Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-08-08
15:55:09 sdague they are coming in from the web, so it's just about how things end up on the glass with navigation patterns
15:55:18 stephenfin sdague: It kind of does because the filesystem manifests itself in the URL structure
15:55:43 sdague stephenfin: which is ignored by 99% of consumers
15:55:46 dansmith mriedem: I dunno, I think that test looks at a lot of RT stuff doesn't it?
15:55:56 stephenfin and I was pretty sure the plan was to still have central "Admin guide" and "Install guide" pages that deep link into each folder
15:56:02 stephenfin sdague: True
15:56:23 bauzas jaypipes: cdent: gibi: I clarified my opinion in https://review.openstack.org/#/c/491529/5/nova/tests/functional/integrated_helpers.py@75 about using all the filters
15:56:31 stephenfin Documentation is hard work.
15:56:35 sdague stephenfin: for sure
15:56:48 bauzas jaypipes: cdent: gibi: I'm okay with using all of them by default, just not in the same change, that's it
15:57:06 sdague stephenfin: it's possible we get there, there is so much near duplication of content from these different sources though, it's going to take a lot of massaging to figure it out
15:57:09 asettle stephenfin sdague - I'm going to print screen that you two are both agreeing that docs are hard work and cherish that forever
15:57:22 asettle Do you know how many times I've heard that docs aren't necessary and my job is worthless :D
15:58:21 sdague asettle: zomg, that's terrible.
15:58:46 cfriesen asettle: anyone making such a statement is obviously not worth listening to.
15:58:54 asettle *shrug* it is what it is :)
15:59:02 asettle cfriesen: in a nutshell ;0
15:59:04 asettle ;) *
15:59:19 asettle My favourite argument is that their code is so pure no docs are required for it's beauty
15:59:30 asettle Because 'you should just be able to read the code' ;)
15:59:45 cfriesen having written pure and beautiful code and then gone back 5 years later.....
16:00:39 asettle hahahhahaha
16:01:28 sean-k-mooney asettle: cfriesen that argument may fly for perl. i here its considered a write only language so need to read it again 5 years later :P
16:02:10 asettle stephenfin: uhhhhh who wrote that? :P did we?
16:02:18 asettle sean-k-mooney: pretty much goes for all code, no?
16:02:20 sean-k-mooney stephenfin: that is when you use qemu without kvm/xen i think
16:02:57 sfinucan sean-k-mooney: Doesn't look like it http://user-mode-linux.sourceforge.net/
16:03:14 sean-k-mooney asettle: there is some code that is self documenting and it you coreccletly decompose your codded into small fuctions it can be but ingererall comment help alot
16:03:17 sdague sfinucan: it's what came before xen
16:03:18 stephenfin Oops - wrong nick
16:03:30 stephenfin sdague: Is it still relevant?
16:03:38 sdague stephenfin: I wouldn't think so
16:03:45 cfriesen sean-k-mooney: no, actually it's different and funky
16:03:52 edleafe asettle: Docs aren't necessary - to the person writing the code. To everyone else, though...
16:03:57 sdague user mode linux was run linux kernels as user processes
16:04:04 asettle edleafe: I think that's the point at the end, right? :p
16:04:10 asettle I wrote my first little python script so I feel ya
16:04:16 asettle It is obviously perfect
16:04:16 edleafe asettle: :)
16:04:21 asettle And makes sense to *everyone*
16:04:27 asettle :p
16:04:28 edleafe but of course
16:04:29 sdague circa late 90s
16:04:47 sean-k-mooney stephenfin: huh that is weird but ok if nova supports it thne great if there is not a ci for it though its broken by default
16:04:59 sdague honestly, uml is kind of docker before docker
16:06:20 cfriesen sdague: wouldn't that be lxc?
16:06:22 sean-k-mooney sdague: so its like lxc/openvz container. e.g a sandboxed system level container/chroot?
16:06:55 sdague sean-k-mooney: yeh, except pre namespaces
16:07:18 cfriesen sean-k-mooney: sdague: I think it's actually linux implemented using POSIX as the architecture.
