| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add OpenConfig classes for LACP https://review.opendev.org/c/openstack/networking-baremetal/+/835161 | 03:12 |
|---|---|---|
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Device management driver iface https://review.opendev.org/c/openstack/networking-baremetal/+/835170 | 03:13 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Device management driver iface https://review.opendev.org/c/openstack/networking-baremetal/+/835170 | 03:52 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Device management driver iface https://review.opendev.org/c/openstack/networking-baremetal/+/835170 | 04:01 |
| rpittau | good morning ironic! Happy Friday! o/ | 05:52 |
| arne_wiebalck | Good morning rpittau and Ironic! | 07:07 |
| rpittau | hey arne_wiebalck :) | 07:07 |
| janders | good morning rpittau arne_wiebalck and Ironic o/ | 07:34 |
| janders | Happy Friday | 07:34 |
| rpittau | hey janders :) | 07:34 |
| arne_wiebalck | hey janders o/ | 07:36 |
| rpittau | TheJulia: I've updated the ironic release in the ironic-website https://github.com/OpenStackweb/ironic-website/pull/48 | 07:42 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove get_remote_client & check_vm_connectivity methods https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834124 | 08:40 |
| dtantsur | morning folks! finally Friday! | 08:58 |
| rpittau | good morning dtantsur :) | 08:59 |
| *** mgoddard- is now known as mgoddard | 09:11 | |
| dtantsur | would love some reviews on https://review.opendev.org/c/openstack/ironic/+/835098 please | 09:11 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove create_timestamp method https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834127 | 09:20 |
| rpittau | mmm I thought I reviewed it earlier, forgot to press SEND :P | 09:37 |
| dtantsur | :) | 09:39 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove create_floating_ip & get_server_ip methods https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834129 | 10:54 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove _get_router & _create_router methods https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834151 | 10:54 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove _create_network & _get_server_port_id_and_ip4 methods https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834424 | 11:15 |
| opendevreview | Merged openstack/ironic-tempest-plugin master: Remove setup_clients method https://review.opendev.org/c/openstack/ironic-tempest-plugin/+/834425 | 11:15 |
| opendevreview | Merged openstack/ironic master: [doc] Document dynamic network burn-in https://review.opendev.org/c/openstack/ironic/+/835079 | 11:15 |
| iurygregory | good morning Ironic | 13:00 |
| admiyo | good morning | 13:06 |
| admiyo | OK, debugging a PXE failure again. I can see the DHCP request via TCP dump on the controller. I don't see the response, but I think that is due to my tcpdump filter. The cleaning node does not see the response. What should happ[en? I assume conductor somehow tells the dnsmasq container that there is a new node. How? Is it a file that gets put somewhere in the dnsmasq containers /var/lib... _data directory or | 13:22 |
| admiyo | something else? | 13:22 |
| TheJulia | admiyo: so kind of depends on your configuration | 13:28 |
| TheJulia | neutron integrated? | 13:28 |
| TheJulia | also, good morning everyone | 13:28 |
| TheJulia | wrt ironicbaremetal.org, is anyone going to do a summary of the changes? | 13:29 |
| TheJulia | please say yes | 13:29 |
| admiyo | Goooood Morning TheJulia | 13:30 |
| admiyo | neutron is integrated, yes | 13:31 |
| admiyo | THis is a kolla deploy, with Nova | 13:31 |
| admiyo | so I got a VM to deploy and I can talk to it, but that is an all-in-one, so simple on the controller | 13:31 |
