Which log to tail ?

[ozgurerdogan]

My proxmox lately hangs or loose connections and I have to reset node to become online. But I can not see any failure in logs. Actually not sure which log to look at. Can you please guide ? I look at all logs files but can not see anything.

 

[sb-jw]

I look at all logs files but can not see anything.

If you do so, there is no other hidden log files if you looked all logs under /var/log.

Yoi should check the monitor output with a KVM / IPMI.

 

[ozgurerdogan]

But anything I would see via kvm, must be also in log files? I am thinking of moving to a new server.

 

[sb-jw]

If the Disks cant write anything at this moment, you can't find it in the logs.

You can use kernel dump to check what's going on. But afaik there might be some problems with using it in combination with PVE.

Do you already checked your hardware? You only lost network?

 

[ozgurerdogan]

Server is located at hetzner and they say it will take around 14 hours to check hardware which is too long for me. I loose all connection to server ssh, ping etc. and I send a hardware reset from panel and it becomes online again.

Only things I found in logs are; but in two months, this happened three time only.

Jun  2 12:15:33 hs1 kernel: [4508657.319058] e1000e 0000:00:1f.6 enp0s31f6: Detected Hardware Unit Hang:
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   TDH                  <0>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   TDT                  <4>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   next_to_use          <4>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   next_to_clean        <0>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] buffer_info[next_to_clean]:
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   time_stamp           <1432df892>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   next_to_watch        <0>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   jiffies              <1432e0238>
Jun  2 12:15:33 hs1 kernel: [4508657.319058]   next_to_watch.status <0>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] MAC Status             <80083>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] PHY Status             <796d>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] PHY 1000BASE-T Status  <3800>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] PHY Extended Status    <3000>
Jun  2 12:15:33 hs1 kernel: [4508657.319058] PCI Status             <10>
Jun  2 12:15:33 hs1 kernel: [4508657.350813] e1000e 0000:00:1f.6 enp0s31f6: Reset adapter unexpectedly

 

[sb-jw]

Is the NIC onboard or AIC? Do you have a second LAN port? If yes, can you change the cable to the other port?

 

[ozgurerdogan]

Ok I think I found the reason. It seems to be bug with OS.
See: https://wiki.hetzner.de/index.php/Low_performance_with_Intel_i218/i219_NIC/en

 

[oguz]

hi,

can you mark this thread [SOLVED] if the workaround in the link solves your issue? (you need to advanced edit your first post)

 

[ozgurerdogan]

not yet as I need to wait for couple of days. But so far so good..

 

[oguz]

not yet as I need to wait for couple of days. But so far so good..

update?

 

[ozgurerdogan]

How do I mark as Solved ?

 

[oguz]

edit your first post with advanced edit or find thread tools