbaas | 06:43 | |
| *** aojea has joined #openstack-lbaas | 06:43 | |
| *** JudeCross has quit IRC | 06:49 | |
| *** pcaruana has joined #openstack-lbaas | 06:57 | |
| *** JudeCross has joined #openstack-lbaas | 06:58 | |
| *** rcernin has quit IRC | 07:04 | |
| *** maciejjozefczyk has joined #openstack-lbaas | 07:05 | |
| *** rtjure has quit IRC | 07:09 | |
| *** rtjure has joined #openstack-lbaas | 07:12 | |
| *** abaindur has joined #openstack-lbaas | 07:14 | |
| *** dosaboy has joined #openstack-lbaas | 07:38 | |
| *** abaindur has quit IRC | 07:47 | |
| *** velizarx has joined #openstack-lbaas | 07:48 | |
| maciejjozefczyk | hello! | 08:03 |
| maciejjozefczyk | My amphorea which is build on Xenial is playing with me, sometimes VIP interface is renamed from ethX to ensX and then renamed back. After some seconds its missing in the system. | 08:04 |
| maciejjozefczyk | After a big digging it said on the net to try Bionic | 08:04 |
| maciejjozefczyk | So I tried amphore'a with Bionic and while attaching VIP interface to amp on amphore-agent logs I see | 08:05 |
| maciejjozefczyk | [2018-10-05 07:57:46 +0000] [970] [DEBUG] POST /0.5/plug/vip/192.168.0.51 | 08:05 |
| maciejjozefczyk | ::ffff:10.10.10.54 - - [05/Oct/2018:07:57:46 +0000] "POST /0.5/plug/vip/192.168.0.51 HTTP/1.1" 500 123 "-" "Octavia HaProxy Rest Client/0.5 (https://wiki.openstack.org/wiki/Octavia)" | 08:05 |
| maciejjozefczyk | How to debug it? | 08:05 |
| maciejjozefczyk | In conf i have defiend the debug=True | 08:05 |
| maciejjozefczyk | I need to do manual CURL with certificate validation to check that/ | 08:06 |
| maciejjozefczyk | ok… got it, syslog is the key | 08:07 |
| maciejjozefczyk | sorry for spamming ;) | 08:07 |
| *** salmankhan has joined #openstack-lbaas | 08:08 | |
| *** velizarx has quit IRC | 08:11 | |
| *** velizarx has joined #openstack-lbaas | 08:12 | |
| *** JudeCross has quit IRC | 08:26 | |
| *** JudeCross has joined #openstack-lbaas | 08:43 | |
| *** aojea has quit IRC | 08:44 | |
| *** JudeCross has quit IRC | 08:47 | |
| *** aojea has joined #openstack-lbaas | 08:50 | |
| rm_work | maciejjozefczyk: i am not sure how much Bionic is actually tested/functional yet for amps | 08:53 |
| rm_work | maciejjozefczyk: but it's interesting you're suffering from the net naming issue, as we specifically turn that off | 08:53 |
| maciejjozefczyk | rm_work: my fault | 08:53 |
| maciejjozefczyk | rm_work: I need a duck, basically :) | 08:54 |
| rm_work | did you get it working with xenial? | 08:55 |
| maciejjozefczyk | yep | 08:57 |
| maciejjozefczyk | im just curious, because I'm testing Active-Standby with my network architecture | 08:58 |
| maciejjozefczyk | and I see there is no ability to set "HA mode" while creating LB? | 08:58 |
| maciejjozefczyk | the only way is config parameter loadbalancer_topology? | 08:58 |
| rm_work | are you using my L3 network patch? | 08:59 |
| rm_work | yes | 08:59 |
| rm_work | we would need flavor support | 08:59 |
| rm_work | to allow for configurable topology | 08:59 |
| maciejjozefczyk | rm_work: unfortunately im not using the L3 network driver | 08:59 |
| rm_work | ah | 08:59 |
| maciejjozefczyk | due I was not able to select other subnet for VIP and VRRP | 08:59 |
| maciejjozefczyk | ok, but anyway about this patch | 09:00 |
| rm_work | i'm still confused about that | 09:01 |