| TheJulia | So basically what happens is when we move the node to cleaning, we setup records for PXE booting the node via the extra dhcp options on the port | 13:32 |
| TheJulia | so if you do openstack port list and Identify the port for the machine (use the mac address!) | 13:32 |
| TheJulia | you should see options there | 13:32 |
| TheJulia | neutron handles all of that and when it fails, we remove the records | 13:32 |
| admiyo | There are 5 ports for that one mac address. Does that mean they are not getting cleaned up on a failuer? | 13:33 |
| TheJulia | the dnsmasq associated with ironic is likely for introspection/discovery of hardware | 13:33 |
| TheJulia | oh my | 13:33 |
| TheJulia | that... shouldn't... be | 13:33 |
| TheJulia | so. | 13:33 |
| TheJulia | Neutron is *supposed* to have a unique constraint on mac addresses | 13:33 |
| admiyo | Hey. Its Me. I break things. This is OK | 13:34 |
| admiyo | I think I lied. The macs are diffferent | 13:34 |
| iurygregory | TheJulia, yes I'm working on it today =) | 13:35 |
| TheJulia | admiyo: for a moment, I was going to find the best corgi sideeye picture on the internet | 13:35 |
| iurygregory | also, good morning | 13:35 |
| admiyo | OK, I am going to delete the one baremetal node I have and see what it looks like at the start | 13:35 |
| TheJulia | delete the neutron port, generally deleting a node out of ironic is one of those last resort sort of things | 13:35 |
| TheJulia | but it *sounds* like your physical network is not bridged through to neutron with happiness | 13:36 |
| TheJulia | I'd honestly be trying to tcpdump inside neutron namespaces to verify connectivity | 13:36 |
| admiyo | OK...no BM nodes. Only the controller. No VMs. I have 5 ports. None of them match the MAC address for the node I am about to enroll.... | 13:36 |
| TheJulia | okay | 13:36 |
| TheJulia | okay | 13:36 |
| TheJulia | then you should be okay, I think | 13:36 |
| admiyo | | 15d2503c-acfa-4c57-9180-39b2f68e3ff8 | | 0c:42:a1:49:a0:f4 | ip_address='192.168.97.184', subnet_id='cb7eb95a-bbd8-45ea-bc50-a1b7d6e7bc28' | DOWN | | 13:39 |
| admiyo | that is the new port | 13:39 |
| admiyo | If I go to the controller, what artifact should I see that tells the dhcp server about that new thing | 13:40 |
| admiyo | ugh....I love it when the network connection to the server is so slow that you see a visiable lag when I type. And I type fast. | 13:40 |
| admiyo | OK, so my dhcp is (I believe) docker ps | grep dns | 13:41 |
| admiyo | cd6e06efcdfa eng13sys-r111.scc-lab.amperecomputing.com:5000/kolla/ubuntu-source-dnsmasq:13.0.1 "dumb-init --single-…" 19 hours ago Up 15 hours ironic_dnsmasq | 13:41 |
| admiyo | and so the files for that can be seen via inspect and looking at the volumes | 13:42 |
| TheJulia | ymm hold on | 13:44 |
| TheJulia | did you create that port? | 13:44 |
| TheJulia | so, neutron-dhcp-agent manages the dhcp services in a namespace | 13:44 |
| TheJulia | so it gets the update internally in neutron and updates dnsmasq | 13:44 |
| janders | see you next week Ironic, have a great weekend everyone o/ | 13:46 |
| admiyo | Yeah, create a port | 13:46 |
| TheJulia | g'night janders | 13:46 |
| rpittau | bye janders o/ | 13:47 |
| admiyo | openstack baremetal node create ... then openstack baremetal port create ${ironic_pxe_interface_mac} --node ${NODE_UUID} | 13:47 |
| TheJulia | and then? | 13:47 |
| admiyo | openstack baremetal node validate ${NODE_UUID} | 13:48 |
| admiyo | penstack baremetal node manage ${NODE_UUID} | 13:48 |
| admiyo | openstack baremetal node provide ${NODE_UUID} | 13:48 |
| admiyo | with a few "shows" in between to make sure things are happening | 13:48 |
| TheJulia | okay | 13:48 |
| admiyo | Node is in clean wait | 13:49 |
| TheJulia | what network_interface are you using on your ironic node? | 13:49 |
