Consecutive Full GCs but no OOM

[Bernd Eckenfels]

Hello Thomas,

Is it possible that the default is too small or it should not apply to full heap but oldgen instead? If one survivor and potential eden is empty (or equivalent region in G1) it still should consider consecutive full GCs (triggered by full oldgen), or did I missunderstand?

Gruss
Bernd
--
http://bernd.eckenfels.net
________________________________
From: hotspot-gc-use <hotspot-gc-use-bounces at openjdk.java.net> on behalf of Thomas Schatzl <thomas.schatzl at oracle.com>
Sent: Friday, May 25, 2018 7:19:48 PM
To: Binoy Dalal
Cc: hotspot-gc-use at openjdk.java.net
Subject: Re: Consecutive Full GCs but no OOM

Hi Binoi,

On Fri, 2018-05-25 at 22:35 +0530, Binoy Dalal wrote:
> Thomas,
> Thanks for your response.
> How did you calculate 4% remaining?
> Is there some error in my calculations because the total amount of
> heap cleared is 679K which is less than 2%(180879.36K) of the total
> heap (9043968K)? Or am I using incorrect parameters while
> calculating?

"Probably the most important flag here would be GCHeapFreeLimit:
simplified, if free space after full gc is below this percentage of
total heap, that full gc does not count against GCOverHeadLimit."

Total heap is the size given by -XX:MaxHeapSize.

I.e. -XX:MaxHeapSize=9663676416 = 9216MB

The total heap after full gc is e.g. 9043968K, i.e. 8832 MB.

1 - 8832/9216 = 4.17%.

Thanks,
  Thomas

_______________________________________________
hotspot-gc-use mailing list
hotspot-gc-use at openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/hotspot-gc-use
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/hotspot-gc-use/attachments/20180525/fdbbfbed/attachment.html>