Full GC being triggered, but cant understand why

Question

This is what is there in one of the entries in my gc.log
{Heap before GC invocations=302:
Heap
PSYoungGen      total 116160K, used 555K [0x10010000, 0x17810000, 0x17810000)
eden [0x10010000,0x10010000,0x16af0000)
from [0x17180000,0x1720afd0,0x17810000)
to   [0x16af0000,0x16af0000,0x17180000)
PSOldGen        total 401408K, used 253915K [0x17810000, 0x30010000, 0x30010000)
object [0x17810000,0x27006e48,0x30010000)
PSPermGen       total 131072K, used 92569K [0x30010000, 0x38010000, 0x38010000)
object [0x30010000,0x35a767c8,0x38010000)
1103.451: [Full GC 254471K->222947K(517696K), 5.0458606 secs]
Heap after GC invocations=302:
Heap
PSYoungGen total 116288K, used 0K [0x10010000, 0x17810000, 0x17810000)
eden [0x10010000,0x10010000,0x16b30000)
from [0x171a0000,0x171a0000,0x17810000)
to [0x16b30000,0x16b30000,0x171a0000)
PSOldGen total 401408K, used 222947K [0x17810000, 0x30010000, 0x30010000)
object [0x17810000,0x251c8dd8,0x30010000)
PSPermGen total 131072K, used 92569K [0x30010000, 0x38010000, 0x38010000)
object [0x30010000,0x35a767c8,0x38010000)
I cant understand what made the garbage collector decide to do a Full GC in this scenario.
My options to the GC are as follows
-Xms512m -Xmx512m -Xmn120m -XX:PermSize=128m -XX:MaxPermSize=128m -XX:+UseParallelGC
-XX:+PrinGCDetails
This a machine that has 4GB RAM and 4 intel processors. Any help would be greatly appreciated. I am new to Garbage Collection tuning
By The way. I am using the Sun HotSpot 1.4.2_04 with weblogic 813
Edited by: Shivreet on Sep 16, 2007 12:23 AM

Answer 1

Shivreet wrote:
This is what is there in one of the entries in my gc.log
{Heap before GC invocations=302:
Heap
PSYoungGen      total 116160K, used 555K [0x10010000, 0x17810000, 0x17810000)
eden [0x10010000,0x10010000,0x16af0000)
from [0x17180000,0x1720afd0,0x17810000)
to   [0x16af0000,0x16af0000,0x17180000)
PSOldGen        total 401408K, used 253915K [0x17810000, 0x30010000, 0x30010000)
object [0x17810000,0x27006e48,0x30010000)
PSPermGen       total 131072K, used 92569K [0x30010000, 0x38010000, 0x38010000)
object [0x30010000,0x35a767c8,0x38010000)
1103.451: [Full GC 254471K->222947K(517696K), 5.0458606 secs]
Heap after GC invocations=302:
Heap
PSYoungGen total 116288K, used 0K [0x10010000, 0x17810000, 0x17810000)
eden [0x10010000,0x10010000,0x16b30000)
from [0x171a0000,0x171a0000,0x17810000)
to [0x16b30000,0x16b30000,0x171a0000)
PSOldGen total 401408K, used 222947K [0x17810000, 0x30010000, 0x30010000)
object [0x17810000,0x251c8dd8,0x30010000)
PSPermGen total 131072K, used 92569K [0x30010000, 0x38010000, 0x38010000)
object [0x30010000,0x35a767c8,0x38010000)
I cant understand what made the garbage collector decide to do a Full GC in this scenario.
My options to the GC are as follows
-Xms512m -Xmx512m -Xmn120m -XX:PermSize=128m -XX:MaxPermSize=128m -XX:+UseParallelGC
-XX:+PrinGCDetails
This a machine that has 4GB RAM and 4 intel processors. Any help would be greatly appreciated. I am new to Garbage Collection tuning
By The way. I am using the Sun HotSpot 1.4.2_04 with weblogic 813
Edited by: Shivreet on Sep 16, 2007 12:23 AMIs there any reason you must stay with that version of the JVM?
If you could upgrade to a more recent version of the JDK, in particular,
JDK 6u2, please try that.
If, for some reason, you are unable to upgrade, then try the following
option and let us know if it helped:-
-XX:-UseAdaptiveSizePolicy