| admiyo | I opened SOL console on it, saw it PXE request, time out, go to UEFI and stall | 13:49 |
| TheJulia | (and yes, this unfortunately kind of does matter) | 13:49 |
| TheJulia | okay, that helps | 13:49 |
| admiyo | --deploy-interface ? | 13:50 |
| TheJulia | --network-interface | 13:50 |
| TheJulia | the conductor will assign the default if you don't populate it | 13:50 |
| TheJulia | or what it determines is the first in the ordered list | 13:51 |
| admiyo | I only specify (by mac) in the port create | 13:51 |
| admiyo | | network_interface | flat | 13:52 |
| TheJulia | if flat, and there is no network assigned for cleaning/deployment in the conductor configuration, you'll need to attach the a vif to the baremetal node. But, your getting further than that because your actually booting. So Ironic in this case creates a neutron port, populates the dhcp configuration, and goes from there | 13:52 |
| admiyo | that is what comes back from show | 13:52 |
| TheJulia | well actually, I take that back you don't need to pre-attach a vif at all | 13:52 |
| TheJulia | okay | 13:52 |
| admiyo | there should be a cleaning network | 13:52 |
| TheJulia | so, in this stage, you should have a new port in neutron | 13:52 |
| TheJulia | look for it | 13:53 |
| admiyo | cleaning_network = baremetal-datapl | 13:53 |
| admiyo | hmmmmm | 13:53 |
| TheJulia | so | 13:53 |
| admiyo | | cb7eb95a-bbd8-45ea-bc50-a1b7d6e7bc28 | baremetal-dataplane-subnet | b0ae0494-91d7-4e5d-be99-1e151c2262ac | 192.168.97.0/24 | 13:54 |
| admiyo | and on the controller, there is a nic on that network | 13:54 |
| TheJulia | okay | 13:54 |
| admiyo | enP4p4s0f0 inet 192.168.116.62/24 | 13:55 |
| admiyo | ACH! | 13:55 |
| TheJulia | so do you know how to look through network namespaces? | 13:55 |
| admiyo | 97 vs 116 | 13:55 |
| TheJulia | well | 13:55 |
| TheJulia | hold up | 13:55 |
| admiyo | Da Komrade! | 13:55 |
| TheJulia | it doesn't necessarilly have to have an address bound on the host | 13:55 |
| admiyo | yeah, but at a minimum it should be a correct addresss...let me sort this | 13:55 |
| TheJulia | well | 13:55 |
| admiyo | (some bad terraform....) | 13:55 |
| TheJulia | that might still be right | 13:56 |
| admiyo | and by bad, I mean Copy Pasta error from me | 13:56 |
| TheJulia | look at your namespaces, ip netns list | 13:56 |
| TheJulia | and then look inside each of them for the neutron dhcp agent | 13:56 |
| admiyo | I would not expect that to work...its outside the netmast | 13:56 |
| TheJulia | qdhcp-blah-blah | 13:56 |
| admiyo | # ip netns list | 13:56 |
| admiyo | qrouter-f7b297ee-8980-4cc7-9027-079e60aef4fd (id: 2) | 13:56 |
| admiyo | qrouter-fa51bfcc-550d-4cf9-acb4-0cc6b2ae9a6a (id: 1) | 13:56 |
| admiyo | qdhcp-5348a828-a126-40ca-8fc9-34467d2d0e72 (id: 0) | 13:56 |
| admiyo | Now we are beyond my knowledge... | 13:57 |
| TheJulia | ip netns exec qdhcp-5348a828-a126-40ca-8fc9-34467d2d0e72 ip addr | 13:57 |
| TheJulia | (this is how neutron does it's magic, network namespaces) | 13:57 |
| admiyo | ip netns exec qdhcp-5348a828-a126-40ca-8fc9-34467d2d0e72 ip addr | 13:58 |
| admiyo | 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 | 13:58 |
| admiyo | link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 | 13:58 |
| admiyo | inet 127.0.0.1/8 scope host lo | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| admiyo | inet6 ::1/128 scope host | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| admiyo | 29: tapf5ebcdce-db: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc noqueue state UNKNOWN group default qlen 1000 | 13:58 |
| admiyo | link/ether fa:16:3e:8d:e1:3a brd ff:ff:ff:ff:ff:ff | 13:58 |
