Topic: SME7.0 RC1 Crashing too often

[seabro]

Hi,

My SME 7.0RC1 has been installed for a month and has crashed twice.

Where can I look to see what is causing it?  Client is moaning at me.  My 6.01 server stay up for 6 months no probs.  I told him I would downgrade him to 6.01 to stop the crashing but that doesnt sound like the proper way forward?

When it crashes, the root console is still there and the server looks ok but you cannot ping it, you cannot get pop mail and you cannot get webmail or even the primary web page.

There are only about  4 users using it for file storage and email.

It is running on a P4/2.8/512 PC.


Where should I begin to troubleshoot?  

Thanks,

seabro

[psoren]

Hi,

My SME 7.0RC1 has been installed for a month and has crashed twice.

Where can I look to see what is causing it?  Client is moaning at me.  My 6.01 server stay up for 6 months no probs.  I told him I would downgrade him to 6.01 to stop the crashing but that doesnt sound like the proper way forward?

When it crashes, the root console is still there and the server looks ok but you cannot ping it, you cannot get pop mail and you cannot get webmail or even the primary web page.

There are only about  4 users using it for file storage and email.

It is running on a P4/2.8/512 PC.


Where should I begin to troubleshoot?  

Thanks,

seabro

You could begin by upgrading to RC3. It's still not a final release so there can be bugs.
If the problem is still there look in /var/log/messages and then the good old word from Charlie: Please report bugs in the bugtracker.

Per

[RedBeard]

Sounds like a hardware problem to me.  Check /var/log/messages, look for eth0 or eth1 going down.  Seem like NIC is the most likely candidate since everything looks ok from the console.  I have had this happen to me.  

You didn't say if you could ping it from the LAN or WAN when it "crashes".  So the next time it is down see if you can ping it from the LAN or WAN side.  If one works and the other doesn't replace the NIC on the side that is not working.  

If both are down it is more likely to be something else besides the NIC, but I have had multiple NICs fail at once.  I have also had a bad NIC in one of the workstations cause problems with the network.  It was flooding the network with garbage.  You could run ‘tcpdump -ieth0’ to check to see if the internal network is being flooded with bad packets.


I would also run 'service  --status-all|more' (hit the space key to advance to the next screen) to see if any services aren’t running that should be.

Run ‘top’ from the command line and see if any zombied processes are shown near the top.  This means that something has crashed and gotten hung up.  If so, run ‘ps –aux|grep defunct’ to see if any processes have crashed and gotten hung up.  This will help you narrow the problem down.  

Someone with more experience might have some better pointers, but this is how I would start narrowing it down.

Good Luck

[CharlieBrady]

Where should I begin to troubleshoot?  

All problems with any release should be reported via the Bug Tracker.

[smeghead]

.. maybe check the mobo bios revision in case there is a new bios with some tweaks/bugfixes that could be relevant eg: bus timings changes that could affect the NIC's??

I would also swap out the PSU as a low 5V rail can create all sorta havoc.

HTH

[smeghead]

.. maybe check the mobo bios revision in case there is a new bios with some tweaks/bugfixes that could be relevant eg: bus timings changes that could affect the NIC's??

I would also swap out the PSU as a low 5V rail can create all sorta havoc.

HTH