| rm_work | why would you need to select a subnet? | 09:01 |
| rm_work | to make a FLIP, you only select a network? | 09:01 |
| rm_work | I thought | 09:01 |
| maciejjozefczyk | I have usecase when client has his own private tenant network, lets name it 'private' | 09:01 |
| maciejjozefczyk | and under this 'private' network client has some backends | 09:01 |
| maciejjozefczyk | I would like to have VRRP working inside this private network | 09:02 |
| rm_work | hmm | 09:02 |
| maciejjozefczyk | and on VIP port I would like to create a FLIP, which is from other network (External network) | 09:02 |
| rm_work | yes, that is how it works now, I believe | 09:03 |
| rm_work | though actually i can't remember how i decide which network to put the VRRP ports on | 09:03 |
| maciejjozefczyk | basically its similiar to usecase given here: https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/13/html/networking_guide/sec-octavia | 09:03 |
| rm_work | since it can't be the FLIP network | 09:03 |
| rm_work | ah yeah probably my driver just selects the management network for the VRRP network | 09:05 |
| rm_work | because I also assumed "no tenant networks" in the design | 09:05 |
| maciejjozefczyk | yep, thats right | 09:05 |
| rm_work | ok, i'll think about that | 09:05 |
| maciejjozefczyk | so basically vrrp port, VIP port and FIP are from the same subnet | 09:05 |
| rm_work | not sure if there's an easy way to allow that | 09:05 |
| rm_work | err, that can't be true i think | 09:06 |
| rm_work | but i'll check | 09:06 |
| maciejjozefczyk | please check :) | 09:06 |
| rm_work | you can always submit a patch after mine in the chain, and change what you need :P | 09:07 |
| maciejjozefczyk | for now I tested the allowed_access_pairs driver under my tenant network with floating IP that points on VIP port, and it worked perfecto! | 09:07 |
| rm_work | ah, cool | 09:07 |
| rm_work | i wonder if it will work on failover though | 09:07 |
| maciejjozefczyk | yea, sure, rm_work we're like touching the topic now, like a little research | 09:07 |
| maciejjozefczyk | rm_work: im checking it | 09:07 |
| rm_work | if you don't have the same L2, it shouldn't | 09:07 |
| maciejjozefczyk | router and FIP along with VRRP works in the same L2 | 09:08 |
| maciejjozefczyk | so the only thing is that keepalived needs to move the alias on VRRP interface from one amphorea to another | 09:08 |
| maciejjozefczyk | it must work ;) | 09:08 |
| maciejjozefczyk | anyway, we're going in OVH do deploy octavia soon | 09:09 |
| rm_work | ah yeah if you have guaranteed L2 adjacency, you're good | 09:09 |
| maciejjozefczyk | we have dedicated team for service similiar to Octavia and they will adopt Octavia to our use-case | 09:09 |
| maciejjozefczyk | so for sure we gonna upstream some patches | 09:09 |
| rm_work | cool | 09:10 |
| maciejjozefczyk | for nowe we have like 2 or 3 cases that needs to be solved, but little ones | 09:10 |
| maciejjozefczyk | for now* | 09:10 |
| *** salmankhan has quit IRC | 09:25 | |
| *** Emine has joined #openstack-lbaas | 09:32 | |
| *** salmankhan has joined #openstack-lbaas | 09:32 | |
| *** JudeCross has joined #openstack-lbaas | 09:38 | |
| *** JudeCross has quit IRC | 09:43 | |
| maciejjozefczyk | rm_work: HA works perfect! | 09:51 |
| maciejjozefczyk | <1 sec of downtime | 09:51 |
| *** yamamoto has quit IRC | 10:07 | |