| admiyo | inet 10.0.0.2/24 brd 10.0.0.255 scope global tapf5ebcdce-db | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| admiyo | inet 169.254.169.254/32 brd 169.254.169.254 scope global tapf5ebcdce-db | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| admiyo | inet6 fe80::a9fe:a9fe/64 scope link | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| admiyo | inet6 fe80::f816:3eff:fe8d:e13a/64 scope link | 13:58 |
| admiyo | valid_lft forever preferred_lft forever | 13:58 |
| TheJulia | so that is likely your baremetal dataplane network right? | 13:58 |
| admiyo | sorry for the spam | 13:58 |
| admiyo | so nothing on 192.168 anyway | 13:58 |
| TheJulia | paste.openstack.org :) | 13:58 |
| admiyo | yeah, yeah...I'm a bad citizen today | 13:59 |
| TheJulia | so if you do something like `ovs-vsctl show` on the host, you should see tapf5ebcdce-db in the bindings | 13:59 |
| TheJulia | and to verify your getting traffic into the namespace, you *really* need to tcpdump for traffic inside the namespace | 13:59 |
| TheJulia | so instead of ip addr on the ip netns exec command, use tcpdump there | 13:59 |
| TheJulia | if you *don't* see the tap interface bound through to a physical interface, then you have a misconfiguration someplace. Neutron might have the wrong input in term of the physical networks and the bindings | 14:00 |
| TheJulia | but if you see any traffic, then that is an awesome sign | 14:00 |
| admiyo | # ovs-vsctl show | 14:00 |
| admiyo | f174cf05-6ce0-4282-a814-26020351df0d | 14:00 |
| admiyo | ovs_version: "2.13.5" | 14:00 |
| TheJulia | https://external-content.duckduckgo.com/iu/?u=https%3A%2F%2Fi.pinimg.com%2Foriginals%2F95%2F40%2Fb9%2F9540b955723aaa6c65ad51a02d6b280e.jpg&f=1&nofb=1 | 14:01 |
| TheJulia | no ovs output? | 14:01 |
| admiyo | nope | 14:01 |
| TheJulia | could you be using ovn? | 14:02 |
| TheJulia | ... but that should still show something there | 14:02 |
| admiyo | Its a Kolla install, defaults. Ubuntu 21. I did not yet specify, but I can look | 14:02 |
| TheJulia | you have routers, which makes me think you have a running l3 agent | 14:03 |
| TheJulia | but it sounds like neutron-ovs-agent is not running | 14:03 |
| TheJulia | I'd check its logs | 14:03 |
| admiyo | neutron_openvswitch_agent container is running | 14:04 |
| TheJulia | time to check logs, since you don't have any ovs records | 14:04 |
| admiyo | no stack trace in the ovslog | 14:05 |
| TheJulia | errors? | 14:05 |
| TheJulia | warnings? | 14:05 |
| admiyo | Nope | 14:05 |
| admiyo | blissfully unaware | 14:05 |
| TheJulia | "critical hit, 1000 points, your cloud is down" messages ? | 14:06 |
| admiyo | nope | 14:06 |
| TheJulia | is it *actually* running in the container | 14:06 |
| TheJulia | or is it having a super long nap of "no cloud for you" | 14:06 |
| TheJulia | I'd try to restart the container, tbh | 14:07 |
| TheJulia | maybe it will do the needful, but it sounds like it is basically not operational | 14:07 |
| admiyo | OK. I want to clean up the terraform etc first and then I will. THis gives me somefodder for my Banthas | 14:07 |
| TheJulia | yay, electrical inspector nor contractor is here yet | 14:08 |
| opendevreview | Verification of a change to openstack/ironic master failed: Start API after conductor is started in the combined executable https://review.opendev.org/c/openstack/ironic/+/835098 | 14:35 |
| TheJulia | cbouchar_: o/ you around? | 14:38 |
| cbouchar_ | TheJulia: I sure am! | 14:38 |
| TheJulia | cbouchar_: do you have the exception from when you put a folder path in image_source ? | 14:38 |
| cbouchar_ | TheJulia. I can easily reproduce. I'll forward via email when done. | 14:39 |
| TheJulia | cbouchar_: thanks! | 14:40 |
| cbouchar_ | TheJulia: so I need to put traceback into my kickstart correct? | 14:41 |
| TheJulia | cbouchar_: no, not in your kickstart config | 14:41 |
| TheJulia | we never even get that far when you use a path without a trailing / | 14:41 |
