Engine Polling Failure after polling/sql split

Hi guys

This is my database server:

Web Server	VITALSTATISTIX
Operating System	Windows 2003 Standard Edition
OS Version	5.2.3790
Service Pack	2.0

Database Server	192.168.10.67
Catalog	SolarwindsData
Database Engine	Microsoft SQL Server
SQL Version	Microsoft SQL Server 2005 - 9.00.4035.00 (X64) Nov 24 2008 16:17:31 Copyright (c) 1988-2005 Microsoft Corporation Standard Edition (64-bit) on Windows NT 6.0 (Build 6002: Service Pack 2)

Web Server VITALSTATISTIX

Operating System Windows 2003 Standard Edition

OS Version 5.2.3790

Service Pack 2.0

Database Server 10.1.1.2 (fake)

Catalog SolarwindsData

Database Engine Microsoft SQL Server

SQL Version Microsoft SQL Server 2005 - 9.00.4035.00 (X64)

Nov 24 2008 16:17:31

Standard Edition (64-bit) on Windows NT 6.0 (Build 6002: Service Pack 2)

and here is my polling engine details:

Polling Engine on VITALSTATISTIX

Engine Status Polling Engine Down

Type of Polling Engine Primary

Polling Engine Version Engine Version 9.1.0 - SolarWinds Orion Network Performance Monitor 9.1 SP5

IP Address 10.1.1.1 (fake)

Last Restart 23-May-10 05:04 PM

Last Database Sync 33 seconds ago

Last Fail-Over Never

Elements 8979 Elements

So, the problem here is that when you check under the admin settings on the webgui interface the polling engine status goes down/up/down within a matter of seconds, this causes false positives on our side and totally messes up the SLA agreements we have with our hotspots.

Everything was working fine (amount of elements are rougly the same as before the split) when the NPM and the SQL was running on the same box, but after the split this started happening.

Do you have any ideas where to start looking?

Find more posts tagged with

polling_engine_down_monitoring_false_positives

Accepted answers

All comments

nding

Hi Warlynx,

You might want to check the performance of the SQL server that you have moved to: www.solarwinds.com/.../gaps.htm

And also, can you share the RAID configuration of the SQL server, whether it is on SAN/NAS, or that it is even sitting in a Virtual Environment?

Warlynx

Hi Warlynx,
You might want to check the performance of the SQL server that you have moved to: www.solarwinds.com/.../gaps.htm
And also, can you share the RAID configuration of the SQL server, whether it is on SAN/NAS, or that it is even sitting in a Virtual Environment?

Hi nding,

I checked you link out, everything is fine on this side, polling is standing at 99.18% so the server is coping with the polling, and we didn't experience any gaps in the data.

I also checked out the Polls per Second and changed it to a lower value to that of which the machine suggest.

Max Polls = 120 (Machine says it can handle 179)
Max Stats = 40 (Machines says it can handle 45)

I had to stop the Netperfmon service and restart it again, but now when I am opening the NPM System Manager from the box itself, it takes ages just to load up, not talking about baseline calculation here.

The loading screen form NPM took 10min

nding

Hi Warlynx,

It is normal, after the restart of the NetPerfMonService.exe. It needs to populate the information that it needs, not mentioning about baseline calculation.

99.18% is quite ok, but I would say that it will be slightly low...

You haven't answered me on "And also, can you share the RAID configuration of the SQL server, whether it is on SAN/NAS, or that it is even sitting in a Virtual Environment?"

And you might want to try posting your perfmon results: [Start | Run | perfmon]

Warlynx

Hi nding,

Ah, sorry about that, raid happens on the box itself. No SAN/NAS setup.

Here is the result for the perfmon