| *** yamamoto has joined #openstack-lbaas | 10:08 | |
| *** yamamoto has quit IRC | 10:09 | |
| *** yamamoto has joined #openstack-lbaas | 10:10 | |
| *** yamamoto has quit IRC | 10:13 | |
| *** yamamoto has joined #openstack-lbaas | 10:14 | |
| *** yamamoto has quit IRC | 10:16 | |
| *** yamamoto has joined #openstack-lbaas | 10:17 | |
| *** Emine has quit IRC | 10:17 | |
| *** yamamoto has quit IRC | 10:21 | |
| *** yamamoto has joined #openstack-lbaas | 10:27 | |
| *** yamamoto has quit IRC | 10:32 | |
| *** yamamoto has joined #openstack-lbaas | 10:33 | |
| *** yamamoto has quit IRC | 10:38 | |
| *** annp has quit IRC | 10:53 | |
| *** yamamoto has joined #openstack-lbaas | 11:08 | |
| *** rpittau_ has quit IRC | 11:17 | |
| *** yamamoto has quit IRC | 11:18 | |
| *** yamamoto has joined #openstack-lbaas | 11:19 | |
| *** yamamoto has quit IRC | 11:23 | |
| *** yamamoto has joined #openstack-lbaas | 11:24 | |
| *** salmankhan has quit IRC | 11:29 | |
| *** JudeCross has joined #openstack-lbaas | 11:39 | |
| *** JudeCross has quit IRC | 11:44 | |
| *** JudeCross has joined #openstack-lbaas | 12:23 | |
| *** JudeCross has quit IRC | 12:28 | |
| *** aojea has quit IRC | 12:32 | |
| *** salmankhan has joined #openstack-lbaas | 12:33 | |
| *** dosaboy has quit IRC | 12:35 | |
| *** dosaboy has joined #openstack-lbaas | 12:36 | |
| *** dosaboy has quit IRC | 12:36 | |
| *** aojea has joined #openstack-lbaas | 12:36 | |
| *** aojea has quit IRC | 12:41 | |
| *** dosaboy has joined #openstack-lbaas | 12:41 | |
| *** dosaboy has left #openstack-lbaas | 12:42 | |
| *** ccamposr has quit IRC | 12:52 | |
| *** dosaboy has joined #openstack-lbaas | 12:56 | |
| *** dims_ has quit IRC | 12:58 | |
| *** celebdor has quit IRC | 13:02 | |
| *** celebdor has joined #openstack-lbaas | 13:16 | |
| *** ramishra has quit IRC | 13:28 | |
| *** velizarx has quit IRC | 13:46 | |
| *** velizarx has joined #openstack-lbaas | 13:50 | |
| *** celebdor has quit IRC | 14:14 | |
| *** fnordahl has joined #openstack-lbaas | 14:16 | |
| *** JudeCross has joined #openstack-lbaas | 14:24 | |
| *** JudeCross has quit IRC | 14:28 | |
| *** numans has quit IRC | 14:47 | |
| *** sabomia has joined #openstack-lbaas | 14:54 | |
| johnsom | FYI, the charms folks are starting work on an Octavia charm: https://review.openstack.org/608283 | 15:03 |
| johnsom | Bummer, so the Bionic kernel does not solve the KVM issues we are seeing.... | 15:15 |
| johnsom | http://logs.openstack.org/39/600539/18/check/octavia-v2-dsvm-scenario/6c294a7/controller/logs/libvirt/qemu/instance-0000000a_log.txt.gz | 15:15 |
| *** Emine has joined #openstack-lbaas | 15:23 | |
| *** pcaruana has quit IRC | 15:39 | |
| cgoncalves | shall we switch OS to centos? ;-) | 15:48 |
| *** velizarx has quit IRC | 15:50 | |
| *** salmankhan has quit IRC | 16:04 | |
| *** ianychoi_ is now known as ianychoi | 16:19 | |
| *** sabomia has quit IRC | 16:27 | |
| *** celebdor has joined #openstack-lbaas | 16:34 | |
| *** pcaruana has joined #openstack-lbaas | 16:42 | |
| johnsom | I am open to the possibility if it fixes our issue..... How is that test patch going? | 16:45 |
| *** dims has joined #openstack-lbaas | 16:46 | |
| openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Enabling SG customization on loadbalancer listerners https://review.openstack.org/602564 | 16:49 |
| *** salmankhan has joined #openstack-lbaas | 16:49 | |
