possibly broken kubernetes configuration

[damianio]

my console is spammed with the following message: IPVS RR: TCP 172.17.105.108.5432 - no destination available. is this ip from the kubernetes cluster that is setup for truenas apps because i can't find records of this ip on the forums (possibly everyone gets a uniquely generated ip for their own clusters?)

problem: apps do not start and are stuck on "deploying" which could indicate that the error spam in my console is because kubernetes is trying to start my apps.
what i did before this error that could have caused it: i don't remember.. i found out after i noticed that my truenas server had a kernel panic and wasn't responding.
i upgraded to the latest truenas scale not too long ago but everything seemed fine for a while.


also i was trying to copy the console error message via a terminal shell from web gui but my web gui didn't load and i got another kernel panic. could this be some buffer overflow type of situation because of the error spam in console? because if it is something else then i'm even more conserned, i can't provide information about the kernel panic because i don't know if and where a log is saved automatically when the system hangs

i've provided the log for the system before it kernel panics. the kernel panic only happens after a long while and isn't reproducable as it is random. but the error just persists constantly and i find not being able to use my truenas apps kinda sad so if someone can please help me, i'd be very grateful

system information is also included in the log but to ensure i'm not breaking forum policies it's here as well:

• Motherboard make and model: supermicro x9sae
• CPU make and model: i7-2600
• RAM quantity: 23 GiB
• Hard drives, quantity, model numbers, and RAID configuration, including boot drives: 2 mirrored ST4000VX013-2XG104 4tb drives with an 80gb intel 320 ssd as boot drive
• Hard disk controllers: not sure because the supermicro server has this hot swap bay that can do hardware raid but it does have a cable for each individual hard drive that just goes to the onboard sata sata controller: marvel group [88se912x]
• Network cards: 2 onboard gigabit intel nics +1 realtek network card " rtl8111/8168/8411 " assigned to a vm via bridge's instead of pci passthrough
• OS Version:TrueNAS-SCALE-22.12.0
• Product:C7P67

 

[FrankWard]

I'm using a SuperMicro X10SLM+-F and experienced the same behavior today. In an effort to get a VM to see NAS, I opened the TN console and followed the instructions:

1. Remove the IP from eth0
2. Create a bridge br0
3. Bind eth0 to br0
4. Assign IP to br0
5. Reboot

On reboot I started to see this message IPVS RR: TCP 172.17.105.108.5432 - no destination available on the console. I noticed the FireFly app stuck on deploying, but my other apps were fine. VMs were not started and trying to start them gave a CallError [EFAULT] error creating macvtap interface macvtap2@eno1 (00:a0:98:1d:7f:24): Device or resource busy.

• I stopped the Firefly app and the messages ceased to show on the console. I restarted the app and it worked without any messages in the console.
• I then had to set the NIC to br0 for each VM, which allowed them to start. The VM could now see the NAS.

So far, after these small changes, everything is working as intended. This information is to assist others that may run into this cryptic issue and search before investigating too deeply.

 

[RJ07]

I solved this today... I had PiHole installed and old config. When change IP's, it went rogue.
Deleted PiHole, now all is normal.