nfs-utils.git
9 years agoexports: Clearly Defining Exports Priorities
James Pearson [Tue, 7 Jun 2011 20:25:13 +0000 (16:25 -0400)]
exports: Clearly Defining Exports Priorities

Added some verbiage to the exports(5) man page
that clearly explains the precedence around
how exports will work with regard to netgroups.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoDocument "-n" for svcgssd
Neil Brown [Tue, 7 Jun 2011 17:18:55 +0000 (13:18 -0400)]
Document "-n" for svcgssd

The svcgssd man page doesn't mention the "-n" flag.

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Fix missing varialble assignment in auth_unix_gid
Pavel Shilovsky [Tue, 7 Jun 2011 17:18:13 +0000 (13:18 -0400)]
mountd: Fix missing varialble assignment in auth_unix_gid

When we get into auth_unix_gid at the second time, groups_len
is not 0 and ngroups variable leave as 0. Then we use ngroups
in getgrouplist that fails in this case. This patch fixes it.

Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfsstat: reorder nfs4 stats for 2.6.38 and up
Benny Halevy [Mon, 23 May 2011 12:37:17 +0000 (08:37 -0400)]
nfsstat: reorder nfs4 stats for 2.6.38 and up

match order in 2.6.38, 2.6.39 (-rc3) and development tree
while at it, get rid of obsolete ds_write and ds_commit

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agosupress socket error when address family is not supported
NeilBrown [Mon, 23 May 2011 12:23:51 +0000 (08:23 -0400)]
supress socket error when address family is not supported

From: Suresh Jayaraman <sjayaraman@suse.de>

It was observed that when ipv6 module was not loaded and cannot be auto-loaded,
when starting NFS server, the following error occurs:
"rpc.nfsd: unable to create inet6 TCP socket: errno 97 (Address
family not supported by protocol)"

This is obviously a true message, but does not represent an "error" when ipv6
is not enabled.  Rather, it is an expected condition.  As such, it can be
confusing / misleading / distracting to display it in this scenario.

This patch instead of throwing error when a socket call fails with
EAFNOSUPPORT, makes it as a NOTICE.

Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemove risk of nfs_addmntent corrupting mtab
NeilBrown [Mon, 23 May 2011 12:19:57 +0000 (08:19 -0400)]
Remove risk of nfs_addmntent corrupting mtab

nfs_addmntent is used to append directly to /etc/mtab.
If the write partially fail, e.g. due to RLIMIT_FSIZE,
truncate back to original size and return an error.

See also https://bugzilla.redhat.com/show_bug.cgi?id=697975
(CVE-2011-1749) CVE-2011-1749 nfs-utils: mount.nfs fails to anticipate RLIMIT_FSIZE

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoexportfs: getexportent interprets -test-client- as default options
Ben Myers [Mon, 23 May 2011 12:07:00 +0000 (08:07 -0400)]
exportfs: getexportent interprets -test-client- as default options

With commit 1374c3861abdc66f3a1410e26cc85f86760b51dd Neil added a
-test-client- export to test the exportability of filesystems when exportfs
is run.  When using the old cache controls (i.e. /proc/fs/nfsd is not
mounted) exportfs will read /proc/fs/nfs/exports to process existing
exports and find these test client entries.  The dash at the beginning of
-test-client- will be cause getexportent to look for default options in the
rest of the string, which test-client- will not match:

exportfs: /proc/fs/nfs/exports:1: unknown keyword "test-client-(rw"

This patch resolves that problem (as Steve suggested) by not processing any
default options if we are reading the list of existing exports from the
kernel.  Default options are converted to individual exports by exportfs so
the kernel won't have any regardless.

