cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Hannes Schmidt <han...@eyealike.com>
Subject Re: Native heap leaks?
Date Mon, 09 May 2011 17:03:14 GMT
> I have not looked into smaps before. But it actually seems odd that that
> mmaped Index files are taking up so *little memory*.  Are they only a
> few kb on disk?

The sum of the sizes of all *-Index.db files in /var/lib/cassandra is 2924kb.

> Is this a snapshot taken shortly after the process
> started or before the OOM killer is presumably about to come along.

It was taken after a day of considerable write load.

>  How long does it take to go from 1.1 G to 2.1 G resident?

Two days of active use with considerable write load.

>  Either way, it
> would be worthwhile to set one node to standard io to make sure it's
> really not mmap causing the problem.

Will do that. I am currently a/b testing OpenJDK vs. Sun/Oracle JDK.
Next test will be for standard vs. auto.

> Anyway, assuming it's not mmap, here are the other similar threads on
> the topic.  Unfortunately none of them claim an obvious solution:
>
> http://www.mail-archive.com/user@cassandra.apache.org/msg09279.html
> http://www.mail-archive.com/user@cassandra.apache.org/msg08063.html
> http://www.mail-archive.com/user@cassandra.apache.org/msg12036.html
> http://mail.openjdk.java.net/pipermail/hotspot-dev/2011-April/004091.html
>

Thank you. I wasn't aware of these threads. Must have googled wrong.

Mime
View raw message