At 10:26 AM 27/02/2006, Danny Butroyd wrote:
OK, do the changes affect both server and client and is there any
documentation on the changes?
cvs commits and reading this mailing is the best place to track what
has changed in detail. Unfortunately, there is no middle ground
source for information. I use a procmail script to sort all RELENG_6
commits into a file that I can then search through after the fact.
The descriptions generally offer an OK description as to what was
changed / fixed improved. If that fails, going back to the original
commit message might offer more detail. e.g. a quick perusal of just
the nfs code commits shows a couple of obvious bug fixes.
maxim 2005-12-15 18:10:37 UTC
FreeBSD src repository
Modified files: (Branch: RELENG_6)
sys/nfsclient nfs_socket.c
Log:
MFC rev. 1.134: fix for a bug where NFS/TCP would
not reconnect (in the case where the server FIN'ed).
PR: kern/88833
Requested by: Roman V. Palagin
Approved by: Mohan Strinivasan
Revision Changes Path
1.125.2.5 +1 -1 src/sys/nfsclient/nfs_socket.c
tegge 2006-01-14 01:05:22 UTC
FreeBSD src repository
Modified files: (Branch: RELENG_6)
sys/nfs4client nfs4_vfsops.c
Log:
MFC: Obtain mount point lock before restarting sync loop if vget() failed.
Revision Changes Path
1.20.2.1 +1 -0 src/sys/nfs4client/nfs4_vfsops.c
rwatson 2006-02-14 00:06:32 UTC
FreeBSD src repository
Modified files: (Branch: RELENG_6)
sys/nfsclient nfs_lock.c
Log:
Merge nfs_lock.c:1.43 from HEAD to RELENG_6:
In nfs_dolock(), GC now under-used ioflg, rendered obsolete when we moved
from using a fifo to talk to rpc.lockd to using a special device node.
Approved by: re (scottl)
rees 2006-02-16 02:39:53 UTC
FreeBSD src repository
Modified files: (Branch: RELENG_6)
sys/nfsclient nfs_socket.c
Log:
MFC rev 1.135:
Don't log an error on tcp connection reset, even if we don't get ECONNRESET.
Submitted by: [EMAIL PROTECTED]
Approved by: re (scottl)
Revision Changes Path
1.125.2.6 +2 -2 src/sys/nfsclient/nfs_socket.c
>
> BTW, if you are using tcp mounts, and you have devices not on the same
> subnet, setting
> net.inet.tcp.inflight.enable=0
> helps with performance. However, this is a separate issue to the
> problems you are seeing.
>
Yeah, I saw this in the list archives :)
Danny
> ---Mike
>
>
>> Danny
>
> _______________________________________________
> [email protected] mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"
>
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
