PERFCOUNT issues in 5.11.1?

Hi,

I upgraded to Serverscheck 5.11.1 this afternoon- and since upgrading and rebooting have had some strange alerts from the PERFCOUNT checks I run.



For example;

I monitor DNS requests and responses to our main DNS server. Serverscheck report these to be over 300 a second.

I monitor the same counters locally on the server using Performance Monitor and it peaks at around 40 a second...

Serverscheck is alerting this to be much more.



I then checked my other PERFCOUNT alerts, which monitor Network InterfaceBytes Total per second on several servers. Since the upgrade this has jumped up too, and now reports 300T a second!!

(This was around 30mb a sec using version 5.10.5).



Is this a bug in the new version? Is anyone else having these issues?



Thanks

I checked the development log and nothing was changed on the PERFCOUNT check (version file is the same as in 5.10.5).



The component doing it is from January 2005. Try following (in order to Debug).



Create a file with notepad in your agents subfolder and called performance.conf with following content:

[performance.conf]

Server=servername

Username=Administrator

Password=password

Counter=counter



Via the command prompt, launch following in the agents subfolder:

perfcount_check performance



It should create a file called performance.log

What is the content of that file?

Changed settings to UK English and a test with following counter:

LogicalDisk(C:)Free Megabytes

which reported the same value as the DriveSpace check:

1240



To date no issues on the Perfcount have been reported with release 5.11.1



(I apologize to say this): we have had situations were conflicts were caused due to incomplete/errors during the upgrade process. Issues have been fixed by uninstalling/installing the software. Now you don't need to go that far.

Simply do following:

delete the monitoring_manager.exe and monitoring_rule.exe and download it from:

http://members.chello.be/cr50591/update/monitoring_manager.exe

http://members.chello.be/cr50591/update/monitoring_rule.exe

Good Morning.

As you advised, I replaced the 2 files with the 2 from the links (monitoring_manager.exe and monitoring_rule.exe)



Restarted server, but check values still report the same.



I've then created the performance.conf as advised and below are the entries from the file (after I run it serveral times every few minutes).



"37.1"

"27"

"33.4"

"30.9"

"33.3" - These seem to reflect the values reported from Perf. Monitor - but NOT from serverscheck.



When I run the 'Test Settings' from serverscheck the value is always reported as around 300 (10 times higher then Perf. Monitor)



Also, the value reported from my Network Card checks (Network Interface 3Com Bytes Total/sec) is reporting

304281806967146 - this is also about 10 times more then it was before the upgrade.



What should I do next?



Thanks

I am going to dig into this a bit deeper but it seems to me that the value 30.9 is changed to 309 ...



I am guessing that the "." might be the issue.



Let me check it out right now and I will get back to you within the hour.

We tracked it down. The error slipped apparently in due to changes to non English based platforms. Someone is working on a fix as we speak and it should be ready for you to download within the hour (monitoring_rule.exe component)

You can download the new build from

http://www.serverscheck.com/files/monitoring_rule.zip

EXCELLENT!



Since replacing that file, all the PERFCOUNT checks seem to have returned to believable levels, and show more what Performance Monitor displays.



I'll keep checking it for the rest of the day, but for the past hour or so it seems to have been fixed.



Thanks for resolving it so quickly.

Glad to hear that it is fixed and thank you for your help in reporting and locating the issue.