Signed-off-by: Ben Myers <bpm@sgi.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved compilation warnings from mountd/cache.c
Jim Rees [Wed, 18 May 2011 16:42:02 +0000 (12:42 -0400)]
Removed compilation warnings from mountd/cache.c

Commit 5604b35a6 introduced a number of missing initializer
warnings that were missed. This patch removes those warnings.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfsstat: Output headings mislabled
Steve Dickson [Tue, 26 Apr 2011 17:32:35 +0000 (13:32 -0400)]
nfsstat: Output headings mislabled

The badclnt and badauth headers were reversed
when the server side rpc stats (-s -o rpc) were
displayed.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agorpc.svcgssd: Segmentation fault on error nfs-utils-1-2-4-rc8
Steve Dickson [Tue, 19 Apr 2011 16:31:30 +0000 (12:31 -0400)]
rpc.svcgssd: Segmentation fault on error

Commit 544ed73d introduced a regression that caused
rpc.svcgssd to seg fault on "Wrong principal in request"
errors in gss_accept_sec_context()

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Increase the stdio file buffer size for procfs files
Sean Finney [Tue, 19 Apr 2011 15:04:35 +0000 (11:04 -0400)]
nfs-utils: Increase the stdio file buffer size for procfs files

Previously, when writing to /proc/net/rpc/*/channel, if a cache line
were larger than the default buffer size (likely 1024 bytes), mountd
and svcgssd would split writes into a number of buffer-sized writes.
Each of these writes would get an EINVAL error back from the kernel
procfs handle (it expects line-oriented input and does not account for
multiple/split writes), and no cache update would occur.

When such behavior occurs, NFS clients depending on mountd to finish
the cache operation would block/hang, or receive EPERM, depending on
the context of the operation.  This is likely to happen if a user is a
member of a large (~100-200) number of groups.

Instead, every fopen() on the procfs files in question is followed by
a call to setvbuf(), using a per-file dedicated buffer of
RPC_CHAN_BUF_SIZE length.

Really, mountd should not be using stdio-style buffered file operations
on files in /proc to begin with.  A better solution would be to use
internally managed buffers and calls to write() instead of these stdio
calls, but that would be a more extensive change; so this is proposed
as a quick and not-so-dirty fix in the meantime.

Signed-off-by: Sean Finney <sean.finney@sonyericsson.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Use a dynamic buffer for storing lists of gid's
Sean Finney [Tue, 19 Apr 2011 15:05:47 +0000 (11:05 -0400)]
mountd: Use a dynamic buffer for storing lists of gid's

Previously, in auth_unix_gid, group lists were stored in an array of
hard-coded length 100, and in the situation that the group lists for a
particular call were too large, the array was swapped with a dynamically
allocated/freed buffer.  For environments where users are commonly in
a large number of groups, this isn't an ideal approach.

Instead, use malloc/realloc to grow the list on an as-needed basis.

Signed-off-by: Sean Finney <sean.finney@sonyericsson.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount: add --enable-libmount-mount nfs-utils-1-2-4-rc7
Karel Zak [Wed, 6 Apr 2011 16:39:21 +0000 (12:39 -0400)]
mount: add --enable-libmount-mount

This patch allows to link mount.nfs with libmount from util-linux >=
v2.19. The new libmount based code is enabled by CONFIG_LIBMOUNT and
is stored in mount_libmount.c. The old code is not affected by this
change.

The libmount does not have officially stable API yet, so the
--enable-libmount-mount is marked as experimental in the configure
help output.

The ./configure option is the same as we use in util-linux to enable
support for libmount in mount(8).

The addr= (and some other options necessary for remount/umount) are
stored to /etc/mtab or to /dev/.mount/utab. The utab file is *private*
libmount file. It's possible that some mount options (for example
user=) will be moved to kernel, so the utab will not be necessary.

About libmount:

  * supports systems without and with regular /etc/mtab
  * does not store VFS and FS mount options in userspace
  * manages user= option and evaluate permissions
  * parses VFS mount options and generate MS_* flags
  * parses /etc/{fstab,mtab}, /proc/mounts or /proc/self/mountinfo
  * long-term goal is to use the same code in all mount.<type> helpers

Note, use

   LIBMOUNT_DEBUG=0xffff mount.nfs foo:/path /path

to debug the library.

On systems with util-linux v2.19 the findmnt(8) command uses libmount
to list all/selected mount points:

   $ findmnt /path
   $ findmnt --mtab /path

the --mtab appends userspace mount options (e.g. user=) to the output.