16:08:01 sdague anyway, the crux of it, I'd remove all the references to it in the nova space. I expect someone got somethign working in cactus, but never bothered again after
16:08:06 sdague and the reference lasted until now
16:08:42 cfriesen sdague: yeah, agreed. don't see why anyone would actually use it with nova
16:08:52 stephenfin Yup, they're gone
16:09:02 stephenfin There was only one (that I've found so far)
16:09:14 sdague oh, this is why - https://blueprints.launchpad.net/nova/+spec/austin-user-mode-linux
16:09:27 sdague they were attempting to use it for testing instead of 2nd level qemu
16:09:44 sean-k-mooney sdague: looking at https://wiki.openstack.org/wiki/Nova/UML it looks like its pluged in under livbivrt
16:09:54 stephenfin sdague: Importing this config guide is proving slower than I thought (alll the cross-references). I may need to have you take over it if I don't get much further in the next hour or so
16:10:02 stephenfin (I'm gone tomorrow on)
16:10:04 sdague stephenfin: no prob
16:10:08 mriedem melwitt: i've rebased and cleaned up the conflict in the stable branches for https://review.openstack.org/#/q/I8cd056fa17184a98c31547add0e9fb2d363d0908,n,z
16:10:17 mriedem melwitt: now just need to get the series of changes below that one merged
16:10:58 sdague stephenfin: before you leave, if you can send me an email with the patch url, and what the major reference fixing looks like
16:11:10 sdague just to get me running on it quicker
16:11:21 sdague as I'm about to go out for a bike ride while the getting is good
16:11:40 stephenfin Will do
16:20:59 openstackgerrit Jay Pipes proposed openstack/nova master: remove provider allocs in confirm/revert resize https://review.openstack.org/488510
16:21:00 openstackgerrit Jay Pipes proposed openstack/nova master: placement: refactor healing of allocations in RT https://review.openstack.org/491850
16:21:00 openstackgerrit Jay Pipes proposed openstack/nova master: Resource tracker compatibility with Ocata and Pike https://review.openstack.org/491012
16:21:26 jaypipes dansmith: ok, I did my best on the above but it's icky and may have made a mistake or two.
16:21:34 dansmith okay
16:21:37 jaypipes dansmith: really need your eyeballs on it.
16:22:07 dansmith I guess I didn't think there was that much change needed...
16:24:01 dansmith jaypipes: this replaces the delete with "heal" which doesn't make sense to me
16:24:03 dansmith since we're also still healing in the non-delete path
16:24:50 jaypipes dansmith: the heal will need to recreate allocations for ocata-deleted allocations, which is why I renamed it heal. I know, it's bad
16:25:04 dansmith and it doesn't heal right? it deletes.
16:25:35 jaypipes dansmith: well, this method will need to re-create allocations that an ocata compute node deleted incorrectly.
16:26:20 jaypipes dansmith: if an ocata compute node was the source for a migrate to a pike compute host.
16:26:31 dansmith but that's already done in update_usage_for_instance right?
16:26:46 dansmith in the top patch, conditional on the presence of ocata computes
16:28:14 jaypipes dansmith: ack, sorry, yes, you're right.
16:28:22 jaypipes dansmith: I can rename it back to delete...
16:28:32 dansmith jaypipes: I'm not sure what the point of this patch is though.. just some extra conditions for not calling delete?
16:29:51 jaypipes dansmith: the point of this patch was two things: 1) to only delete allocations when the instance is DELETED or SHELEVED_OFFLOADED and 2) add a bunch of log statements so we can see what state instances are in during this _remove_deleted_instances_allocations() method
16:30:44 dansmith jaypipes: okay I thought we were doing this filtering earlier such that by the time we got here we didn't have anything other than the ones we needed to delete..
16:30:55 jaypipes dansmith: I added the debug log statements when trying to fix the issue alex_xu brought up about evacuate code paths not being correct.
16:31:09 dansmith we've been round and round about this particular conditional, so maybe I missed something
16:31:33 jaypipes dansmith: the evacuate code paths were not being affected by the update_usage_for_instance() path for some reason.
16:31:45 jaypipes dansmith: on the source host at least.
16:32:34 jaypipes dansmith: so when trying to verify the evacuate code path for scenario when source host is restarted, I was trying to get log messages about the instances in this method. sorry, it got really confusing :(
16:32:55 dansmith jaypipes: so vm_states.ALLOW_RESOURCE_REMOVAL excludes things that aren't deleted or shelved_offloaded, so doesn't that cover us?
16:33:07 dansmith oh, not for deleting allocations I guess
16:33:17 jaypipes right, that actually results in InstanceNotFound.
16:33:27 dansmith what does?
16:33:32 jaypipes when an instance is deleted.
16:33:41 jaypipes the Instance.get_by_uuid() returns InstanceNotFound
16:34:47 openstackgerrit Stephen Finucane proposed openstack/nova master: doc: Import configuration reference https://review.openstack.org/491853
16:35:04 mriedem https://docs.openstack.org/releasenotes/nova/unreleased.html#id15
16:35:04 mriedem stephenfin: in case you haven't left yet, there are two duplicate upgrade notes for similar things in the libvirt driver about port filtering capability

Earlier   Later