[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: __alloc_pages: 2-order allocation failed



One possibility here...

I'm currently running qmail with a virtual memory limit of 64 MB (via
ulimit).  I wonder if it's trying to allocate more memory than that...

Interestingly, looking through my logs at the times this has happened
(once on Fri at 6:30 pm and once on Mon at 6:30 pm), a single email has
come in for quite a few local recipients.  Couple that with the fact
that spamc runs once for each user, and I begin to think it's possible
that qmail really *is* trying to go beyond the 64 MB limit.

I'm upping that to see if it affects anything.

One thing that disturbs me, though, is the imapd process that caused
this error.  This is courier imapd, running seperately from qmail,
hence it shouldn't be running into the memory limit imposed on qmail.
It looks like the package uses a virtual memory ulimit of 64 MB by
default, but this shouldn't be dependent on qmail...

Apr 18 18:32:33 psychology kernel: __alloc_pages: 1-order allocation
failed (gfp =0x20/0)
Apr 18 18:32:33 psychology kernel: __alloc_pages: task(spamc)
pid(16043) caller(000000000046ed98)
Apr 18 18:32:33 psychology kernel: __alloc_pages: 1-order allocation
failed (gfp =0x20/0)
Apr 18 18:32:33 psychology kernel: __alloc_pages: task(imapd)
pid(14534) caller(000000000046ed98)

Same time for both of those.  Weird...


On Tue, Apr 19, 2005 at 03:04:14PM -0400, Mike Edwards said:
> 000000000046ea20 T __alloc_pages
> 000000000046ed80 T __get_free_pages
> 000000000046edc0 T get_zeroed_page
> 000000000046ee00 T __free_pages
> 000000000046ee60 T free_pages
> 
> 
> On Tue, Apr 19, 2005 at 11:43:10AM -0700, David S. Miller said:
> > On Tue, 19 Apr 2005 10:34:52 -0400
> > Mike Edwards <sauron-debian-sparc@psychology.rutgers.edu> wrote:
> > 
> > > __alloc_pages: 1-order allocation failed (gfp=0x20/0)
> > > __alloc_pages: task(spamc) pid(15858) caller(000000000046ed98)
> > 
> > What symbols in the System.map of this kernel are near the
> > 0x46ed98 address being output?  It will solve the puzzle.

-- 
Mike Edwards <sauron-debian-sparc@psychology.rutgers.edu>
System Administrator
Psychology Department, Rutgers University, Newark campus



Reply to: