G1 issue: falling over to Full GC

[Simone Bordet]

Hi,

You have one Full GC at 3573.851, and just afterwards a concurrent-mark-abort.
There was a concurrent marking going on, started at 3514.480, but it
was not finished not even after ~60 s.
The same goes for the second full GC at 5393.957, which had a
concurrent-mark-start at 5267.191 and aborted.

While concurrent marking was happening, you had several young GC until
the heap was full, which triggered the full GC.

Now, *why* concurrent marking on a 1 GB head takes > 60 s, I don't
know; it seems an incredibly long time even for a full old gen.
I can only think the GC thread was starved, but perhaps there are
other explanations.

There was only 1 concurrent mark that ended completely (between the 2
full GCs), followed by a:

4972.437: [GC pause (partial), 1.89505180 secs]

that I cannot decypher (to Monica - what "partial" means ?), and no
mixed GCs, which seems unusual as well.
Are you sure you are actually using 1.7.0_u7 ?

Simon
-- 
http://cometd.org
http://webtide.com
Developer advice, training, services and support
from the Jetty & CometD experts.
----
Finally, no matter how good the architecture and design are,
to deliver bug-free software with optimal performance and reliability,
the implementation technique must be flawless.   Victoria Livschitz