[EMAIL PROTECTED] wrote:
>
> Peter,
> If they are only occasional errors then it's fine but if there are
> loads of them the system could hang, but of course not in your case as the
> process is being killed instead. What's happening is the syslogd cannot free
Well, this has happened to me a few times now. Once I tried to use the machine
and all it did was give me this errors. Dumped those messages to all of the
vt's so I couldn't write any commands. All I could do was a cold reboot.
> up pages. Syslogd has directly made a request for free memory which invokes
> the do_try_to_free_pages() function. (Kswapd usually calls this function).
> This function then tries to free kernel cache memory and attempts to shrink
> mapped memory using (shrink_mmap). It tries to free shared out memory. Try
> and increase your swap space. Actually I think there should be a few VM
Shouldn't this be enough:
[root@ws130 /root]# free
total used free shared buffers cached
Mem: 45696 38432 7264 3028 26716 3872
-/+ buffers/cache: 7844 37852
Swap: 80284 1828 78456
??
Maybe I've got a memory leaking application that is running or something...?
> patches out there for this, something to do with increasing the priority
> setting in shrink_mmap or calling swap_out(), can't really remember.
Any ideas where I could find these patches?
Regards,
Peter
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