| cbouchar_ | that's right. I'm all over the place with different subjects. | 14:42 |
| TheJulia | no worries, sorry I've not been able to focus on this so far this week, I've been pulled in a few different directions | 14:45 |
| cbouchar_ | TheJulia: With --instance-info image_source=http://<ip>:<port>/DVD_MNT_RHEL7/, when I deploy I see "Failed to validate deploy or power info for node <node-id>. Error: Validation of image href http://<ip>:<port>/DVD_MNT_RHEL7/ failed, reason: Got HTTP code 403 instead of 200 in response to HEAD request. (HTTP 400)" | 14:47 |
| TheJulia | right... okay | 14:47 |
| TheJulia | thanks | 14:47 |
| TheJulia | cbouchar_: hey, could you remove the trailing / real quick ? | 14:50 |
| cbouchar_ | TheJulia: ??? | 14:51 |
| cbouchar_ | trailing what? | 14:51 |
| TheJulia | the trailing slash on the url | 14:51 |
| TheJulia | I'm trying to make some code to gracefully handle these cases | 14:51 |
| cbouchar_ | got you | 14:51 |
| cbouchar_ | TheJulia: Failed to validate deploy or power info for node <node-id>. Error: Validation of image href http://<ip>:<port>/DVD_MNT_RHEL7 failed, reason: Got HTTP code 301 instead of 200 in response to HEAD request. (HTTP 400) | 14:54 |
| TheJulia | awesome thanks | 14:54 |
| opendevreview | Verification of a change to openstack/ironic master failed: Start API after conductor is started in the combined executable https://review.opendev.org/c/openstack/ironic/+/835098 | 15:09 |
| rpittau | bye everyone, have a great weekend! o/ | 15:49 |
| dtantsur | mmm, too many RETRY_LIMITs | 16:26 |
| * dtantsur reports to #opendev | 16:28 | |
| TheJulia | :( | 17:09 |
| dtantsur | Team, I'll be on PTO next week. See you on the PTG! | 17:22 |
| admiyo | ovs-vsctl show | 17:38 |
| admiyo | f174cf05-6ce0-4282-a814-26020351df0d | 17:38 |
| admiyo | Bridge br-ex | 17:38 |
| admiyo | Port enp1s0f1 | 17:38 |
| admiyo | Interface enp1s0f1 | 17:38 |
| admiyo | Port br-ex | 17:38 |
| admiyo | Interface br-ex | 17:38 |
| admiyo | type: internal | 17:38 |
| admiyo | ovs_version: "2.13.5" | 17:38 |
| admiyo | That is a little more sane, right? | 17:38 |
| admiyo | but still no tap | 17:39 |
| opendevreview | Julia Kreger proposed openstack/ironic master: WIP: Make non-image deploys more sane... https://review.opendev.org/c/openstack/ironic/+/834709 | 17:49 |
| TheJulia | admiyo: much more sane | 17:51 |
| TheJulia | dtantsur: o/ | 17:51 |
| TheJulia | cbouchar_: rloo: ^^^ So I've started on this after some chatting with dtantsur earlier this week. It passes unit tests locally, but I need to write a number of tests. the idea is that with it we *should* be able to allow anaconda/kickstart based usage with directory paths *or* stage2 ramdisks based on image_source. cbouchar_ since it passes the normal ironic unit tests, you might want to give it a quick patch and try, | 17:52 |
| TheJulia | just fwiw. | 17:52 |
| rloo | thx TheJulia! zer0c00l fyi ^^ | 17:53 |
| cbouchar_ | TheJulia: So instead of using image_url, I use image_source. | 17:53 |
| TheJulia | cbouchar_: yes | 17:54 |
| TheJulia | technically my change also makes us tollerant of not blowing up on image redirects | 17:54 |
| TheJulia | which I guess is a good thing in limited quantity | 17:54 |
| TheJulia | zer0c00l: please review, I'm hoping I'm not detonating the purpose of anything, but standalone usage is a prime use case for upstream. Also, with this I *think* we can put CI in place | 17:59 |
| cbouchar_ | TheJulia: I'm in xena branch and this patch doesn't patch cleanly. I could search for dependent patches but not sure what kind of hole I'll be digging into. | 18:09 |
| TheJulia | oh, likely image_type stuffs | 18:10 |
| TheJulia | so backporting it would be deleting some of it I guess | 18:10 |
| TheJulia | but fundimentals should still be good | 18:11 |
