=== frankban|afk is now known as frankban
[08:11] <mwe11> pmautils: @documentation: I had to search within the source code to find out how to create md-raid (set raid level) via commandline. I tried "raid1", "RAID1", "1", "RAID-1" at the end I found out to simply set "raid1" :D
[08:12] <mwe11> It would be nice to have 1. documentation and 2. increased error-messages
[08:14] <mwe11> not only "raid-level parameter is wrong" but "raid-level parameter is wrong. It should be one of "raid-1", raid-5", [...]"
[08:18] <mwe11> It's defined in /usr/lib/python3/dist-packages/maasserver/enum.py
[10:27] <kklimonda> I'm trying to use MAAS 2.2 pods with virsh driver, but shortly after I restart regiond the Virsh option from the Pod Type disappears
[11:19] <magicaltrout> hello folks
[11:19] <magicaltrout> maas testing this week
[11:19] <magicaltrout> Node commissioning failure - 'cloudinit' running config-ntp with frequency once-per-instance <- I see that, which I also see https://bugs.launchpad.net/maas/+bug/1629578 and https://askubuntu.com/questions/872971/maas-2-1-commissioning-failure-in-cloudinit
[11:19] <magicaltrout> anyone got any clues
[11:33] <magicaltrout> /usr/sbin/ntpd: error while loading shared libraries: libcap.so.2: cannot stat shared object: Permission denied
[11:43] <magicaltrout> nm found https://bugs.launchpad.net/maas/+bug/1697209
[13:52] <gimmic> Is there a way to force a rescan of the dhcp environment? I have maas thinking some IPs are allocated to nodes which have a different static allocation
[13:52] <gimmic> maas is providing the dhcp as well
[13:53] <gimmic> I assumed 'map subnet' would take care of it, but apparently not
[14:09] <gimmic> I see I can use maas cli to delete the static allocated interface, but I'm not sure how to remove the 'discovered' one
[14:13] <gimmic> I ended up deleting the associated interface entirely and re-creating it after marking the node 'broken'.
[14:13] <gimmic> Would be nice if the subnet mapping did some reconciling. I suspect that is a bug/oversight
[15:09] <gimmic> Yeah I'm going to have to find a better way to clean this up. The nodes are deployed with static addresses but have 'observed' dhcp leases listed
[15:09] <gimmic> If I delete the interface I have to re-create it(and mark the node as broken etc)
[15:41] <mup> Bug #1703403 opened: regiond workers can use too many postgres connections <MAAS:In Progress by blake-rouse> <MAAS 2.2:Triaged by blake-rouse> <https://launchpad.net/bugs/1703403>
[16:28] <gimmic> Anyone have a good way to clean up the 'observed' ip allocations?
[16:32] <mup> Bug #1703406 opened: MAAS fails to detect Huawei-based server's power type <MAAS:New> <https://launchpad.net/bugs/1703406>
=== frankban is now known as frankban|afk
[19:54] <gimmic> Is there a reason you can't change a machine from state "failed deployment" to "rescue mode" directly?
[19:54] <gimmic> I have to Failed deployment -> mark broken -> rescue mode
[20:09] <gimmic> Just did an update from 2.2 to 2.2.1 now I'm seeing "failed to commission" and "failed to enter rescue mode"
[20:09] <gimmic> gonna reboot maas :|
[21:57] <mup> Bug #1703462 opened: rack controller rejected by region <cdo-qa> <cdo-qa-blocker> <MAAS:Incomplete> <https://launchpad.net/bugs/1703462>