 | | Linux Too Aggressive With File Caching? (megaraid allocation
failure) | Linux Too Aggressive With File Caching? (megaraid allocation
failure) 2004-04-02 - By Kevin Stussman
Back
I 've sent this to the Dell Poweredge mailing list (as it does relate to
the 2650 + Powervault 220S / Perc4 Raid Card) and they basically point
to a kernel issue. I am hoping that someone here knows about this (i.e
is it a known issue, severity , fixes etc..) So far the only problem has
been 100s of messages like the following:
megaraid:allocation failed
but no tailspins (touch wood).
I am 99.99% sure that memory used by processes is not at issue here. The
stats have always remained at this level:
total used free shared buffers
cached
Mem: 6063128 6020328 42800 0 209560
5509172
-/+ buffers/cache: 301596 5761532
Swap: 8385848 102076 828377
OS : 2.4.21-4.ELsmp #1 SMP Fri Oct 3 17:52:56 EDT 2003 i686 i686 i386
GNU/Linux
-- ---- -----
Original Messages
-- ---- -----
> Does anyone have insight into what this means? (in dmesg/syslog)
>
> megaraid:allocation failed
It indicates that the driver could not allocate kernel memory, which is
rare but possible. If the system kept going, this means that it was
eventually able to allocate memory to satisfy the driver 's needs.
> > I have come across what looks to be a problem with the caching
algorithm
> > being too aggressive with RHEL3.0. The OS seems to cause the machine
to
> > swap instead of releasing memory from it 's cache which seems wrong
to me
> > and causes several of my machines to dive into a swapping tailspin
that
> > just kills the machine.
--
Taroon-list mailing list
Taroon-list@(protected)
http://www.redhat.com/mailman/listinfo/taroon-list
|
|
|