Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-11-12
17:42:37 sean-k-mooney become: yes
17:43:02 sean-k-mooney that will alocate 1G on the contoler and 4 on the compute of 2MB hugepages
17:43:12 sean-k-mooney that should be enough
17:43:55 artom Ack
17:55:47 artom sean-k-mooney, I think the problem with using https://zuul-ci.org/docs/zuul-jobs/general-roles.html#role-copy-build-sshkey is that it's 1-user only
17:55:57 artom IOW, 'tempest' can ssh to 'tempest'
17:56:13 artom We need 'tempest' to SSH to 'stack' or some other account that has limitless sudo powers
17:59:25 sean-k-mooney artom: the same key is already copied to root and stack
17:59:32 sean-k-mooney root is done in the pre job
17:59:43 sean-k-mooney stack in the muntinode playbook
18:00:17 sean-k-mooney so if we use it for tempest all 3 users will have the same private and public key
18:00:59 sean-k-mooney you wouldnt want to do that in production but it should be fine for ci
18:01:30 artom *facepalm*
18:01:31 artom I see
18:01:52 artom The devstack orchestrate-devstack role will put the pubkey in stack's authorized_keys for us
18:02:03 artom And we can put the private key in tempest's .ssh
18:02:13 artom In theory, anyways
18:05:53 artom ... do we have access to another project's roles? Not sure...
18:06:10 donnyd artom: the expanded flavors have more memory
18:11:45 sean-k-mooney artom: am we can get acess to them yes there are two ways to do i but your job inherits form the tempest job yes
18:11:52 sean-k-mooney if so then it will do it already
18:12:18 sean-k-mooney you you can just invoke the copy-build-key role twice once for stack and once for tempest
18:28:03 artom https://review.opendev.org/#/c/691062/57 let's see how badly stuff explodes
18:41:17 sean-k-mooney artom: you do not need the ssh keys in that anymore
18:41:36 sean-k-mooney but i think it should be ok
18:41:59 sean-k-mooney its runing in a sperate pool so it should be pretty fast
18:42:15 artom sean-k-mooney, something borked, it's retrying now
18:42:23 artom Heh, oh yeah, I should get rid of the keys
18:42:29 artom I'll wait for it to work first
18:43:00 sean-k-mooney the console for the job is here http://zuul.openstack.org/stream/11de9dc2d4fd4f7d931eedeea276af3e?logfile=console.log
18:43:23 sean-k-mooney looks like tis doing the normal multinode setup
18:44:30 artom That's the second attempt
18:44:40 sean-k-mooney yep
18:44:59 sean-k-mooney ill leave it open on one of my monitors and see what hapens
18:45:49 sean-k-mooney haha
18:45:54 sean-k-mooney you have a typo
18:46:36 sean-k-mooney artom: http://paste.openstack.org/show/785994/
18:47:45 sean-k-mooney you did not include teh whitebox-compute role in the commit
18:47:52 sean-k-mooney or you forgot to remove that line
18:49:27 artom *facepalm*
18:49:37 artom 'git commit -a' isn't magic
18:49:53 sean-k-mooney it speciricly ignore untracted files
18:52:03 artom I know
18:52:42 sean-k-mooney good becaue i apparently cant spell specifcally and im not sure you woudl figure it out if you didnt :P
18:55:33 sean-k-mooney i need to try enabling spell check for irc again but the last time i tried that it was kind of depressing
18:57:29 artom *shrug* I forgive you :)
18:57:48 artom You have other uses besides spelling ;)
18:58:31 sean-k-mooney well if you want a terible way to generate passwords just ask me to spell somthing
18:59:14 artom lulz
19:04:54 sean-k-mooney so looking at the job you are trying to allocate 8G of hugepages on all nodes
19:05:07 sean-k-mooney that will not work with the non expanded flavors
19:05:11 sean-k-mooney *lables
19:07:17 artom derp
19:46:29 mriedem cool, the instance_actions_events table has a detail TEXT column that is never used, and when creating the InstanceActionEvent object we pass the exc_val but it gets passed down to the DB API in a 'message' kwarg rather than 'details' so that column is never populated
19:48:08 mriedem so as a non-admin user you can see a particular event passed or failed but not why, e.g. NoValidHost
19:48:20 mriedem and only admins can see the traceback (like a fault) by default
19:49:39 sean-k-mooney hum ok that sound like the behavior we want. are you going to reuse the colume for something
19:50:20 sean-k-mooney such as storging the traceback?
19:50:32 mriedem the traceback is stored in the traceback column
19:50:40 mriedem the exc_val is not stored at all
19:50:48 mriedem b/c the object is using the wrong column name
19:51:07 sean-k-mooney sure but i was wondering if you planned to chagne that
19:52:03 mriedem i think it would be useful to at least expose to the non-admin owner of the server the exception type, like we do for a fault message
19:53:37 sean-k-mooney ya maybe. as long as its just the type and not the body of the excption its proably safe. although im not sure all operator would be ok with that
19:53:55 sean-k-mooney e.g. it might affect there sla depening on what the falut was
19:54:13 mriedem if the instance goes to ERROR status we're already showing the exception type in the fault message
19:54:25 mriedem if not the formatted message
19:54:30 sean-k-mooney ya that is ture
19:54:47 sean-k-mooney i normally am logged in as root but we get the message in that case at least
19:54:56 sean-k-mooney * admin
19:55:05 sean-k-mooney so im not sure what a non admin sees
20:10:55 sean-k-mooney artom: :) looks like it worked https://openstack.fortnebula.com:13808/v1/AUTH_e8fd161dc34c421a979a9e6421f823e9/zuul_opendev_logs_3ae/691062/59/check/whitebox-multinode-devstack/3ae547c/testr_results.html.gz
20:11:29 sean-k-mooney we still need to adress the skips
20:12:14 sean-k-mooney but the ssh keys worked correctly and it looks liek the hugepages allocation did not cause memory issues
20:23:15 mriedem dansmith: your suggestions here to use the tuples for comparing times doesn't work in all cases https://review.opendev.org/#/c/636224/50/nova/compute/api.py@4956
20:23:51 mriedem and i don't really want to bake a bunch of conditional logic into unreadable tuples
20:25:22 dansmith aight, well, do what you want
20:25:32 dansmith inlining all that in a closure seems bad,
20:25:46 dansmith especially for testability, which has clearly manifested itself already
20:26:09 dansmith that could be a
20:26:16 dansmith def newer_than(obj)
20:26:42 dansmith method on NovaObject and be usable for other things, as well as easier to test
20:27:11 dansmith is there not some way we could do this without times?
20:27:22 dansmith because that's kinda smelly in general,
20:27:33 sean-k-mooney i was going to suggest a before or after method or something but ya the tuple way would be nice if it was not for the value error
20:27:37 dansmith assuming they're all time-synced, you don't have any delays, delayed transactions, etc
20:27:45 sean-k-mooney *type error
20:29:01 sean-k-mooney oh this is in get_newer_job already
20:29:31 sean-k-mooney you could define a lessthan operator on the migration object
20:29:45 sean-k-mooney although i guess that would have the same proably with None
20:29:57 dansmith no, it wouldn't
20:30:08 dansmith but also, less-than can mean lots of things other than time
20:30:20 sean-k-mooney for a migration object
20:30:26 sean-k-mooney and ya i could
20:30:42 mriedem we could just say f it and do like get_all and filter out by uuid after we've already processed one
20:31:13 mriedem this https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2883
20:31:49 dansmith mriedem: yeah
20:32:12 mriedem the migration from the first cell processed in the results is always the one that is returned,
20:32:20 mriedem which could be wrong, but i don't really care at this point
20:32:40 dansmith does it matter/

Earlier   Later