11:13:09 <TheTim0Nat0r> Does anyone know how to format the script if I want SmartOS to recognize it to be a cloud-init netwerk config? 
11:13:24 <TheTim0Nat0r> Should it be something like this: https://pastebin.com/W08q69CW 
11:13:52 <TheTim0Nat0r> Or should I use other format or keyword?
14:03:13 <neuroserve> how would I implement dns lookups via consul in a native zone? consul dns would be consumable on 127.0.0.1:8600 - do I have to run consul as root in order to let it listen on port 53? or can I re-route dns queries to another port via firewall-rules?
15:29:40 <bahamat> Well, there's a few options.
15:30:03 <bahamat> 1. Make it behave as a standard DNS server and listen on port 53. Hopefully all your other DNS requests work too.
15:30:49 <bahamat> 2. An nss plugin, but you'd have to make that yourself since it doesn't already exist and AFAIK consul doesn't provide one (compiling this for the PI may be difficult as well)
15:31:25 <bahamat> 3. Just use dig (or something similar) and query consul directly without making it a system resolver.
15:40:38 <neuroserve> actually I want to test matrix-synapse connecting to a postgres-patronic cluster - and it would be nice to use consul dns in order ot make matrix synapse always connect to the leader database
15:42:11 <neuroserve> but matrix doesn't like to start anyway: socket.gaierror: [Errno 8] node name or service name not known (although I've added the ip/name to /etc/hosts)
22:14:52 <teutat3s-rec> Hello and happy New Year everybody
22:15:21 <teutat3s-rec> I'm trying to recover our headnode from a mayor outage in one of our Triton DCs
22:16:28 <teutat3s-rec> long story short: false fire alarm went off because of a broken sensor in AC system. Loud alarm noise (90dB) from the alarm before the gas tried to extinguish "the fire" damaged various spinning disks
22:16:57 <teutat3s-rec> Offsite backups ftw
22:17:38 <teutat3s-rec> I'm happy I got this far, but now binder seems to want some extra care
22:18:00 <teutat3s-rec> Maybe it's not happy because it was running clustered before and the other two are not up yet?
22:18:39 <teutat3s-rec> Don't know if I should just try to get those two CNs up or if a better way would be to "de-cluster" binder before proceeding?
22:19:01 <teutat3s-rec> I'm not even sure if CNs will boot without a fully functioning headnode?
22:21:22 <teutat3s-rec> in the binder zone, `echo stat | nc 127.0.01 2181; echo ""` shows: This Zookeeper instance is not currently serving requests
22:29:49 <teutat3s-rec> `ruok` in the same command answers `imok`
22:34:37 <teutat3s-rec> Uh oh, this sounds very similar https://smartdatacenter.topicbox.com/groups/sdc-discuss/Tf6fd2fc012e9d9f2-M90c26e43a0be8b30f10ddd9d/datacenter-can-not-boot-up
22:55:44 <teutat3s-rec> just for some more background: all servers were shut down by a technician who panicked and tried to "save" the servers
22:56:55 <teutat3s-rec> So now I'd like to get everything up and running again after doing a backup (ZFS snapshots) restore on the headnode (which seems to have suffered disk damage)
22:57:55 <teutat3s-rec> All CNs are still down, will try to see if I can boot one via PXE and limb onwards