SiteScope Issues

[bilal,aman]

Hi Guys

We are facing issues with lot of false alerts being triggered from 2 diffrent  SiteScope  Servers

We tried to  deploy the same monitors on the other (3rd)Sitescope and the issue did not replicate on this

what can be the cause of this ?

Problematic SiteScope server are running on Windows 2008  and the SiteScope server which is working fine is running on Windows 2003

I also noticed Java vrsdion on [roblematic server is (1.6.0.26) and the Sitescope which is working is running on (1.6.0.18)

Is this due to Java version , I had read about an artiiclae on siteScope page on Facebook about issues with some range of Java version

Can the alert cause because of this ?

Regards

Bilal Aman

[Ben.Shirley]

What is the Sitescope version?

Define 'false alerts'

Which monitors are these from, any particular one?

[bilal,aman]

Ben

The affected monitors are Service Monitor which is used to check Heartbeat of a server

I say these are false because the servers are not down at that particular time when the alert comes in

We keep getting Error Alerts followed by Good then again followed by Error and Good through out  the day

These servers are monitored using SSH  and affected servers are Windows NT Servers

Looking at the Run_Monitor Logs it says unable to exeecute the command  "perfex -timeout 120 -pc =230"

When ever the monitors are in error state  we wil not be able to execute the above command through Putty , and when the monitor state is good , this command works fine

SiteScope Version is 10.14 on all 3 servers

Regards

Bilal Aman

[tibornad]

i don't have a SiteScope system in front of me. so i can not give you any detailed help here. but i could think of a problem with the SSH session being established between the SiteScope server and the target machine. do you have the chance to switch to WMI just in order so see if this improves the situation?

also are you opening a large number of SSH connections at the same time? SiteScope has a bunch of SSH related options in General Preferences you could look at. i think there is also some sort of setting which specifiies the number of concurent SSH sessions you can have. i know this is not ver specific. also you can check if you can still connect to the remote server under Remotes once the monitor is in error state. what does the monitor return in the SiteScopeXXXXX log file as result?

[bilal,aman]

In the Run_Monitor Logs i see this error

Error: failed to execute the command: perfex -timeout 120  -pc =230 on remote and when the monitors are i error i have tried executing this command on Putty and it does not execute (only when the monitor is in error)

Regards

Bilal Aman

[raok]

Hello Bilal,

We are facing similar issues. Service monitors are throwing a lot of false alerts. 

the error i get to see in daily log is :

05:09:54 01/07/2013 error MSSQLSERVERinstanceon Service: SQL Server (MSSQLSERVER) on unavailable error(unable to connect to server) 13:73174 not found (remote monitoring error - unable to connect to server) no data no data 0 0 0 failure

Monitor is working fine now. also since the issue occured sometime ago we do not have information from Runmonitor.

I see that you faced this issue earlier.  Please suggest and advice.

Version : 10.14

Regards,

Kiran Rao

[bilal,aman]

Hi Kiran

The issue what i had reportd here was a different issue and problem was with perfex getting timedout , Isuse was resolved by increasing

1) frequency of the monitor

2) increasing the perfex timeout value

3)Increase the Connection Limit from default 3 to no of monitors  the remote server has.( i increased it upto 10)

3)Increasing NT SSH time out value

But you siad you Connection Method is WMI . so the issue might be different

Regards

Bilal Aman