CC: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount: move generic functions to utils.c and network.c
Karel Zak [Wed, 6 Apr 2011 15:36:40 +0000 (11:36 -0400)]
mount: move generic functions to utils.c and network.c

Move generic code that could be shared between standard mount.nfs and
libmount version to utils.c and network.c.

CC: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Add support to svcgssd to limit the negotiated enctypes
Kevin Coffman [Wed, 6 Apr 2011 15:25:03 +0000 (11:25 -0400)]
nfs-utils: Add support to svcgssd to limit the negotiated enctypes

Recent versions of Kerberos libraries negotiate and use
an "acceptor subkey".  This negotiation does not consider
that a service may have limited the encryption keys in its
keytab.  A patch (http://src.mit.edu/fisheye/changelog/krb5/?cs=24603)
has been added to the MIT Kerberos code to allow an application
to indicate that it wants to limit the encryption types negotiated.
(This functionality has been available on the client/initiator
side for a while.  The new patch adds this support to the
server/acceptor side.)

This patch adds support to read a recently added nfsd
proc file to determine the encryption types supported by
the kernel and calls the function to limit encryption
types negotiated for the acceptor subkey.

Signed-off-by: Kevin Coffman <kwc@citi.umich.edu>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoexports: add a configurable time-to-live for the kernel cache entries
Chuck Lever [Wed, 6 Apr 2011 14:53:57 +0000 (10:53 -0400)]
exports: add a configurable time-to-live for the kernel cache entries

From: Trond Myklebust <Trond.Myklebust@netapp.com>

The fedfs ldap server will specify a ttl for its entries.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This is a refactoring change only.  There should be no change in
behavior.

Original patch had updates to utils/mountd/junctions.c, which no
longer exists.  These are not included here.

Create a macro for the default cache TTL, which is used in several
places besides the export cache.

Make e_ttl unsigned.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agostatd: Remove vestigial "-w" option from man page synopsis
Chuck Lever [Wed, 6 Apr 2011 14:49:52 +0000 (10:49 -0400)]
statd: Remove vestigial "-w" option from man page synopsis

The synopsis of rpc.statd in its man page lists "-w" as a valid
option.  There is currently no support in the source code for a "-w"
option.

BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=199
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Don't leak socket in nfs_ca_sockname()
Chuck Lever [Wed, 6 Apr 2011 14:48:38 +0000 (10:48 -0400)]
mount.nfs: Don't leak socket in nfs_ca_sockname()

Ensure the test socket is always closed before nfs_ca_sockname()
returns.  Otherwise it's orphaned.

BugLink: https://bugzilla.linux-nfs.org/show_bug.cgi?id=197
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved a warning from v4root.c
Steve Dickson [Wed, 6 Apr 2011 14:46:06 +0000 (10:46 -0400)]
Removed a warning from v4root.c

v4root.c:176:9: warning: variable 'ret' set but not used

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved a warning from exportfs.c
Steve Dickson [Wed, 6 Apr 2011 14:39:10 +0000 (10:39 -0400)]
Removed a warning from exportfs.c

exportfs.c:280:29: warning: 'exp' may be used uninitialized in this function

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved a warning from conffile.c
Steve Dickson [Wed, 6 Apr 2011 14:36:30 +0000 (10:36 -0400)]
Removed a warning from conffile.c

conffile.c:258:19: warning: 'j' may be used uninitialized in this function

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoUpdate man pages for /etc/exports.d nfs-utils-1-2-4-rc6
Masatake YAMATO [Mon, 7 Mar 2011 13:36:19 +0000 (08:36 -0500)]
Update man pages for /etc/exports.d

Man page updates for /etc/exports.d.

Signed-off-by: Masatake YAMATO <yamato@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRead /etc/exports.d/*.export as extra export files
Masatake YAMATO [Mon, 7 Mar 2011 13:18:51 +0000 (08:18 -0500)]
Read /etc/exports.d/*.export as extra export files

This patch adding a capability to read /etc/exports.d/*.exports as
extra export files to exportfs.

