Re: nfsd and CPU/performance problem
- From: Nash Nipples <trashy_bumper@xxxxxxxxx>
- Date: Tue, 23 May 2006 06:52:01 -0700 (PDT)
Hi Marko,
Actually i dont find that load critical. I think those lines well tell that actually the process is running 581m42s and now it utilizes 13.48% of available WCPU which is a long run and hopefully successfull if no nfs failures took place.
Im pretty confident that FreeBSD wont let any bad things happen and will allocate the resources where it needs them the most and on time.
I've googled for a few minutes and found this:
http://mail-index.netbsd.org/netbsd-bugs/1994/09/28/0000.html
which sounds like a "kernel tuning issue" if you have excluded nfsserver out of your kernel config last time u were compiling it. if you didnt just skip this part at this time.
please make sure that the following lines do exist
options NFSCLIENT # Network Filesystem Client
options NFSSERVER # Network Filesystem Server
options NFS_ROOT # NFS usable as /, requires NFSCLIENT
or it can be this:
http://lists.debian.org/debian-user/2002/08/msg02884.html
which sounds like umm "daemon aging" issue. is there such thing?
if yes, then i hope someone will share a hint on nfs server maintenance during a long run. I dont really think that restarting it on daily basis is a good thing to do
can i see some more info on nfsd please?
# ps -wux -p `pgrep nfsd`
Sincerely,
Nash
Marko Lerota <marko.lerota@xxxxxxxxxxxxxxxxx> wrote: Marko Lerota writes:
PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
429 root 1 4 0 1204K 820K - 0 581:42 13.48% nfsd
430 root 1 4 0 1204K 820K - 0 10:37 0.00% nfsd
Here is the config
rc.conf
#######################
rpcbind_enable="YES"
portmap_enable="YES"
nfs_server_enable="YES"
nfs_server_flags="-u -t -n 4 -h 10.3.11.43"
mountd_flags="-r"
nfs_client_enable="YES"
#######################
My friend found the 'problem' but I'm not shure who's problem it is.
The clients or the nfs servers. In the handbook section about NFS there
is nothing about this.
These options are added in rc.conf and server now works correctly.
rpc_lockd_enable="YES"
rpc_statd_enable="YES"
But later in the handbook section:
Figure 2-54. Network Configuration Lower-level
The rpcbind(8), rpc.statd(8), and rpc.lockd(8) utilities are all used for
Remote Procedure Calls (RPC). The rpcbind utility manages communication
between NFS servers and clients, and is *required* for NFS servers to operate
correctly.
So I think this should be in the NFS section. Anyone?
The clients are RedHatES4 and servers are FreeBSD 6.1
--
One cannot sell the earth upon which the people walk
Tacunka Witco
_______________________________________________
freebsd-net@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe@xxxxxxxxxxx"
---------------------------------
Feel free to call! Free PC-to-PC calls. Low rates on PC-to-Phone. Get Yahoo! Messenger with Voice
_______________________________________________
freebsd-net@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscribe@xxxxxxxxxxx"
- Follow-Ups:
- Re: nfsd and CPU/performance problem
- From: Marko Lerota
- Re: nfsd and CPU/performance problem
- Prev by Date: Re: How to Quicken TCP Re-transmission?
- Next by Date: (no subject)
- Previous by thread: Re: nfsd and CPU/performance problem
- Next by thread: Re: nfsd and CPU/performance problem
- Index(es):
Relevant Pages
|
|