> From: CrystalCracker [mailto:sudarshan.acha...@gmail.com]
> Subject: Re: Understanding GC Logs
>
> Shouldn't the 'port' object get collected once ?
Don't confuse a reference to an object with the object itself. The port
variable may go out of scope, but
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
CrystalCracker,
On 6/11/2009 12:01 PM, CrystalCracker wrote:
> Upon more profiling, I figured that I have lots of apache.axis objects that
> are not garbage collected.
Does the MyPortType class have a 'close' method or something like that
that must b
”)
}
return account;
}
}
Shouldn't the 'port' object get collected once ?
--
View this message in context:
http://www.nabble.com/Understanding-GC-Logs-tp23862354p23984169.html
Sent from the Tomcat - User mailing list archiv
> From: CrystalCracker [mailto:sudarshan.acha...@gmail.com]
> Subject: RE: Understanding GC Logs
>
> Not sure what you mean by pooling those. I am using JBoss cache.
Which I'm not familiar with, so I can't really offer much advice about it. The
GC log does show some rath
a SoftReference
objects, thus letting GC >>throw them away when space gets tight.
You mean using -XX:SoftRefLRUPolicyMSPerMB= ?
--
View this message in context:
http://www.nabble.com/Understanding-GC-Logs-tp23862354p23871748.html
Sent from the Tomcat - User maili
> From: CrystalCracker [mailto:sudarshan.acha...@gmail.com]
> Subject: Re: Understanding GC Logs
>
> I have a caching layer which is configured to evict objects from memory
> after 3 minutes of idle time. I have profiled and see all those objects
> in cache get evicted after so
me idle time.
Is the pause times really too long compared to the throughput (99.6)? If so
I probably could decrease the size of heap to make GC more frequent.
--
View this message in context:
http://www.nabble.com/Understanding-GC-Logs-tp23862354p23870052.html
Sent from the Tomcat - User mailing lis
CrystalCracker wrote:
> I am just trying to learn and understand the GC logs. (I did read the sun's
> docs on GC tuning).
>
> jdk1.6.0_07
> -Xms2g -Xmx2g -Xss256k
> -XX:+UseLargePages -XX:+PrintGCDetails
> -verbose:gc -Xloggc:/var/log/gc.log
> -XX:PermSize=512m -XX:MaxPermSize=512m
>
> It is r
gGen: 551747K->98346K(574784K)]
1723184K->1271902K(1973568K), 0.2013120 secs] [Times: user=0.32 sys=0.01,
real=0.21 secs]
326378.971: [GC [PSYoungGen: 555050K->57405K(578560K)]
1728606K->1232674K(1977344K), 0.1028040 secs] [Times: user=0.22 sys=0.00,
real=0.11 sec
