Questions[performance-counters](server)

Anyone else often having a need to correlate multiple log sources such as event log or various log files with performance counters when troubleshooting Windows machines?

For example, we had a case of memory and CPU performance peaking periodically, only to realize it was because of one indexing service was accessing data remotely on server. The way we figure that one is by looking into Security event log and see many login events from that indexing service's account during performance peaking periods.

It took us a while to get to this so I am wondering is there a tool that auto-correlates performance counter deviation with logs appearing during that time? I know there are many log analyzers out there, but they mostly seem to be SIEM oriented and none has performance counter correlation feature.

We have a "SQLServer 2005 SP3 32bit Enterprise Edition" running on a 32 bit Windows 2003 32bit Enterprise Edition 12GB RAM with AWE enabled using RAID5(5 pysical disks).

We tuned AWE to enabled and restart sqlserver this afternoon after work, hope the performance will be better than old time.

But there is something that we are very confused.

On working days, SQLServer has a very bad performance. When we are looking for reasons, we check Windows Performance counter.

Avg. Disk Read Queue Lenght > 140
Avg. Disk Write Queue Length < 1
SQL Server Buffer Cache Hit Ratio > 96%
%Processor Time < 30%
SQL Server Total Server Memory < 1.8G

Obviously, without AWE enabled, SQL Server can use only less than 2G memory. My Question is:

1. why "SQL Server Total server Memory" is less than 2G?I think SQL Server will use all 2G process address space. Does this counter count anything out?
2. we known that sql server is sufferring lack of memory, but why "buffer hit ratio“ is as high as 96?

Any advice is welcomed!

I'm doing performance/load testing of my IIS6/Win2K3 server. I setup perfom by adding performance counters for the specific site I'm interested in (ie. w3wp#7).

The problem is -- I have to manually complete these steps for each test run assuming my load test causes an automatic recycle of the application pool (OR if the test server is restarted or an iisreset happens).

Is there an easier way to setup a Counter Log with the specific counters I want that will automatically monitor the process I'm interested in. Obviously the site stays the same, its just the process that sometimes changes.

Any help would be great for productivity.

Thanks!

jg

I'd like to make sure that I'm measuring IOPS correctly across Windows and Linux servers. We're busy trying to figure out which SAN would best suit our environment and measuring IOPS across our servers is the kind of thing I just can't get wrong. Little bit of procurement paranoia kicking in.

On the Windows servers I'm using the \PhysicalDisk(_Total)\Disk Reads/sec and \PhysicalDisk(_Total)\Disk Writes/sec performance counters. On the Linux servers I run vmstat and am looking at the bi (read) and bo (write) counters. I'm pretty sure that I've got this right and that both sets of counters equate to IOPS.

If these counters are correct then I have a problem. The SAN we're going for can handle about 1600 IOPS. Our two mail filter servers, which run MailScanner, MailWatch and Sophos, are hitting around 1500 IOPS collectively. Our two MySQL server run at about 1600 IOPS collectively. We're way over the 1600 IOPS theoretical limit.

Hence the question, am I measuring correctly?

Ossan

I am wondering specifically about the "Process/% Processor Time" counter. If you set it to an interval of say 10 seconds, are the data points a snapshot of what the CPU utilization is at that 10 second interval, or an average of the utilization over the past 10 seconds? It would seem to be to naturally be the former, not the latter, but there has been some confusion amongst myself and my colleagues and I wanted to clarify.