=== meepmeep22_ is now known as meepmeep22 === stokachu_ is now known as stokachu [00:58] Bug #1759944 changed: [2.4, UI] Changing the subnet's fabric/vlan over the UI has no effect [06:36] Bug #1759086 changed: [2.4, websockets] Only send node storage information when needed [06:54] Bug #1759085 changed: [2.4, websockets] Only send node interface data on websocket when needed [07:21] Bug #1760037 opened: [2.4, websockets] GeneralManager is caching all data on load [12:13] Bug #1759085 opened: [2.4, websockets] Only send node interface data on websocket when needed [12:13] Bug #1759086 opened: [2.4, websockets] Only send node storage information when needed [12:19] Bug #1759085 changed: [2.4, websockets] Only send node interface data on websocket when needed [12:19] Bug #1759086 changed: [2.4, websockets] Only send node storage information when needed [12:25] Bug #1759085 opened: [2.4, websockets] Only send node interface data on websocket when needed [12:25] Bug #1759086 opened: [2.4, websockets] Only send node storage information when needed [15:52] Heya folks; need to check my understanding on something. Is the "Cisco UCS Manager" power type completely equivalent to the "UCS Chassis Manager" chassis type? [15:54] I can control individual UCS blades perfectly, but I'm getting "Failed to probe and enlist UCS nodes: Admin implicit props cannot be modified, class=lsbootVirtualMedia, prop=propAcl" in my region log [15:54] when trying to add the entire chassis [16:33] karunamon: yes, it is the same thing, the problem may be that the API changing since first enabled and maas cannot discover all available machines [18:39] looking for parted, fdisk in maas-enlist when enlisting a node drops into initramfs - any help on how I can repartion the disk when this happens [18:39] in busybox [18:39] roaksoax, blake_r how exactly does MAAS determine whether we're booted in legacy or EFI mode during commissioning? [18:40] does it look for EFIVARs or is it some other method? [18:42] bladernr: in <2.3 we look at how the machine is booting *e.g. if it pxe boots and requests efi stuff [18:42] bladernr: in 2.4+ we look at /sys/firmware/efi [18:42] in thepehemral environment [18:42] + what it boots [18:43] hrmmm... so does that mean it checks that every time the node boots? or still only at commissioning time? [18:43] bladernr: every time, but with newer firmware they have broken stuff [18:43] bladernr: making it non-backwards compat [18:43] which the system can be EFI but it can boot on legacy [18:43] but still require efi [18:43] so we added an ptpion to force the boot method in 2.4 [18:44] yeah, that seems to be the case here, trying to determine if they broke something in firmware (which is likely). [18:45] is it possible to try 2.4? [18:46] maas/next is showing me the same 2.3 that is in xenial-updates [18:47] bladernr: that's strange [18:47] bladernr: and beta1 goes out today [18:47] bladernr: which will have the efi stuff [18:47] bladernr: and they have regressed firmware [18:47] bladernr: but manufacturesrs say they dont change it [18:47] because the ipmi spec has updated as such [18:48] https://pastebin.ubuntu.com/p/BphvFT6bjW/ [18:48] that's what apt-cache is telling me after updating it [18:49] bladernr: ah, you are running xneial ? [18:50] yeah [18:50] bladernr: 2.4 is bionic+ [18:50] ugh... [18:55] MAAS enlists gives sda: unknown partition level. Any help on how to resolve? [18:55] level=table [19:48] Bug #1760191 opened: maas-ipmi-autodetect-tool fails during enlistment for cavium thunder 88xx crb v2 [20:03] Bug #1760191 changed: maas-ipmi-autodetect-tool fails during enlistment for cavium thunder 88xx crb v2 [20:09] Bug #1760191 opened: maas-ipmi-autodetect-tool fails during enlistment for cavium thunder 88xx crb v2