src/sys/nfsserver, branch release/6.2.0

MFC: The nfsm_srvpathsiz() macro in nfsrv_symlink() in nfs_serv.c

2007-01-07T13:22:49+00:00

should check length of the pathname in the range 0<=n<=NFS_MAXPATHLEN,
not 0

MFC: sys/nfsserver/nfs_serv.c 1.167

2006-12-02T09:02:30+00:00

Protect nfsm_srvpathsiz() call with the nfsd_mtx lock.

Reviewed by:	mohans
Requested by:	kib
Approved by:	re (rwatson)

MFC rev. 1.166:

2006-11-03T09:16:48+00:00

Fix leak in NAMEI zone caused by nfs server when VOP_RENAME fails.

Approved by:	re (hrs)

MFC: Release references acquired by VOP_GETWRITEMOUNT() and vfs_getvfs().

2006-09-27T00:37:46+00:00

Approved by:	re (kensmith)

MFC revision 1.99 nfs_srvsock.c

2006-09-02T23:58:21+00:00

MFC revision 1.111 mac_vfs.c
MFC revision 1.70 mac.h
MFC revision 1.71 mac_policy.h

MFC the mac_associate_nfsd_label entry point, only replace one of the
mac_policy_ops stub elements to ensure that we dont break compatability
with existing MAC modules.

MFC revision 1.79 nfs.h

2006-09-02T21:54:08+00:00

MFC revision 1.161 nfs_serv.c
MFC revision 1.96 nfs_srvsock.c
MFC revision 1.105 nfs_syscalls.c

date: 2006/01/28 19:24:39;  author: csjp;  state: Exp;  lines: +1 -1
Manage the ucred for the NFS server using the crget/crfree API defined in
kern_prot.c. This API handles reference counting among many other things.
Notably, if MAC is compiled into the kernel, it will properly initialize the
MAC labels when the ucred is allocated.

This work is in preparation for a new MAC entry point which will be responsible
for properly initializing policy specific labels for the NFS server credential.
Utilization of the crfree/crget APIs reduce the complexity associated with
this label's management.

Submitted by:	green (with changes) [1]
Obtained from:	TrustedBSD Project
Discussed with:	rwatson, alfred

[1] I moved the ucred allocation outside the scope of the NFS server lock to
    prevent M_WAIKOK allocations from occurring with non-sleep-able locks held.
    Additionally, to reduce complexity, the ucred persist as long as the NFS
    server descriptor.

MFC: Free the server request cache on unload and use callout_drain().

2006-08-11T19:13:25+00:00

MFC of the temporary fix for nfsd leaking GIANT.

2006-06-13T10:50:29+00:00

src/sys/nfsserver/nfs_serv.c rev. 1.165
src/sys/nfsserver/nfs_srvsubs.c rev. 1.141

Approved by:	pjd (mentor)

MFC r1.18:

2006-04-30T03:55:45+00:00

>  Bump up the NFS server dupreq cache limit to 2K (from 64). With a small
>  duplicate request cache, under heavy load a lot of non-idempotent requests
>  were getting served again, resulting in errors.

Approved by:	re (scottl)

rick says:

2006-04-04T15:29:51+00:00

The following bug was just identified in OpenBSD and it looks like the same
bug exists in the other BSDen NFS servers.

A Linux client (don't know which version, but you can look at
        http://bugzilla.kernel.org/show_bug.cgi?id=6256)
does a Setattr of mtime to the server's time, where the file is mode 0664 and
the client user has group access (ie. caller is not the file owner).

The BSD servers fail the Setattr with EPERM, since the VA_UTIMES_NULL flag
isn't set before doing the VOP_SETATTR.

It seems to me that this should be allowed, since it is allowed for a local
utimes(2). If so, the fix is to set VA_UTIMES_NULL for the
"set-time-to-server-time" cases of setting atime and/or mtime.

MFC of revision 1.140 to RELENG_6.

Submitted by:   rick@snowhite.cis.uoguelph.ca
Reviewed by:	cel
Approved by:	re (scottl), silby