Confusing server metrics

[sknopp]

Hi,

I want to write a Python-Program for monitoring the PTC Integrity server. Since i haven't found a way to read the gc.out file directly

I settled for using <im diag --diag=metrics --target=server> but I am a little confused regrading the returned information. For each part of the

Heap there is a "current" and "peak" value: is this current in use/ currently available and is peak the total storage or the peak of the

used storage in the past? I am also interested in knowing if the multiple values for the "average response time" were taken

just for this command/result or were they collected over a longer period and so only the last value gives information about the

current state?

I hope that someone has already some experience with this kind of output and can help. Thanks!!

[JoeBartlett · Accepted solution]

Hi Stephan,

Have you considered using PSM to monitor your Integrity Server? It allows you to see current and historical graphs of heap usage, GC activity, track user operations and performance, and much more. For further details see the PSM page here. If you're really set on monitoring in a script, you could try the heapstats diagnostic. This will give you output of the amount of that allocation which is used, the current amount allocated for the heap, and the maximum heap setting. I have prettied-up the output a bit to make it simpler to understand but the 3 values you get are in this order:

im diag --diag=heapstats

Amount of Allocation Used	Current Allocation	Max Heap
321489096   (320MB)	682522424 (681MB)	1004011520 (1024MB)

Regarding your question about the Response Times, this is taken from the last 100 operations to make a call to the database and simply shows the average of those values.

[sknopp]

Ok, thank you!