Re: Memory profiling in Production environment

Hi all,

I've observed from that the GC logs that there is an unmistakable
upward trend that does not come down to it's original levels. This
has led me to suspect that there is a potential memory leak in the
codes.

I've tried to profile the codes in a test environment but to no
avail. This could be because the leak takes place in a section of the
code which is not actively tested in the test environment. As there
are quite a number of applications deployed within the container, and
automated testing is not available, it may prove rather difficult to
go through all test cases in a brute force manner in an attempt to
replicate the issue.

As such, I would like to ask if there is any way to (periodically)
capture memory snapshots of the Production heap that can subsequently
be downloaded for offline analysis. I understand there may be some
performance penalty and I would be grateful if anyone could advise if
the overhead is negligible.

Please feel free to also offer me advise if there are better ways of
achieving my aims.