Date: Mon, 1 Aug 2005 17:14:45 -0400 (EDT) From: rick@snowhite.cis.uoguelph.ca To: kris@obsecurity.org Cc: fs@freebsd.org, openbsd-nfsv4@sfobug.org Subject: Re: Re: FreeBSD6.0-BETA1 panics Message-ID: <200508012114.RAA37699@snowhite.cis.uoguelph.ca>
next in thread | raw e-mail | index | archive | help
> It didn't panic yet, but when doing a multiple simultaneous cvs update > with NFS3 mounted repo the clients eventually got stuck in the nfsreq > state. > > FreeBSD had a similar problem for a while, after a change made last > December, but it was fixed early this year. Is your server based on > an old version of the freebsd server code that is missing this (and > possibly other) fixes, or is this a new bug introduced in your > version? The short answer is that the low level socket handling code for the FreeBSD port is pretty well cloned from the FreeBSD server, so I'll take a look at the bugfix stuff on the web site, to see what that was. I'll also check my code against the current FreeBSD server stuff. If you could email me the following on the server at the time of the hang, it could be useful: # netstat -a <-- to see if the connection has data in the send or rcv queue # ps axl | fgrep newnfsd <-- to see if any are hung and where Again, thanks for the testing, rick
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200508012114.RAA37699>