Date: Fri, 16 Apr 2004 11:33:44 -0500 From: Dan Nelson <dnelson@allantgroup.com> To: Brian Henning <b1henning@hotmail.com> Cc: freebsd-questions@freebsd.org Subject: Re: nfs cache Message-ID: <20040416163334.GA2146@dan.emsphone.com> In-Reply-To: <BAY15-DAV50CIFoPkG0000215e1@hotmail.com> References: <BAY15-DAV50CIFoPkG0000215e1@hotmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In the last episode (Apr 16), Brian Henning said: > I am trying to get some info on the cache for nfs on the nfs client. > I have looked at nfsstat > > where are the settings located for the nfs cache? I think they're just part of the system's regular disk cache, except for the sysctl below. > the setattr percentages are really high on this server. The is the > main reason i want to know what is going on with the cache. Getattr, you mean? That just means that the client has stat()'ed a file twice within the timeout specified by the vfs.nfs.access_cache_timeout sysctl, and has pulled the value from its cache instead of querying the server. > Version 2: (185583924 calls) > null getattr setattr root lookup > 0 0% 175265837 94% 66975 0% 0 0% 6407008 3% So here, 94% of the getattr calls were pulled from cache. -- Dan Nelson dnelson@allantgroup.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040416163334.GA2146>