[00:53] <NickZ> can someone please point me to where i can get ACPI throttling working on ubuntu server before the fans on this thing drive me mad
[06:11] <lordievader> Good morning
[11:22] <tobias-urdin> coreycb: another upgrade issue but is probably only rdo packages, gnocchi 4.3 requires pyparsing >= 2.2.0 if you want to verify for ubuntu :)
[11:28] <dpawlik> coreycb: hi, quick question: cloud archive team will provide nova-conductor package with start scripts that will configure nova superconductor and also service for cell conductor or it can not be done using package?
[11:36] <coreycb> dpawlik: you can just use nova-conductor
[11:36] <coreycb> tobias-urdin: i think we're good, bionic has python-pyparsing 2.2.0 as well
[11:37] <dpawlik> ok, but when you create few other cells, other conductors need to "read" another config
[11:46] <coreycb> dpawlik: there's shouldn't be any difference in nova-conductor's per cell afaik. each cell has it's own database.
[11:47] <dpawlik> coreycb: as I see superconductor should read /etc/nova/nova.conf file and cell should e.g./etc/nova/nova_cell1.conf
[11:47] <dpawlik> so init scripts should take new config
[11:53] <coreycb> dpawlik: i'm guessing that is something that's done in devstack for all in one testing of cells?
[12:39] <dpawlik> coreycb: devstack is configuring infra using super conductor
[12:39] <dpawlik> so  by default they are using multiple cells configuration
[12:40] <dpawlik> its a really good example how e.g. conductor services should work
[12:41] <coreycb> dpawlik: but i think the only reason you'd need to rename nova.conf per cell is if you are installing all of your cells on the same machine which seems like something that wouldn't be that useful
[12:43] <dpawlik> https://docs.openstack.org/nova/latest/user/cellsv2-layout.html#multiple-cells
[12:44] <dpawlik> coreycb: so you can install on controller cell1 conductor and also cell2, its just a service using different configuration.
[12:45] <dpawlik> coreycb: so you propose to not install cell0 controller and cell1 controler on the same host?
[12:48] <coreycb> dpawlik: correct, i'm not sure why you'd use cells on the same machine
[12:48] <coreycb> dpawlik: since the whole point of them is to be able to scale
[12:48] <coreycb> multiple cells on the same machine, that is
[12:52] <dpawlik> coreycb: I did not look on cells from such way
[12:54] <dpawlik> coreycb: now I need to replan deployment of my dev zone ;(
[16:26] <rbasak> nacc: opinion on https://code.launchpad.net/~racb/usd-importer/+git/usd-importer/+merge/355449 please? AFAICT, this is a reasonable thing to do. Though I'm not sure why git ubuntu tag requires the tree to be clean in the first place, but we can address that separately later.
[16:26] <rbasak> (in case there's some deep complex reason for it that needs a more complex cleanup)
[16:41] <nacc> rbasak: looking
[16:57] <nacc> rbasak: pygit2 won't let you create a tag if the tree is dirty, iirc.
[16:57] <nacc> rbasak: we're not using pygit2's tag api, but i think at some point we were?
[16:57] <rbasak> I see, thanks
[16:58] <nacc> rbasak: that's my short recollection, at least
[16:58] <nacc> rbasak: i also recall pygit2's repo.status() returning non-zero even when `git status` doesn't (even with --ignored and other flags)
[16:58] <nacc> i never debugged that further at the time, as this was a reasonable workaround
[18:44] <havenstance> just love when you do-dist-upgrade and it changes your custom settings lol
[18:44] <havenstance> I had to remember how to configure lubuntu to auto login....
[18:44] <havenstance> wrong chan sry ya'll
[18:44] <nacc> havenstance: if that was something controlled by a system config file, it would have prompted you about it, unless you have it configured not to
[18:45] <havenstance> nacc, nah files were moved, just a simple lightdm --show-config told me where the new one was
[18:45] <havenstance> all is well
[18:46] <nacc> havenstance: ah ok
=== JanC_ is now known as JanC