Am 06.12.2010 schrieb "Mike Sievers <saturngeist@xxxxxxxxxxxxxx>":
> Hi List,
> my message.log shows:
> The following is only an harmless informational message.
> Unless you get a _continuous_flood_ of these messages it means
> everything is working fine. Allocations from irqs cannot be
> perfectly reliable and the kernel is designed to handle that.
> gzip: page allocation failure. order:1, mode:0x20, alloc_flags:0x30
> Pid: 15317, comm: gzip Not tainted 184.108.40.206-0.2-xen #1
> Call Trace:
> [<ffffffff80009b95>] dump_trace+0x65/0x180
> [<ffffffff803508e6>] dump_stack+0x69/0x73
> Is this really harmles?
This has nothing to do with xen. It's a linux kernel thing.
It shows that your kernel was temporarily out of memory.
In the call trace (you snipped this) you should see which instance failed in
trying to allocate the memory. If you see no other messages later then
everything should be ok.
> The server is xen domU
Company details: http://ts.fujitsu.com/imprint.html