| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-06-05 | |||
| 16:08:03 | TheJulia | I'm suddenly reminded of the one time I literally burned out a breaker | |
| 16:08:40 | johnthetubaguy | sean-k-mooney: yeah, anti affinity totally breaks things still, +1 that | |
| 16:09:18 | sean-k-mooney | its because we calusate the alternitive host up front so once one thing files and retire it can cause another thing to fail | |
| 16:09:39 | johnthetubaguy | it gets worse with PCI passthrough | |
| 16:09:48 | johnthetubaguy | or seems to | |
| 16:09:56 | sean-k-mooney | so unless we call the sculder again instead when we do the retry your screwed | |
| 16:10:04 | sean-k-mooney | johnthetubaguy: yes it will | |
| 16:10:58 | sean-k-mooney | johnthetubaguy: that just because we dont track the pci device in placment so when we claim the new candiate we dont calim the pci device and that can cause a reschdule | |
| 16:11:08 | johnthetubaguy | ack | |
| 16:11:20 | sean-k-mooney | johnthetubaguy: we dont need placement to fix that we could allocate the deivce in the conductor be we dont today | |
| 16:12:35 | sean-k-mooney | the pci devices table has 3 states available allocated and claimed i think and the difference between the last too is one is a reservation for a vm that has not yet been created on the host and the other is for when the vm is actully using it | |
| 16:12:52 | johnthetubaguy | oh my... custom resources using the provider yaml totally helps with those races, I think | |
| 16:13:07 | sean-k-mooney | it could | |
| 16:13:21 | sean-k-mooney | but we have no way to associate the RPs with the pci devcices | |
| 16:13:27 | sean-k-mooney | so its close but wont work | |
| 16:13:32 | sean-k-mooney | at least not today | |
| 16:13:42 | johnthetubaguy | because we pick the specific one too early? | |
| 16:14:08 | sean-k-mooney | no we pick the specific on on the compute node but that code has no knolage of placment | |
| 16:14:19 | sean-k-mooney | so it would not pick form the device claimed in palcment | |
| 16:14:25 | sean-k-mooney | it would pick form all devices on the host | |
| 16:15:05 | johnthetubaguy | anyways, I think it fixes a case I am worried about, but we haven't merged that feature (must review that again soon) | |
| 16:17:00 | sean-k-mooney | it will fix it for anything that is not manage by nova direclty | |
| 16:17:09 | sean-k-mooney | like cache allocation | |
| 16:17:35 | sean-k-mooney | or power/termal based scudling | |
| #openstack-nova - 2020-06-06 | |||
| 12:44:53 | lvdombrkr | hello guys, trying to inject admin_password (inject_password=true ) into image but when creating instance get this error in nova-compute.log | |
| 12:45:09 | lvdombrkr | 2020-06-06 14:53:50.188 6 WARNING nova.virt.disk.api [req-94f485ca-944c-40e9-bf14-c8b8dbe09a7b 052d02306e6746a4a3e7e5449de49f8c 413a4cadf9734fca9ec3e5e6192a446f - default default] Ignoring error injecting admin_password into image (a bytes-like object is required, not 'str') | |
| 12:45:14 | lvdombrkr | Train + Centos8 | |
| 12:45:33 | lvdombrkr | looks something with related with pytho3 | |
| 12:45:41 | lvdombrkr | any clues? | |
| 13:55:36 | lvdombrkr | guys anyone here? | |
| 16:06:02 | frickler | lvdombrkr: that part of nova code seems never to have worked with python3 and is also lacking proper testing, I can reproduce on master, will try to get a patch up soon | |
| 16:12:33 | lvdombrkr | frickler: ok i see. thanks for detailed answer. do i need open bug in https://bugs.launchpad.net/nova/ | |
| 16:12:34 | lvdombrkr | ? | |
| #openstack-nova - 2020-06-07 | |||
| 09:38:40 | openstackgerrit | Jens Harbott (frickler) proposed openstack/nova master: Fix injection for python3 https://review.opendev.org/733982 | |
| 12:13:24 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove hacking rules for python 2/3 compatibility https://review.opendev.org/733987 | |
| 12:20:31 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.PY2 and six.PY3 https://review.opendev.org/726512 | |
| 12:21:02 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.add_metaclass https://review.opendev.org/726565 | |
| 12:21:12 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.reraise https://review.opendev.org/726898 | |
| 12:27:36 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.moves https://review.opendev.org/727224 | |
| 12:28:22 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.iteritems/itervalues/iterkeys https://review.opendev.org/727757 | |
| 12:28:55 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.iteritems/itervalues/iterkeys https://review.opendev.org/727757 | |
| 12:29:20 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.byte2int/int2byte https://review.opendev.org/727777 | |
| 12:29:36 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.binary_type/integer_types/string_types https://review.opendev.org/728094 | |
| 12:30:21 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.text_type (1/2) https://review.opendev.org/728109 | |
| 12:31:20 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.text_type (1/2) https://review.opendev.org/728109 | |
| 12:31:38 | openstackgerrit | Takashi Natsume proposed openstack/nova master: Remove six.text_type (2/2) https://review.opendev.org/728117 | |
