OMR running on Mini PC from Amazon suddenly crashing & rebooting after years of great performance #3783
Labels
Comments
|
So, maybe you should check the CPU + RAM usage on that load. We don't have an crystal orb to tell ya, whats gonig on. Could also be the internal storage of the mini pc (reached end of life)... P.S. Using DHCP on the WAN-Interfaces is not recommended, i think and can lead to problems. |
|
Hello @darkman1983 and thank you for your response. as far as I can tell, CPU load seems to be fine... looking at web interface is quite sluggish thou... how can I determine if the storage is dying? I believe it's a SSD memory and I can easily replace that if that's the case, but how I can determine that? also I am getting lots of these messages in the log: I have modified one of the Starlinks to use public IP and set it directly to the interface based on your recommendation... is that the correct way to do it? In Unifi interface, I see these weird really high spikes on the graph: 
|
|
now the mini PC running OMR seems to be crashing constantly it reboots... runs for few seconds and reboots again |
|
If there is no updates of OpenMPTCProuter, backup configuration first.
It's working correctly without unifi ? |
|
it's really weird in a sense that if I switch WANs in Unifi router so OMR is on WAN2 as failover where there is little to no load, it seems to be working fine... once I switch it back to WAN1 where there is about 50 to 100Mbps traffic, it starts to crash almost immediately last night, I have set public IP on Starlink which is Master interface and set it as static in OMR and it worked fine the whole night now, I have done the same thing for all the other WANs and started crashing again so it's really unpredictable... like I said originally, it used to work just fine for about 2 years it's in remote location, so I have to drive there, but yeah... I will take screen with me, since the logs are too short so there is not much info there when it crashes... For example this: that seem like the log from after the reboot after crash but it's missing the beginning |
|
now I have it on WAN2 as failover and works fine... |
|
in the afternoon I will get screen and go there and try to capture whats happening when it's connected as main source of the internet for the network and crashes |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Expected Behavior
to work as it did for 2 years
Current Behavior
I have 6 port Mini PC from Amazon connected to 5 Starlink WANs
[Micro Firewall Appliance, 6 Intel I225-V 2.5GbE NIC Ports Fanless Mini PC, Intel Core i3 10110U Dual Core, Support AES-NI,TPM2.0 1xDP 1xHDMI 1xCOM 6xUSB, 16GB DDR4, 128GB M.2 SSD]
LAN port is supplying bonded internet to UDM-SE thru 2.5Gbps
Initially (2 years ago when I first set it up) there was a problem with the interfaces, but this was resolved by selecting "Force link" on each interface
Since then, this has been working pretty much flawlessly.
Lately I started getting lots of internmitent outages on the Unifi graph... and was not sure why.
And today, the unit became pretty much unusable... it reboots/crashes every minute or less. One can tell by the uptime counter constantly restarting. And the internet not working of course.
There was no change in configuration. It only happens when I set in Unifi to use this as primary WAN.
If I set it as failover, it doesn't crash. I thought it might have something to do with the load. But when I set it as failover, SSH into the terminal and do omr-test-speed, it pulls over 500Mbps sustained for multiple minutes without hiccup. The moment I switch the WANs in Unifi and set it as primary, it starts crashing again. And the load is between 50 and 80 Mbps, so nowhere near the speed test.
here is system log: https://pastebin.com/raw/cML4sH8r
here is kernel log: https://pastebin.com/raw/mygpEXV1
Specifications
The text was updated successfully, but these errors were encountered: