Server Hung State -What is the monitoring Solution ?

[Issac08]

Hi All,

 

 What is solution for monitoring server hung state proactive ? (Mostly windows )Still now i cant find any solutions .Server is responding to icmp and all services are running fine and there is no spike in performance too . How do we overcome this ? Request to share the experience on this

[Christophe_Sperandio]

What do you mean by "hung state"? What is the observable state? What happens? If services are still up, that means service manager is still responding, and ICMP up means the network interface is still responding...

[Issac08]

Sometimes the windows will be in loading state ,but user cant login the server. 

[Yu Ishitani]

i am not perfect and how about net_connect based port monitoring over tcp 3389 for windows.

[Issac08]

No Yu, It doesn't help because ping and services will be active only.

[RowanCollis]

There might be something in the event log, so that is worth monitoring with ntevl.

Terminal service/remote desktop service is running with ntservices?

C: drive disk usage with cdm to check for zero space.

When you find what goes wrong when this happens, you might be in a better position to setup the monitoring.

Hope this helps

[Garin Walsh]

The approach we take is to monitor for the lack of something expected happening in a QoS table. 

 

Essentially it's a fancy heartbeat and there's a couple ways of going about it.

 

The approach is to use logmon (or one of the other probes) to run a script/test that exercises one of the functions that your end user would perceive as "hanging". Change in log file size, disk space free measure, URL get, etc. Or even just a batch file that posts a QoS record periodically.

 

Then on your central hub, have a scheduled job query that QoS table for robots that haven't reported a QoS update recently.

 

It's not especially elegant but it does work.

 

-Garin 

[Issac08]

Thanks Garin for Suggestion !!  Again we need to do some custom solution where it must be addressed as default capability .

[Christophe_Sperandio]

Another way could be to have a probe which issues a certain callback on a couple of probes installed on the robot, to see if you get a result. If you don't, that might mean it's frozen.

[Marco Ippati]

Hi Issac08 Praveen Venkatesan, Is your question answered?