| #openstack-nova - 2020-06-08 | |||
| 07:20:36 | bauzas | good morning Nova | |
| 07:30:43 | lvdombrkr | morning nova. is there other ways to inject password except inject_password=true ? | |
| 07:53:55 | gibi | bauzas: good morning | |
| 07:54:26 | gibi | lvdombrkr: you can use cloud init in the guest and you can use the config driver or the metadata service to provide input for cloud init | |
| 07:54:28 | bauzas | first time I don't feel jetlagged or fried after a PTG | |
| 07:55:01 | bauzas | gibi: didn't we deprecated password injection btw. ? | |
| 07:55:04 | gibi | bauzas: I don't miss the jetlag part of the PTG :) | |
| 07:55:26 | bauzas | gibi: next one, we won't have jetlag too | |
| 07:55:27 | bauzas | :) | |
| 07:55:39 | gibi | bauzas: I think we did https://blueprints.launchpad.net/nova/+spec/deprecate-file-injection | |
| 07:55:58 | gibi | ohh this is file injection not password injection | |
| 07:56:42 | lvdombrkr | gibi: thanks for answer. but if i will use metadata, then passwortd will be stored in metadata in plain text, right? | |
| 07:56:51 | bauzas | ah my bad | |
| 07:57:12 | bauzas | I remembered we did deprecated some injection, but like you, it was the wrong way | |
| 07:58:04 | gibi | lvdombrkr: you can inject public keys | |
| 07:58:42 | lvdombrkr | gibi: yes yes about public keys i know, but i need excatly password | |
| 07:58:58 | bauzas | lvdombrkr: question, why can't you use a config tool like Ansible for injecting such things if you don't wanna use the tool we provide ? | |
| 07:59:52 | openstack | Launchpad bug 1882421 in OpenStack Compute (nova) "inject_password fails with python3" [Undecided,In progress] - Assigned to Dr. Jens Harbott (j-harbott) | |
| 07:59:52 | frickler | bauzas: because it is broken? see https://launchpad.net/bugs/1882421 | |
| 08:00:12 | lvdombrkr | bauzas: i want use config you provided (inject_password) but its not working with Centos8 Python3 | |
| 08:00:13 | bauzas | frickler: ah, gtk | |
| 08:00:57 | bauzas | then, I'd recommend to go thru external tools until it's fixed | |
| 08:03:29 | bauzas | frickler: I can take a look at your change if you will | |
| 08:03:38 | lvdombrkr | bauzas frickler : thanks for answers. how long time can take fix this? | |
| 08:06:56 | gibi | bauzas, lvdombrkr, frickler: I'm reviewing the fix now (while on a parallel phone meeting so it will take time) | |
| 08:10:29 | lvdombrkr | gibi bauzas frickler: perfect guys, keep me updated please :) | |
| 08:11:16 | gibi | lvdombrkr: sure | |
| 08:13:14 | aarents | good morning nova | |
| 08:13:27 | aarents | gibi: bauzas just FYI, I've made some stupid max attachement contest on a devstack with machine-type q35: | |
| 08:13:28 | gibi | aarents: good morning | |
| 08:13:34 | aarents | nova interface-list 6cc30912-c4ae-493d-8209-e19f16f101bc | grep -c ACTIVE | |
| 08:13:37 | aarents | 220 | |
| 08:13:45 | aarents | it is only possible because I override in code num_pcie_ports.max to 224 instead of 28 (we force max to 28 because of limitation with aarch64/virt), without doing it max attachment on q35 is 24 (28 - 4 reserved) and on pc it is 26, so we are still better on pc than on q35 regarding max attachement with current code. | |
| 08:13:46 | bauzas | aarents: good morning | |
| 08:14:22 | bauzas | that reminds me a bell, but I don't remember exactly what | |
| 08:14:30 | gibi | aarents: so we need to make the num_pcie_ports.max dynamic based on the machine typ | |
| 08:14:33 | gibi | e | |
| 08:14:41 | aarents | gibi: I think | |
| 08:15:06 | bauzas | didn't we had an issue like this that was raised before ? | |
| 08:15:17 | bauzas | or maybe it was for usb devices, I can't recall | |
| 08:17:42 | gibi | bauzas: I don't have memory about it | |
| 08:20:40 | bauzas | frickler: this could help you to understand the distinguo between the bytes and the str types in py3 https://eli.thegreenplace.net/2012/01/30/the-bytesstr-dichotomy-in-python-3 | |
| 08:21:34 | bauzas | frickler: tl;dr you shouldn't explicitely tell a string as byte-coded | |
| 08:31:08 | frickler | bauzas: yes, this is clear, the question for me is, when a function needs bytes, is it better to make the caller supply bytes or have the function convert str arguments to bytes. that happens multiple times in the inject_* code | |
| 08:48:09 | bauzas | frickler: I definitely prefer the latter | |
| 08:48:28 | bauzas | bytes type should only be used when you need | |
| 09:02:20 | bauzas | frickler: I think i found why you need to horribly encode all the strings | |
| 09:03:30 | bauzas | tl;dr: the guestfs import seems to be py3 | |
| 09:03:32 | bauzas | py2* | |
| 09:05:49 | bauzas | that should come from libguestfs not having the py3 bindings | |
| 09:07:35 | frickler | bauzas: well I tested locally on focal with python3-guestfs installed, surely no py2 involved there | |
| 09:08:03 | bauzas | what I don't understand is why you get a bytes-encoded FS | |
| 09:09:13 | frickler | bauzas: I think that's pretty normal, files contain bytes, you'd have to tell guestfs the encoding in order to be able to convert them to strings | |
| 09:10:11 | bauzas | sure, of course | |