Image using 2GB memory, can SDF file help determine what is using so much memory?

[Wayne Johnston]

Our server image typically takes 0.5GB RAM, but for some reason in one environment it got up to 2.1GB then failed to allocate more memory.  Can looking at the SDF file help me determine what is using so much memory?  I know of

https://www.instantiations.com/docs/922/wwhelp/wwhimpl/js/html/wwhelp.htm#href=sv/serve114.html

but I would really like to know how many instances there are of each class, and how much memory is consumed by instances of each class - like EsMemoryUseReport.

[Mariano Martinez Peck]

Hi Wayne, 

Quick question, which exact VAST version are you using? 

--
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to va-smalltalk...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/c84fc3b5-1961-415d-ae7d-12f6c9f7f6d4o%40googlegroups.com.

--

Mariano Martinez Peck
Software Engineer, Instantiations Inc.

Email: mp...@instantiations.com

Twitter: https://twitter.com/MartinezPeck

LinkedIn: www.linkedin.com/in/mariano-martinez-peck

Blog: https://marianopeck.wordpress.com/

[Wayne Johnston]

This happens to be 9.2.1.

[Mariano Martinez Peck]

Greetings Wayne, 

There was one leak (in SocketCommunicationsInterface) to our knowledge that was introduced in 9.2.0 and fixed in 9.2.1 and that's why I asked. But since you are in 9.2.1, then forget it. 

I don't think SDF would be much of a help here. But what happens when you get the "failed to allocate more memory"? Does it just exits the VM? Do you have any walkback or vmtrp to share? 

Using EsMemoryUseReport could  be a good idea. I am CCing Adriaan, which is a customer of ours, that as far as I know they were recently hunting a leak and they have a kind of watchdog process that would dump the info of EsMemoryUseReport every X time. You may try that as well. 

The other question is how reproducible is this. You said "in one environment"...so that means that it happens only in that environment but not in others? How frequently does that scenario happen? 

Best, 

Mariano

On Mon, Aug 17, 2020 at 11:06 AM Wayne Johnston <wa...@thejohnstons.com> wrote:

This happens to be 9.2.1.

--
You received this message because you are subscribed to the Google Groups "VA Smalltalk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to va-smalltalk...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/2e810c9f-2500-4af1-a5e1-1192312e3b3co%40googlegroups.com.

[Wayne Johnston]

We got a nice walkback, but it looks very normal except for the high amount of memory usage.  The machine doesn't have much memory available at this point, but it's largely because of our image using so much memory.  We fail to allocate just 4MB.  

Yes, we do a EsMemoryUseReport on a configurable basis, but for some reason it was turned off in this one environment.  It's been turned on in case the problem recurs.

Thanks.  Oh well, was just hoping to find the culprit in the SDF with no lead.

[Mariano Martinez Peck]

Hi Wayne,

Sounds good. Let us know what do you find with the EsMemoryUseReport if it recurs. 

Best, 

--

You received this message because you are subscribed to the Google Groups "VA Smalltalk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to va-smalltalk...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/va-smalltalk/512727a6-229a-42dc-a6af-1cc5e2e90e65o%40googlegroups.com.