mbuf leak

Matthew Dillon dillon at apollo.backplane.com
Fri Apr 16 11:29:26 PDT 2004


:Pete Carah wrote:
:> I'm seeing on several machines a leak of mbufs.  This started the middle of 
:> last week.  It was much worse last week than it is now; apparently a set 
:> of commits to tcp_* on Monday fixed part (the gradual part) of the problem.  
:> Now things run steadily for a day or two and *suddenly* run out of mbufs no 
:> matter what I set nmbclusters to.
:
:
:I just started witnessing the same problem as well.
:
:-Mike

    All right.  There's a good chance it is related to packet fragment
    handling or packetes related to dead connections.  But the more info
    we can get on the types of packets causing the problem the easier it 
    will be to track down.

    I think I see a bit of leak on my machines too now that you mention it.
    I'll try to track it down on my end as well.  Once we get it fixed I 
    think it's finally time for me to create a (non-branch) cvs tag to track
    known-to-be-probably-stable builds :-).

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>





More information about the Kernel mailing list