If one wants to add or remove an export entry in a script, currently
one may have to use sed or something tool for adding or removing the
line for the entry in /etc/exports file.

With the patch, adding and removing an entry from a script is much
easier.
cat<<EOF... or mv can be used for adding. rm can be used for removing.

Signed-off-by: Masatake YAMATO <yamato@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoCleaned up a warning in rpcdispatch.c
Steve Dickson [Sat, 5 Mar 2011 21:17:01 +0000 (16:17 -0500)]
Cleaned up a warning in rpcdispatch.c

rpcdispatch.c:40:20: warning: comparison between signed and unsigned
integer expressions

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount: Remove MOUNT_CONFIG warnings
Steve Dickson [Sat, 5 Mar 2011 21:13:01 +0000 (16:13 -0500)]
mount: Remove MOUNT_CONFIG warnings

The following changes are needed to remove compile warnings when
MOUNT_CONFIG is not defined

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount: Recognize zero as a valid value for the port= option
Chuck Lever [Thu, 3 Mar 2011 22:26:33 +0000 (17:26 -0500)]
mount: Recognize zero as a valid value for the port= option

While zero is not a valid IP port number, zero does represent a valid
value for "port=".  It means "query rpcbind to discover the actual
non-zero port number to use".  So the parsing functions that handle
"port=" should not flag zero as an invalid value.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoGssd: modify wrong err message at handle_gssd_upcall
Mi Jinlong [Wed, 9 Feb 2011 16:29:42 +0000 (11:29 -0500)]
Gssd: modify wrong err message at handle_gssd_upcall

Modify wrong err message at handle_gssd_upcall when
sscanf encryption types fail.

Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoSupport AD style kerberos automatically in rpc.gss
Jason Gunthorpe [Wed, 9 Feb 2011 16:27:19 +0000 (11:27 -0500)]
Support AD style kerberos automatically in rpc.gss

An Active Directory KDC will only grant a TGT for UPNs, getting
a TGT for SPNs is not possible:

$ kinit -k host/ib5@ADS.ORCORP.CA
kinit: Client not found in Kerberos database while getting initial
credentials

The correct thing to do for machine credentials is to get a TGT
for the computer UPN <HOSTNAME>$@REALM:
$ kinit -k IB5\$
$ klist
12/22/10 11:43:47  12/22/10 21:43:47  krbtgt/ADS.ORCORP.CA@ADS.ORCORP.CA

Samba automatically creates /etc/krb5.keytab entry for the computer UPN,
this patch makes gssd_refresh_krb5_machine_credential prefer it above
the SPNs if it is present.

The net result is that nfs client works automatically out of the box
if samba has been used to setup kerberos via 'net ads join' 'net ads
keytab create'

Tested using Windows Server 2003 R2 as the AD server.

Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoFixed segfault in rpc.mountd
Steve Dickson [Wed, 26 Jan 2011 12:49:19 +0000 (07:49 -0500)]
Fixed segfault in rpc.mountd

A unallocated piece of memory, instead of a NULL point, was being
used to initialize a ->next point in the mount link list which
caused a segfault after a few remote accesses via the showmount
command.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoImprove debugging in svcgssd nfs-utils-1-2-4-rc5
Steve Dickson [Fri, 14 Jan 2011 15:12:28 +0000 (10:12 -0500)]
Improve debugging in svcgssd

Added in gss_display_error() which translates the GSS error into the
actual GSS macro name. Currently only the translation of these errors
are logged. Since those translations are buried deep in the kerberos
library code, having the actual GSS macro name makes it easier to
follow the code.

Moved the nfs4_init_name_mapping() call into main() so if debug is
enabled the DNS name and realms will be logged during start up.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnsm.a: modify return value to false from 0 at nsm_drop_privileges()
Mi Jinlong [Tue, 4 Jan 2011 16:16:45 +0000 (11:16 -0500)]
libnsm.a: modify return value to false from 0 at nsm_drop_privileges()

At nsm_drop_privileges(), for improving readability, unify
the return value.