| *** pcaruana has quit IRC | 16:50 | |
| openstackgerrit | Luis Tomas Bolivar proposed openstack/octavia master: Enabling SG customization on loadbalancer listerners https://review.openstack.org/602564 | 16:54 |
| ltomasbo | johnsom, cgoncalves ^ I updated it with your suggestions | 16:55 |
| johnsom | Cool, so you figured out a work around? Sorry, I got pulled into an internal exercise so, haven't followed closely | 16:56 |
| ltomasbo | johnsom, me? I just added the option to have it disable by default, as well as a whitelisting option (cgoncalves' suggestion) to even if it is active, only enable it for certain projects | 16:58 |
| johnsom | hmmm, not sure I am a fan of the whitelisting | 16:58 |
| johnsom | I will have a look though. Thanks for the patience on this, it's a tricky one | 17:00 |
| *** JudeCross has joined #openstack-lbaas | 17:00 | |
| ltomasbo | johnsom, great! thanks! and don't worry, I understand it is not the right thing to do | 17:01 |
| ltomasbo | johnsom, it is a pity that changing the SG on the VIP was not working, that would have been a better compromise | 17:02 |
| johnsom | Agreed | 17:02 |
| johnsom | Or even a way to "AND" SGs so that we could own one and the tenant could own the other.... | 17:03 |
| ltomasbo | thanks for the feedback (and for the time spent on the weekly meetings on this) | 17:03 |
| *** yamamoto has quit IRC | 17:08 | |
| *** salmankhan has quit IRC | 17:09 | |
| *** yamamoto has joined #openstack-lbaas | 17:09 | |
| *** yamamoto has quit IRC | 17:14 | |
| xgerman_ | I am spending my morning with LB in Error… OSP shutting down rabbit with each upgrade is really a bummer… | 17:24 |
| johnsom | That is fixed with the HM perf patch | 17:24 |
| xgerman_ | ok, well, they won’t backport to OSP 12 - so there’s that | 17:25 |
| johnsom | Sigh | 17:25 |
| *** celebdor has quit IRC | 17:38 | |
| *** salmankhan has joined #openstack-lbaas | 17:53 | |
| *** salmankhan has quit IRC | 17:55 | |
| *** logan- has left #openstack-lbaas | 17:55 | |
| *** salmankhan has joined #openstack-lbaas | 17:55 | |
| *** yamamoto has joined #openstack-lbaas | 17:59 | |
| *** logan- has joined #openstack-lbaas | 18:08 | |
| *** Swami has joined #openstack-lbaas | 18:36 | |
| *** sabomia` has joined #openstack-lbaas | 19:37 | |
| sabomia` | hey guys, what could explain for an otherwise functional load balancer to remain in operational status offline? | 19:40 |
| johnsom | Which object is marked operating status offline, the LB itself? | 19:45 |
| sabomia` | yes the LB itself and the listener | 19:45 |
| sabomia` | the pool is online | 19:45 |
| sabomia` | and the LB seems to function as expected | 19:45 |
| sabomia` | running OSA Queens 17.0.7 | 19:45 |
| johnsom | Check that your health manager process is actually getting the health heartbeats. | 19:46 |
| johnsom | for example, you could look in the octavia database a query the amphora_health table; | 19:46 |
| johnsom | https://www.irccloud.com/pastebin/3ytFpIiu/ | 19:46 |
| johnsom | If it is empty or very old dates, you may have a problem with the heartbeats getting back to the controller(s). | 19:47 |
| *** Emine has quit IRC | 19:47 | |
| sabomia` | http://paste.openstack.org/show/731600/ | 19:50 |
| sabomia` | this eems unrelated to the current lb though | 19:50 |
