Yeah I saw that too and was trying to figure out how to get that to work. It would be sweet to be able to modify a docker-compose file to match MikroTik standards, upload it, then deploy without having to do it the more manual way via /container.
Hi,
I have 1 Proxmox VE Host. It has 3 QEMU VMs and 3 LSC Containers, one of which is Corebit.
Corebit takes 9.25 seconds to poll this host using API Authentication, so I have to set 10 seconds polling time as anything less gives overruns.
I would like to Poll network devices more often than 10 seconds (5 seconds).
1: Is it normal for a nearly ten second poll for Proxmox VE?
2: Can we have a feature to set custom polling for individual devices?
3: Am I doing something wrong withe the proxmox host?
Running corebit version build 159 on Debian LXC.
I have 5 Proxmox servers + 280 mixed devices, and my polling cycle takes 11 seconds.
But even though they are running in parallel, the cycle depends on all devices getting done, so the interval is dependent of the worst device. (Aslo devices that are down, takes longer to poll, as it then takes the time in “Probe Timeout”.
I will try do add logging for porobe times, to see which devices is the slow ones, and then look into making custom polling (If possible).
FWIW, see
I mention this since one nice thing about The Dude is it's essentially a setting in RouterOS. So being able to run CoreBit, easily, on stuff like RB5009 etc be handy.
Also convenient for deploying a custom group of containers as /apps to your own gear (or that of your customers) without having to wait for MikroTik approval by way of a new software release.
I have added probe time in Build 161
This is one of my Proxmox servers. (I use API key to retrieve data).
Good feature, thanks.
My Proxmox host is a Supermicro server of reasonable spec, and I dont think the slowness is there. I’m starting to suspect my LXC for Corebit. Although it shows system load of 0.75 and 65% RAM use. 2GB Ram and 3Ghz 4 CPU Cores
Any other things I can look through?
I notice that the polling time is present for Proxmox and SNMP and Ping but not for Mikrotik via API device. Is that intentional? I can open a bug on github if not.
All the best,
Alex
I have researched some, and it seems that using login to the proxmox rather than an API user is rather slow. Have you tried to switch to API key?.
Hi,
I am actually using API token.
Which version Proxmox are you using? I have seen a few comments that v9.1.x can be slower than 8.x proxmox
I use 9.0.17
Try the api in Postman, or Curl to see if host responds quick. That might help finding the cause.






