Suddenly slow lstat syscalls on CURRENT from Juli

Kostik Belousov kostikbel at gmail.com
Sat Jan 1 16:13:02 UTC 2011


On Sat, Jan 01, 2011 at 05:00:56PM +0100, Beat G?tzi wrote:
> On 01.01.2011 16:45, Kostik Belousov wrote:
> > Check the output of sysctl kern.maxvnodes and vfs.numvnodes. I suspect
> > they are quite close or equial. If yes, consider increasing maxvnodes.
> > Another workaround, if you have huge nested directories hierarhy, is
> > to set vfs.vlru_allow_cache_src to 1.
> 
> Thanks for the hint. kern.maxvnodes and vfs.numvnodes were equal:
> # sysctl kern.maxvnodes vfs.numvnodes
> kern.maxvnodes: 100000
> vfs.numvnodes: 100765
> 
> I've increased kern.maxvnodes and the problem was gone until
> vfs.numvnodes reached the value of kern.maxvnodes again:
> # sysctl kern.maxvnodes vfs.numvnodes
> kern.maxvnodes: 150000
> vfs.numvnodes: 150109
The processes should be stuck in "vlruwk" state, that can be
checked with ps or '^T' on the terminal.

> 
> As the directory structure is quite huge on this server I've set
> vfs.vlru_allow_cache_src to one now.
Did it helped ?

> 
> > You did not specified how much memory your machine have, but I assume it
> > is > 1GB. Anyway, increase of maxvnodes on i386 should be done very
> > cautiously, since it is easy to exhaust KVA.
> 
> The server has 4GB of RAM. Is it possible to check how much i could
> increase kern.maxvnodes without exhausting KVA?
> 
The RAM check is mostly to make sure that you have enough physical RAM
to cover whole possible KVA (1GB). Too aggressive settings for maxvnodes
would cause panic.

Default settings of 100000 is more or less optimal for the i386 arch.
If you need more, you should switch to amd64.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20110101/f55e9975/attachment.pgp


More information about the freebsd-current mailing list