Signed-off-by: Mi Jinlong <mijinlong@cn.fujitsu.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnsm.a: sm-notify sometimes ignores monitored hosts nfs-utils-1-2-4-rc4
Chuck Lever [Mon, 13 Dec 2010 19:50:45 +0000 (14:50 -0500)]
libnsm.a: sm-notify sometimes ignores monitored hosts

Monitored host information is stored in files under /var/lib/nfs.
When visiting entries in the monitored hosts directory, libnsm.a
examines the value of dirent.d_type to determine if an entry is a
regular file.

According to readdir(3), the d_type field is not supported by all
file system types.  My root file system happens to be one where d_type
isn't supported.  Typical installations that use an ext-derived root
file system are not exposed to this issue, but those who use xfs, for
instance, are.

On such file systems, not only are remote peers not notified of
reboots, but the NSM state number is never incremented.  A statd warm
restart would not re-monitor any hosts that were monitored before
the restart.

When writing support/nsm/file.c, I copied the use of d_type from the
original statd code, so this has likely been an issue for some time.

Replace the use of d_type in support/nsm/file.c with a call to
lstat(2).  It's extra code, but is guaranteed to work on all file
system types.

Note there is a usage of d_type in gssd.  I'll let gssd and rpcpipefs
experts decide whether that's worth changing.

Fix for:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=193

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnsm.a: Replace __attribute_noinline__
Chuck Lever [Mon, 13 Dec 2010 19:47:42 +0000 (14:47 -0500)]
libnsm.a: Replace __attribute_noinline__

Replace the __attribute_noinline__ form with

  __attribute__((__noinline__)).

Even though the compiler didn't complain about __attribute_malloc__,
also replace those in order to maintain consistent style throughout the
source file.

Fix for:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=194

Reported-by: "Gabor Z. Papp" <gzp@papp.hu>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agosm-notify: Make use of AI_NUMERICSERV conditional
Chuck Lever [Mon, 13 Dec 2010 19:36:15 +0000 (14:36 -0500)]
sm-notify: Make use of AI_NUMERICSERV conditional

Gabor Papp reports nfs-utils-1.2.3 doesn't build on his system that
uses glibc-2.2.5:

make[3]: Entering directory
`/home/gzp/src/nfs-utils-1.2.3/utils/statd'
gcc -DHAVE_CONFIG_H -I. -I../../support/include   -D_GNU_SOURCE -Wall
-Wextra -Wstrict-prototypes  -pipe -g -O2 -MT sm-notify.o -MD
-MP -MF .deps/sm-notify.Tpo -c -o sm-notify.o sm-notify.c
sm-notify.c: In function 'smn_bind_address':
sm-notify.c:247: error: 'AI_NUMERICSERV' undeclared (first use in this
function)
sm-notify.c:247: error: (Each undeclared identifier is reported only
once
sm-notify.c:247: error: for each function it appears in.)
make[3]: *** [sm-notify.o] Error 1

According to the getaddrinfo(3) man page, AI_NUMERICSERV is available
only since glibc 2.3.4.  getaddrinfo(3) seems to convert strings
containing a number to the right port value without the use of
AI_NUMERICSERV, so I think we can survive on older glibc's without it.
It will allow admins to specify service names as well as port numbers
on those versions.

There are uses of AI_NUMERICSERV in gssd and in nfs_svc_create().  The
one in nfs_svc_create() is behind HAVE_LIBTIRPC, and the other is a
issue only for those who want to deploy Kerberos -- likely in both
cases, a more modern glibc will be present.  I'm going to leave those
two.

Fix for:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=195

Reported-by: "Gabor Z. Papp" <gzp@papp.hu>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agorpc.mountd: Checking RPC Procedure ID before process it
Sid Moore [Fri, 3 Dec 2010 14:19:06 +0000 (09:19 -0500)]
rpc.mountd: Checking RPC Procedure ID before process it

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnfs.a: fix a bug when parse section's arg nfs-utils-1-2-4-rc3
Mi Jinlong [Mon, 29 Nov 2010 15:59:10 +0000 (10:59 -0500)]
libnfs.a: fix a bug when parse section's arg

