I got a new server today and I'm trying to determine the feasibility of
knfsd in our mixed environment of Linux and Solaris (and many versions of
these OSes).
I have installed kernel 2.1.131 + DAC960(Mylex driver) + HJ's knfsd-981122
(all except quota part, UNIX socket part, sunrpc part - which seems
obsolete now - and a part of the root patch, a 1 liner which patches
now non-existent code). System is PII 450 UP, 256MB, DAC960 Raid, all
compiled w/gcc version 2.7.2.3, RedHat 5.1.
There are a few questions I have, but only one real "showstopper." After:
(spike = above machine, sparky = Solaris 2.5 w/big patches from Sun)
On spike:
[root@spike /root]# kexportfs -o rw,no_root_squash sparky:/home
On sparky:
bash# mount -o rw,rsize=4096,wsize=4096 spike:/home /mnt
bash# bonnie -s 50m
File './Bonnie.2068', size: 52428800
Writing with putc()...NFS write failed for server spike: error 2 (RPC:
Can't decode result)
Bonnie: drastic I/O error (putc): I/O error
And in the logs on sparky:
Dec 3 19:35:01 sparky unix: NFS write failed for server spike: error 2
(RPC: Can't decode result)
Dec 3 19:35:01 sparky unix: NFS write error on host spike: I/O error.
Dec 3 19:35:01 sparky unix: (file handle: 403939cf 3780000 1780000
4300000 4300000 2000000 0 0)
<repeating>
I tried again on a Solaris 2.6 box (marvin):
On spike:
[root@spike /root]# kexportfs -o rw,no_root_squash marvin:/home
On marvin:
bash# mount -o rw,rsize=4096,wsize=4096 spike:/home /mnt
bash# bonnie -s 50m
File './Bonnie.26154', size: 52428800
Writing with putc()...done
Rewriting...done
Writing intelligently...done
Reading with getc()...done
Reading intelligently...done
Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done...
-------Sequential Output-------- ---Sequential Input--
--Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block---
--Seeks---
Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec
%CPU
50 5428 46.1 2799 6.8 1662 5.9 3509 30.9 109964 100.9
5216.8 58.7
It completed! *BUT* in the logs:
Dec 3 19:48:15 marvin unix: NFS write failed for server spike: error 2
(RPC: Can't decode result)
Dec 3 19:48:15 marvin unix: NFS write error on host spike: I/O error.
Dec 3 19:48:15 marvin unix: (file handle: 80aec4cf 2780000 1780000
4300000 4300000 2000000 0 0)
<also repeats about 3 times>
The only thing in the linux server logs is:
Dec 3 12:40:27 spike mountd[624]: authenticated mount request from
sparky.cobite.com:950
Dec 3 12:40:27 spike kernel: nfsd: connect from unprivileged port:
cf8e880b:46209<4>nfsd: accept failed (err 11)!
Dec 3 12:40:27 spike kernel: nfsd: accept failed (err 11)!
Dec 3 12:40:27 spike kernel: svc: unknown program 100227 (me 100003)
<repeats once for each mount by a Solaris client>
What can I do to track this down? tcpdump w/some special options?
Thanks,
David
P.S. Why do I have to explicitly give the 'rw' option in the export and
the mount, isn't this the default??
-- /==============================\ | David Mansfield | | david@cobite.com | \==============================/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/