[03:30] Bug #1677382 changed: [2.1.5] Failure to PXE boot Cavium crb-1s Servers - PXE-E99: Unexpected network error. Checkpoint D9 ERROR: Boot option loading failed [03:30] Checkpoint AD [03:42] Bug #1677382 opened: [2.1.5] Failure to PXE boot Cavium crb-1s Servers - PXE-E99: Unexpected network error. Checkpoint D9 ERROR: Boot option loading failed Checkpoint AD [03:45] Bug #1677382 changed: [2.1.5] Failure to PXE boot Cavium crb-1s Servers - PXE-E99: Unexpected network error. Checkpoint D9 ERROR: Boot option loading failed [03:45] Checkpoint AD [12:53] hello, how can i force the 'noatime' to be set on the root partition? [12:53] i tried setting it as a mount_option, but that doesn't seem to have an effect [12:58] BlackDex: try this: maas machine get-curtin-config " and pastebinit [12:58] BlackDex: and we will see if the mount options are eing sent [13:12] roaksoax: https://paste.ubuntu.com/24511084/ [13:12] Seems like it [13:20] BlackDex: could be a curtin bug then [13:22] meh [13:23] well ill leave it there for now. I'm going to add this machine now with juju, lets see what it does [13:25] cool [13:26] that machine failed me during a deployment. As it whas only a nova-compute/ceph-osd unit i removed it, tested the mount option, and i'm now going to add it again. Ill let you know if it added the value or not, or if that i can see something in the logs :) [13:26] sounds good [13:39] roaksoax: Nope, didn't work. No entry in any log on maas or the system it self about the mount options [13:42] BlackDex: rharper ^^ [15:27] roaksoax: mpontillo: was it one of you who said you would verify if user_data metadata pass through worked for Custom images? [15:29] xygnal: it should work for custom images... we dont really differentiate the metadata service [15:30] roaksoax ty. [15:30] xygnal: my apologies if I said I would; I had to make some last minute travel arrangements so it may have slipped my mind [15:30] it's been a crazy week so I honestly don't recall =) [16:56] mpontillo: its okay. testing out the code in trunk for fixing our no dhcp host map issue. [16:56] but it does not seem to have taken affect [16:57] mpontillo : what serices would need to be restarted> === daniel1 is now known as Odd_Bloke [17:17] xygnal: that isn't yet fixed. I had some code under development for that but it wasn't fully tested yet. I'm going to try to work on that this afternoon though. [17:18] mpontillo ok ty. sorry to jump the gun. pressure here to test working fix. [17:18] xygnal: restarting maas-rackd is a workaround since that will force the hostmaps to regenerate [17:19] xygnal: very annoying bug though; I'm sorry it's impacting you. [17:20] mpontillo: yep. the primary problem is that we have a post-script that converts from hostmap hcp ip to static local file [17:20] mpontillo and if the node comes up with the wrong IP after deploy, it will set the wrong IP for its static file =p [17:20] thus, I can't test that until the bug is fxed [17:21] mpontillo beyond that the retard of maas-rackd works just fine [17:21] restart* [17:21] xygnal: ouch. well, maybe you can change your script to scrape the MAAS API or database? Just a thought [17:21] mpontillo I was thinking about that. had not approached it yet. [17:39] mpontillo:  [17:55] Bug #1688359 opened: Node fails to commission completely. === frankban is now known as frankban|afk [20:55] Bug #1685399 changed: [2.2] MAAS should delete cached apt lists before running apt update