2

I've recently installed a server in colocation, and my server is rebooting every 4.5-5.5 minutes. Regardless of the OS I run, it reboots. I have all ECC memory in the server, so it should correct errors if there is a bad bit in the memory, right? It's weird because it always happens about 4.5-5.5 minutes after bootup. My motherboard is a Supermicro X8DTL-iF. I read on a blog that another person had the problem, and supermicro recommended to do a BIOS update. Is this the right course of action?

Improve this question

3 Answers 3

Reset to default
6

Fixed this problem by disabled the Watchdog function in my BIOS. It was rebooting my server every five minutes for some really weird reason. Oh well, at least no hardware is bad! :)

Improve this answer
4
  • Shouldn't this one be marked as the answer then? :)
    – emtunc
    Jan 16, 2011 at 9:54
  • 6
    Probably because you did not isntall... the driver for it. Without a driver this is waht the watchdoc does. The driver is supposed to reset the timer every period (like 10 seconds). In case the computer freezes, the watchdog reboots. Without timer... ;) the watchdog reboots, too.
    – TomTom
    Jan 16, 2011 at 11:24
  • 1
    @TomTom Bingo. This is a feature, not a bug. The whole point of the watchdog is to reboot your computer if it stops responding. 'Responding' is determined by having a piece of software poking the watchdog hardware at least every five minutes. If you didn't set up the software, the watchdog hardware will trip and reboot the machine, thinking that it has hung.
    – Ian Howson
    Mar 20, 2014 at 6:09
  • But the watchdog is usually enabled by the driver, so this shouldn't happen. Without the driver, the watchdog sits idle and does nothing.
    – Jonathon Reinhart
    Sep 30, 2019 at 5:51
0

Consistent lockups/reboots are often caused by heat problems. Make sure all your fans are spinning and nothing is blocking the airflow.

After that, certainly try the bios update, and plan on swapping out the memory too - you could have bad memory that's causing the reboots. ECC won't fix that.

Improve this answer
2
  • Temperatures aren't a problem. The sensors are showing normal amounts of heat. The fans are spinning at a normal speed too. I'll take a trip (1.5hr drive.. urgh) down to the datacenter to do a bios update. I'll bring some extra RAM down to test with. I've emailed supermicro too, but I've heard their support sucks.
    – Andy
    Jan 16, 2011 at 9:09
  • You could always double check that heat is not the problem by running several CPU intensive tasks to see if it still reboots in the 4.5-5min period
    – emtunc
    Jan 16, 2011 at 9:53
0

An anonymous user tried to edit another answer to include:

I had the exactly same issues with that motherboard and found this old posting. Well it's an old MB too, i guess. If anyone else is having the same problem of continious reboots every 4-5 min, please make sure not just to disable the watchdog function in the BIOS, but also remove the jumper like it's discribed in the MB manual.

This little jumper caused the multiple reboots in my case. Now everything is working fine for me.

If anyone still uses this old MB, like me, i hope i was able to help you with this addition to this already helpfull post.

MB-Manual take a look at p. 45.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .