Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I've got a pretty strange problem.

I'm a sysadmin and manage dozens of Xeon servers but one of them (with a low user load) is behaving very strangely.

Server specs:

CPU: Intel Xeon E3 1230 V2
MB: Gigabyte Z77N-WIFI, Bios Version F2 (08/13)
RAM: 16GB DDR3 Dualchannel NON-ECC (2x Kingston 99U5471-037.A00LF)
OS: Windows Server 2012 R2 Standard

This Server hosts two Hyper-V machines. One is Linux Debian and the other Windows 7

The problem: The server freezes randomly every hour or so for 5-40 seconds. By freeze I mean no more IO comes from the server, RDP connections are reset, remote SSH connections from the VMs are reset and all active connections get timed out.

When the server is back online I can see in the Task Manager that the CPU dropped to 0% while frozen.

The event log never has any entries from the freezes and afterwards no error messages from any service or system are logged.

What I can say about the freezes:

  • There is no pattern in the freezes
  • The freezes occur equally random with high or low CPU utilization
  • They occur with low and with high RAM utilization
  • It's not a NIC problem since the CPU drops to zero and this even happens when it's not plugged in
  • The HDDs are in good shape

I made a ping script that records the response times (500 means timeout) and let it run the last day. Ping statistics

Has anyone ever encountered something like that?

share|improve this question
Sorry, not an answer, but I like the way that ping tool looks. How did you generate that image? – jski Jul 18 '14 at 14:32
It's just a quick hack I made for displaying pings with highcharts. here you go: Note: you'll have to change the code in pinger_win.php because it's configured for the german version of CMD – Christian Jul 18 '14 at 14:45
It's sounding like hardware deadlock. Any logs from the Linux machine guest? Its kernel might report something up. What about running with only one VM running? No VMs (to eliminate the hypervisor as the culprit)? – Nathan C Jul 18 '14 at 15:44
Various antivirus products have done this over the years. When they fail / have problems they often lockup whatever subsystem they are intercepting - file access, networking etc without exposing the error or logging any problems. – Brian Jul 18 '14 at 16:52
I'll check the syslog after the next freeze. Maybe I'll find something there It's definitely not the AV since I removed it when I noticed the lags. Didn't change things though – Christian Jul 18 '14 at 19:08
up vote 0 down vote accepted

The problems have stopped. I'm not sure what the final Solution was but it was one of those things:

  • I bought a new PSU since it was underscaled (was 200W is now 400W)
  • I re-installed all Intel chipset drivers
  • I re-installed the nic drivers
  • I doubled the RAM

And now I didn't have a single lag for over a month.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.