Date: Wed, 16 Jun 1999 18:19:35 -0700 (PDT) From: Studded <Studded@gorean.org> To: "David O'Brien" <obrien@NUXI.com> Cc: freebsd-current@freebsd.org Subject: Re: Heavily loaded amd gets stuck Message-ID: <Pine.BSF.4.05.9906161803010.27012-100000@dt054n86.san.rr.com> In-Reply-To: <19990616180131.A54575@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 16 Jun 1999, David O'Brien wrote:
> > without the proto=udp option and see what kind of results I get.
>
> Rather than (or in addtion to) not using "proto=udp" use "proto=tcp".
> The default in 3.2 is UDP, so removing the directive should not do
> anything.
Well, I've already done the tests, but thanks for the tip. :)
Recapping, I have the following map file:
/defaults type:=nfs;opts:=rw,nosuid
* rhost:=IP${key};rfs:=/Space/${key}
Adding to /defaults both the options "vers=2,proto=udp" yielded the kernel
crash and ddb traceback I posted previously. With the above and just
"proto=udp" I got the following:
Xresume1()
--- interrupt
splx()
nfs_nget()
mountnfs()
nfs_mount()
mount()
syscall()
Xint0x80_syscall()
The last 6 are identical in the first test, this test, and the
test below. With the above map and adding only the option "vers=2" I got
the following traceback:
Xresume1()
--- interrupt
zfreei()
nfs_reclaim()
vclean()
vgonel()
getnewvnode()
nfs_nget().... (last 6 same as above).
If there is any more information you need, ddb commands to do,
whatever just let me know. I will run a test now with the above map and
add proto=tcp and see what I get. I'm only going to be at work tonight
till 6:30 PDT, but (at this point) I have the Ok to work on it till it's
fixed, so I can start again tomorrow morning. I am *highly* motivated to
fix it since I've been pushing to use FreeBSD for several months, and now
it's not working. To add insult to injury, the linux box that is my
"competition" is up and running, using the amd-utils 6.0 (the same, yes?)
and not having any problems.
In case it matters, the box is a dual PIII-500 with a gig of ram.
SMP is running fine, all the ram is recognized, completed a make world ok,
etc. It's got an intel etherexpress pro 100 card running at 100 full
duplex. Any other details you need, just let me know.
Also, should I be considering a move to -current for this box? Is
-current stable enough right now to run a fairly heavily loaded web
server? If the NFS in -current is going to be doing better than what's in
-stable it will be worth a little headache to change, since our structure
depends on it heavily.
Thank you very much for the help,
Doug
--
*** Chief Operations Officer, DALnet IRC network ***
On account of being a democracy and run by the people, we are the only
nation in the world that has to keep a government four years, no matter
what it does.
-- Will Rogers
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.9906161803010.27012-100000>