When parsing section's arg at configure file, the pointer
should stop when fetch ']', and give the warning message.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: nfsstat: has_stats() does not function correctly for NFSv4 client stats
Harshula Jayasuriya [Mon, 22 Nov 2010 16:22:31 +0000 (11:22 -0500)]
nfs-utils: nfsstat: has_stats() does not function correctly for NFSv4 client stats

The NFSv4 client procs/ops in "struct rpc_procinfo nfs4_procedures" is
used to generate the NFS client stats interface:
------------------------------------------------------------
net 0 0 0 0
rpc 15 0 0
proc2 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
proc3 22 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 0
proc4 42 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
0 0 0
0 0 0 0 0 0 0
------------------------------------------------------------
Note, for proc4, the number 42. That is the number of stats that follow
on the same line. Currently nfsstat's has_stats() relies on this number
to be equal to CLTPROC4_SZ. Unfortunately this is not the case. I have
changed has_stats() not to rely on these two values being equal. This
should also allow nfsstat to work with different kernel versions that
expose a different number of NFS client ops.

* Fix has_stats()
* Stop print_clnt_list() printing server stats!
* Describe the option -3 and -4 completely in the nfsstat manpage.

Signed-off-by: Harshula Jayasuriya <harshula@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoEnable nfsidmap to compile
Steve Dickson [Sat, 20 Nov 2010 20:01:21 +0000 (15:01 -0500)]
Enable nfsidmap to compile

Only enable the compilation of nfsidmap when libnfsidmap support it.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoAdd the new nfsidmap program
Bryan Schumaker [Fri, 19 Nov 2010 17:01:10 +0000 (12:01 -0500)]
Add the new nfsidmap program

This patch adds the nfsidmap program to nfs-utils.  This program is
called by the nfs idmapper through request-keys to map between
uid / user name and gid / group name.

Signed-off-by: Bryan Schumaker <bjschuma@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved a couple warnings from utils/mount/stropts.c
Steve Dickson [Mon, 22 Nov 2010 16:33:37 +0000 (11:33 -0500)]
Removed a couple warnings from utils/mount/stropts.c

stropts.c:740:6: warning: 'ret' may be used uninitialized in this function
stropts.c:653:6: warning: 'ret' may be used uninitialized in this function

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs(5): Document remount behavior nfs-utils-1-2-4-rc2
Chuck Lever [Fri, 29 Oct 2010 16:56:21 +0000 (12:56 -0400)]
nfs(5): Document remount behavior

It appears that, for a long while, NFS "remount" mounts have
completely wiped the existing mount options in /etc/mtab for a given
mount point.  This is a problem for umount.nfs, since it reads its
options out of /etc/mtab to find out how to do the unmount.

The mount(8) command provides the NFS mount subcommand with the mount
options to perform the remount.  There are four cases to consider:

  1. Both the device and mount directory are specified on the
     command line, and the target mount point is in /etc/fstab

  2. Only one of the device and mount directory is specified on
     the command line, and the target mount point is in
     /etc/fstab

  3. Both the device and mount directory are specified on the
     command line, and the target mount point is not in /etc/fstab

  4. Only one of the device and mount directory is specified on
     the command line, and the target mount point is not in
     /etc/fstab

Currently only case 4 works correctly.  In that case, mount(8)
provides the correct set of mount options to the mount.nfs
subcommand and it can update /etc/mtab correctly.

Cases 1 and 3 replace all mount options in /etc/mtab with the options
provided on the command line during a remount.  Case 2 replaces the
mount options in /etc/mtab with a mix of options from /etc/fstab and
/etc/mtab.

Cases 1 and 3 are historical behavior.  Basically this is a formal
interface to allow administrators to replace the mount options in
/etc/mtab completely, instead of merging in new ones.  The present
patch documents that behavior in nfs(5), and provides best practice
for remounting NFS mount points.

There are near-term plans to address case 2 by fixing mount(8)
(provided by utils-linux-ng in most distributions).

