FreeBSD rebooting every 5 or 6 minutes

C

cybercoke

Hi guys,

Since yesterday one of my FreeBSD 8.1-RELEASE machines start to automatically reboot after some minutes of uptime.

Looking at the logs I just see normal info except for these lines in bold:

Code: 
Mar 12 08:40:43  kernel: WARNING: / was not properly dismounted
Mar 12 08:40:43  kernel: WARNING: /tmp was not properly dismounted
Mar 12 08:40:43  kernel: WARNING: /usr was not properly dismounted
Mar 12 08:40:43  kernel: WARNING: /usr/local/samba was not properly dismounted
Mar 12 08:40:43  kernel: ffs_snapshot_mount: old format snapshot inode 17
Mar 12 08:40:43  kernel: WARNING: /usr/local/squid was not properly dismounted
Mar 12 08:40:43  kernel: WARNING: /var was not properly dismounted
Mar 12 08:40:43  kernel: /var: mount pending error: blocks 192 files 0
Mar 12 08:40:45  root: /etc/rc: WARNING: failed to start vtunclient
Mar 12 08:40:57 fwdentalblumenau squid[1375]: Squid Parent: child process 1377 started
Mar 12 08:40:57 fwdentalblumenau squid[1375]: Squid Parent: child process 1377 exited with status 0
Mar 12 08:41:02 fwdentalblumenau kernel: rl0: promiscuous mode enabled
Mar 12 08:41:07 fwdentalblumenau kernel: :
Mar 12 08:41:09 fwdentalblumenau kernel: tap0: Ethernet address: 00:bd:af:5d:02:00
Mar 12 08:41:09 fwdentalblumenau kernel: tap0: link state changed to UP
[B]Mar 12 08:41:59 fwdentalblumenau kernel: A[/B]
Mar 12 08:42:11  fsck: /dev/ad4s1d: Reclaimed: 0 directories, 1 files, 0 fragments
Mar 12 08:42:11  fsck: /dev/ad4s1d: 212 files, 10 used, 1013005 free (61 frags, 126618 blocks, 0.0% fragmentation)
[B]Mar 12 08:43:25  kernel: 1[/B]
Mar 12 08:43:35  fsck: /dev/ad4s1g: 260269 files, 1528750 used, 13702568 free (32144 frags, 1708803 blocks, 0.2% fragmentation)
[B]Mar 12 08:46:31 kernel: :[/B]

After this the error is always like
Code: 
Kernel: 1
or
Code: 
Kernel:2
or
Code: 
kernel:p
or some like this, it hangs an reboot.

What should I do?

Thank you.
 
cybercoke said:
Since yesterday one of my FreeBSD 8.1-RELEASE machines start to automatically reboot after some minutes of uptime.
Click to expand...
FreeBSD 8.1 has been end-of-life since July 2012 and is not supported anymore.
 
That looks like a hardware problem but there's no way to tell for sure since you're giving us so little information.
 
The hardware is:

Code: 
CPU: Pentium(R) Dual-Core  CPU      E5400  @ 2.70GHz (2693.26-MHz 686-class CPU)
Origin = "GenuineIntel"  Id = 0x1067a  Family = 6  Model = 17  Stepping = 10
real memory  = 1073741824 (1024 MB)
avail memory = 1026699264 (979 MB)

When I look at the hardware info in the message log, I see this:

Code: 
kernel: ACPI Warning: Incorrect checksum in table [OEMB] - 0xC1, should be 0xB4 (20100331/tbutils-354)

It could be something related to the problem?

Thank you.
 
You need to upgrade to a supported FreeBSD release, like 8.3-RELEASE, which is supported until April 2014. We like to invest the efforts on this forum into supported versions.
 
Hi DuchDaemon, I will, but now I need to solve this first because this server is miles away!

Please help me.

Thank you.
 
If nothing changed in the software it's most likely a hardware issue.
 
Hi guys, I changed the memory and the power supply. Same problem.

I caught a message like this in one of the reboots:

Code: 
panic fna pact u2: bad block

It was after the background check started at boot.

I'm trying to check the HDD with an external program.

Thank you for help.
 
Hi,

Main question is: rebooting or crashing? Do you have dumpdev enabled? Do you see core files in /var/crash (or whatever your dumpdir is set to)?

If the server is rebooting, you may have a problem with PSU or the actual power outlet. I'm not sure if "current staggering" is a good translation to English; I tried to google around. It's when the voltage in outlet is not always 230V (or the normal voltage as agreed in your country) but is moving up and down more or less randomly. Worth checking and maybe plugging into different circuit if possible.

As the crash/reboot occurs quite often, you can monitor system with top, iostat to see if something suspicious comes up.

It might be worth disabling any unneeded system services (tunnels, taps, squid and basically all except SSH and syslog).

But those syslog messages of yours are weird ..

EDIT: I did google around for that weird messages in the syslog of yours. Are you using IPFW on that server? Some results show it may come from that. If so, maybe also keep an eye on traffic coming to the server.
 
cybercoke said:
When I look at the hardware info in the message log, I see this:

Code: 
kernel: ACPI Warning: Incorrect checksum in table [OEMB] - 0xC1, should be 0xB4 (20100331/tbutils-354)

It could be something related to the problem?
Click to expand...
No, it just means that the BIOS has a minor problem. Windows doesn't report those anywhere users will notice them, so the BIOS / motherboard manufacturers don't have a real incentive to fix these cosmetic issues.

The first thing to do if you want to get rid of this is to check to see if there's a newer BIOS for your motherboard, and if there is, install it and see if the problem is fixed.

If the message still shows up with the latest BIOS, you can try getting the manufacturer to fix it. You may or may not be successful - I've been able to get SuperMicro to fix some of them, but other manufacturers may not be so willing.
 
Hi @Terry_Kennedy,

Thank you for your answer.

Hi @matoatlantis, the machine is crashing and then rebooting like something gave a command to do that.

I can't see more information because the machine is in another city and the connection breaks every two minutes. I'll try to bring it to me to make the tests.

Thank you.
 
Last edited by a moderator:
cybercoke said:
the machine is crashing and then rebooting like something gave a command to do that.
Click to expand...

Then it is going into a panic (i.e., autoreboot when something goes wrong). This reinforces the fact that there could be an hardware problem.
 
You must log in or register to reply here.
Back
Top