NFS: freeze during copy

[rvenne at dental-on-line.fr]

Cheffo wrote:

> rvenne at dental-on-line.fr wrote:
>
>> Hi
>>
>> I try to connect NFS client.
>>
>> on the server:
>> /etc/rc.conf:
>> rpcbind_enable="YES"
>> nfs_server_enable="YES"
>> mountd_flags="-r"
>>
>> 10029  ??  Ss     0:00.02 rpcbind
>> 10031  ??  Is     0:00.03 nfsd: master (nfsd)
>> 10032  ??  I      0:00.00 nfsd: server (nfsd)
>> 10033  ??  I      0:00.00 nfsd: server (nfsd)
>> 10034  ??  I      0:00.00 nfsd: server (nfsd)
>> 10035  ??  I      0:00.00 nfsd: server (nfsd)
>> 10037  ??  Is     0:00.01 mountd -r
>>
>> cat /etc/exports
>> /usr/data/backup -maproot=admin:wheel 192.168.254.2
>>
>> client:
>> 83  ??  DL     0:00.03 [nfsiod 0]
>>   84  ??  DL     0:00.00 [nfsiod 1]
>>   85  ??  DL     0:00.00 [nfsiod 2]
>>   86  ??  DL     0:00.00 [nfsiod 3]
>>
>> /etc/rc.conf: nfs_client_enable="YES"
>>
>> I was able to mount nfs mounting point and to copie 16K of 7G data. 
>> and, it goes to freezing. I can't kill neither cp process neither 
>> mount unmount nfs (umount command goes freezing too). tcpdump tells 
>> me that server doesn't give any response in spite of client's udp 
>> queries.
>>
>> need helps. thx
>
> Not sure that this can help, but do you use pf on client ?

humm....ipf.
pass in quick on bge0 from 192.168.254.1/32 to 192.168.254.2 keep state
pass out quick on bge0 from 192.168.254.2/32 to any keep state
192.168.254.2 is client

> I had a problem before using properly NFS with PF and "scrub in all"
> and removing normalizer fix NFS client.
> And please do not break threads ;)

>> _______________________________________________
>> freebsd-stable at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
>> To unsubscribe, send any mail to 
>> "freebsd-stable-unsubscribe at freebsd.org"
>
The client keep sending udp write packets, but any answers from 
server.... I've to fix something with sysctl?

-- 
Richard VENNE
www.dental-on-line.com
Phone: 01 43 27 94 24
fax: 01 43 27 66 85