[Gc] "mmap remapping failed" in long-running server

Julian Hall jules@acris.co.uk
Thu, 06 Nov 2003 09:44:16 +0000


Kenneth C. Schalk wrote:

>I'm responsible for maintaining a server that's linked with the
>garbage collector.  (It's part of Vesta: http://www.vestasys.org/).
>
>Where I work (which probably has the heaviest loaded such server
>anywhere), we've been having some problems.  With increasing frequency
>(recently about every 24-48 hours), the server has been dying with the
>message "mmap remapping failed" (which is printed at os_dep.c:1844).
>When this happens, there's clearly more memory on the system, and the
>server is usually significantly below its peak total memory size.
>[...]
>- The OS is Linux.  The kernel is 2.4.9, and the rest of the system is
>a derivative of RedHat 7.1.  (I know, it's rather old, but these are
>externally imposed constraints.  We hope to be moving to a 2.4.20
>kernel soon.)
>  
>
Can I suggest, if at all possible, that you may want to accelerate that 
upgrade.  If my memory serves correctly, 2.4.9 was the release before 
Linus decided that the new virtual memory manager that had been 
implemented for the 2.4 series was too unstable, and switched to an 
alternative design.  This instability could have something to do with 
the problems you are experiencing: I'm not certain what the specific 
problems with the VMM were, but there must have been very good reasons 
for the switch to a completely new implementation in the middle of a 
supposedly stable kernel series...

Hope this helps

Jules