On Thu, May 10, 2007 at 10:28:23AM +0200, Rik Theys wrote: > I've created a new package based on the etch package with the above file > patched, but it doesn't seem to help: the memory leak is still there.
OK, so my guess was right -- it's not that fix, it's something else. > I reproduced the problem on a test system like this: I used your scripts (although only from one machine), and I'm not able to reproduce it with nfs-kernel-server from etch. The memory count goes up a bit, but then stabilizes. > Should I repeat this with 1.0.12? Do you think 1.0.12 will get accepted > in 4.0r1 (if it fixes the problem)? Your best guess at this point is probably Valgrind. Recompile the etch version with debug information (remove the "dh_strip" line in debian/rules), then run mountd with "valgrind --leak-check=full rpc.mountd -F -d all". After it's started leaking, just Ctrl-C it and see what comes out. You're probably most interested in the "definitely leaked" part. /* Steinar */ -- Homepage: http://www.sesse.net/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