| sabomia` | I had a networking issue on it before, so that may explain it. I got that sorted out and somewhat successfully deployed the first VM | 19:50 |
| sabomia` | *LB | 19:50 |
| johnsom | Yeah, ok, so the health heartbeats are not making it back. Things to check: | 19:52 |
| johnsom | https://docs.openstack.org/octavia/latest/configuration/configref.html#health_manager.controller_ip_port_list | 19:52 |
| johnsom | Make sure the controller_ip_port_list is accurate to your health manager processes. | 19:53 |
| johnsom | Example: controller_ip_port_list example: 10.0.0.1:5555, 10.0.1.2:5555 | 19:53 |
| johnsom | Make sure your amps can send UDP back on port 5555 to the controllers. | 19:54 |
| sabomia` | thanks johnsom, I'll look at that in a sec. | 20:09 |
| *** celebdor has joined #openstack-lbaas | 20:33 | |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix IPv6 in Active/Standby topology https://review.openstack.org/589292 | 20:33 |
| *** salmankhan has quit IRC | 20:49 | |
| *** rtjure has quit IRC | 21:07 | |
| johnsom | sabomia` They can have IPs from a different subnet, but that is unusual. Typically the lb-mgmt-net will have the same addresses both on the controller-ip list and the amphora-agent interface | 22:04 |
| colin- | johnsom: any suggestions for where i can search for evidence of what's throttling my throughput wrt what we discussed yesterday? | 22:06 |
| colin- | resource utilization on all concerned hosts is healthy throughout and i don't see anything obvious in journal | 22:06 |
| colin- | cherry picking the changes you linked yesterday for testing | 22:07 |
| johnsom | colin- I would start by eliminating the parts between point A and B. So, co-locate your source instance, amp, and target on the same host. This removes the NICs, etc. from the picture. | 22:07 |
| colin- | ah, good idea. thanks | 22:07 |
| colin- | that will vastly simplify things | 22:07 |
| johnsom | Checking the splicing stuff I mentioned yesterday are good. Then I would work back, like adding OVS, NICs, etc. | 22:08 |
| sabomia` | johnsom: that may be my issue here? I couldn't get it to work though with the amphoras and management ips on the same bridge. Once a amphora VM would get deployed I would lose connectivity on my bridge. I followed xgerman_ install instructions using separate subnets from https://github.com/xgerman/rpc-octavia/blob/master/INSTALLATION.md | 22:11 |
| sabomia` | Following his instructions got my load balancers to work, I am just still not getting any health monitoring msgs | 22:12 |
| *** aojea has joined #openstack-lbaas | 22:12 | |
| johnsom | sabomia` Those instructions have one subnet for the lb-mgmt-net | 22:12 |
| sabomia` | they refer to 172.29.248.0/22 and br-lbaas in the provider network and 10.0.252.0/22 + lbaas-mgt in the octavia management network | 22:14 |
| johnsom | I think that is a mistake | 22:21 |
| *** aojea has quit IRC | 22:45 | |
| openstackgerrit | Merged openstack/octavia stable/queens: Add note to lower constraints for Jinja and pyOpenSSL https://review.openstack.org/606142 | 22:47 |
| *** celebdor has quit IRC | 22:48 | |
| *** Swami has quit IRC | 23:04 | |
| *** aojea has joined #openstack-lbaas | 23:17 | |
| openstackgerrit | Merged openstack/octavia master: HTTPS HMs need the same validation path as HTTP++ https://review.openstack.org/608099 | 23:20 |
| *** aojea has quit IRC | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!