| TheJulia | I did just leave some comments on the change | 18:11 |
| cbouchar_ | yes | 18:13 |
| cbouchar_ | Let me keep looking | 18:13 |
| cbouchar_ | The diff it is dependent on is somewhat big and I suspect it may also have dependencies. I'll have to be patience for your changes to bubble down. | 18:21 |
| cbouchar_ | git show a813c769e8 | 18:21 |
| TheJulia | yup, the image_type work | 18:24 |
| TheJulia | we shoudl be able to unwind and separate that but likely only when we go to backport | 18:24 |
| TheJulia | I'll go highlight what can likely go in the changeset too | 18:24 |
| TheJulia | cbouchar_: rloo found an issue. | 18:34 |
| TheJulia | rloo: replied, cbouchar_ commented on the image type stuff and noted the two lines you'll likely need to move to manager.py on Xena | 18:37 |
| * TheJulia goes and writes unit test to detect what ruby found first | 18:37 | |
| TheJulia | oh wait, it is already checked, it is my redirect stuffs... yeah. | 18:40 |
| TheJulia | cbouchar_: you around? | 19:01 |
| cbouchar_ | yes | 19:02 |
| TheJulia | last week when we were chatting, I remember you did curl -X HEAD http://path/to/your/loopback/ | 19:03 |
| TheJulia | What was the error/headers it responded with? | 19:03 |
| cbouchar_ | I don't think I have it right. This is what I did and see. curl -X HEAD http://<ip>:<port>/DVD_MNT_RHEL7/ | 19:06 |
| cbouchar_ | Warning: Setting custom HTTP method to HEAD with -X/--request may not work the | 19:06 |
| cbouchar_ | Warning: way you want. Consider using -I/--head instead. | 19:06 |
| TheJulia | Yeah, it was -I that you used | 19:10 |
| TheJulia | doh | 19:10 |
| TheJulia | I could have sworn you got an error, but I just tried it with a centos-stream mirror and didn't get an error, so I'm slightly confused | 19:17 |
| TheJulia | then again, it could just be individual server configuration | 19:18 |
| cbouchar_ | Here is comes: curl -I http://<ip>:<port>/DVD_MNT_RHEL7/ | 19:20 |
| cbouchar_ | HTTP/1.1 403 Forbidden | 19:20 |
| cbouchar_ | Server: nginx/1.20.1 | 19:20 |
| cbouchar_ | Date: Fri, 25 Mar 2022 19:19:09 GMT | 19:20 |
| cbouchar_ | Content-Type: text/html | 19:20 |
| cbouchar_ | Content-Length: 153 | 19:20 |
| cbouchar_ | Connection: keep-alive | 19:20 |
| TheJulia | so index just not permitted I guess | 19:20 |
| TheJulia | so if the path ends with / *and* 403 forbidden, then we're likely okay, path wise | 19:20 |
| TheJulia | or... is that just nginx | 19:21 |
| cbouchar_ | without trailing / I get HTTP/1.1 301 Moved Permanently | 19:22 |
| TheJulia | yeah | 19:24 |
| * TheJulia feels sad that the home NAS does everything *but* http | 19:32 | |
| TheJulia | :628 | 21:22 |
| TheJulia | doh | 21:22 |
| * TheJulia calls it a day | 22:13 | |
| * TheJulia will resume on the make things more sane patch tomorrow or sunday | 22:13 | |
| admiyo | ip netns exec qdhcp-338799b4-9090-4653-bea9-16caff89de4f ip addr | 22:17 |
| admiyo | only comes back with the loopback now. Is thatprogress or regress? | 22:17 |
| admiyo | egress | 22:18 |
| TheJulia | regression :\ | 22:22 |
| TheJulia | sorry | 22:22 |
| admiyo | What is supposed to create things in that namespace? | 22:37 |
| admiyo | Am I am making life impossible for myself by doing a Kolla Nova install and Ironic at the same time? Would I be better off with a bifrost install? Is that better documentated? | 22:39 |
| admiyo | Documentated being the word of the day | 22:39 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Device management driver iface https://review.opendev.org/c/openstack/networking-baremetal/+/835170 | 23:09 |
| opendevreview | Harald Jensås proposed openstack/networking-baremetal master: Add netconf-openconfig device driver https://review.opendev.org/c/openstack/networking-baremetal/+/835324 | 23:10 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