This is a partial fix for:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=188

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs(5): Grammar and style fixes
Chuck Lever [Thu, 28 Oct 2010 17:15:22 +0000 (13:15 -0400)]
nfs(5): Grammar and style fixes

Clean up grammar and style issues introduced by recent updates.  Also,
I'm not certain inappropriate options are always ignored.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: mnt_freq and mnt_pass are always zero
Chuck Lever [Thu, 28 Oct 2010 17:13:19 +0000 (13:13 -0400)]
mount.nfs: mnt_freq and mnt_pass are always zero

Clean up.

No need to pass constant zeros to add_mtab() from its only call site.
Ensure that initialization of a struct mntent is consistent in both
places that it is done.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Fix memory leak in nfs_sys_mount()
Chuck Lever [Thu, 28 Oct 2010 17:10:48 +0000 (13:10 -0400)]
mount.nfs: Fix memory leak in nfs_sys_mount()

This appears to have been left behind by last year's adjustments to
how the extra_opts string is constructed.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount: Fix compiler warning in nfs_parse_retry_option()
Chuck Lever [Thu, 28 Oct 2010 17:09:38 +0000 (13:09 -0400)]
mount: Fix compiler warning in nfs_parse_retry_option()

stropts.c: In function nfs_parse_retry_option:
stropts.c:131: warning: conversion to unsigned int from long int may
alter its value

Make it more clear what the second argument is for, and flag the
switch fallthrough case.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Remove all uses of AI_ADDRCONFIG
Chuck Lever [Thu, 28 Oct 2010 16:12:12 +0000 (12:12 -0400)]
nfs-utils: Remove all uses of AI_ADDRCONFIG

It was reported that, if only "lo" is up,

  mount.nfs 127.0.0.1:/export /mount

fails with "Name or service not known".

"man 3 getaddrinfo" says this:

  If hints.ai_flags includes the AI_ADDRCONFIG flag, then IPv4
  addresses are returned in the list pointed to by res only if the
  local system has at least one IPv4 address configured, and IPv6
  addresses are only returned if the local system has at least
  one IPv6 address configured.

The man page oversimplifies here.  A review of glibc shows that
getaddrinfo(3) explicitly ignores loopback addresses when deciding
whether an IPv4 or IPv6 address is configured.

This behavior around loopback is a problem not just for mount.nfs,
but also for RPC daemons that have to start up before a system's
networking is fully configured and started.  Given the history of
other problems with AI_ADDRCONFIG and the unpredictable behavior it
introduces, let's just remove it everywhere in nfs-utils.

This fix addresses:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=191

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: fix default value for --enable-tirpc
Jeff Layton [Thu, 28 Oct 2010 13:18:33 +0000 (09:18 -0400)]
nfs-utils: fix default value for --enable-tirpc

We need $enable_tirpc to be a tristate. 'yes' means that someone
explicitly requested building with tirpc. 'no' means that it was
explicitly disabled. Anything else means that no one specified a value.

Fix it by setting the value to a blank string so that the default is
properly undefined.

Reported-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoUpdated rpc.mountd man page
Steve Dickson [Tue, 19 Oct 2010 19:54:35 +0000 (15:54 -0400)]
Updated rpc.mountd man page

Updated the rpc.mountd man page to no longer reference
v3 as the "newer" version and also mentioned v4 as
a supported version.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoCleared up the sync option in exportfs man page
Steve Dickson [Tue, 19 Oct 2010 19:54:04 +0000 (15:54 -0400)]
Cleared up the sync option in exportfs man page

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRemoved duplicate entries in export man page
Steve Dickson [Fri, 15 Oct 2010 21:20:28 +0000 (17:20 -0400)]
Removed duplicate entries in export man page

The man page's paragraphs about "refer=" and "replicas="
each appear twice.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Clear mountd registrations at start up nfs-utils-1-2-4-rc1
Chuck Lever [Thu, 14 Oct 2010 14:33:25 +0000 (10:33 -0400)]
mountd: Clear mountd registrations at start up

Clear stale MNT registrations before mountd tries to create fresh
listeners, to ensure that mountd starts.  This is also what statd does.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agobehavior as file systems that use the monolithic /sbin/mount command.
Chuck Lever [Wed, 13 Oct 2010 17:57:52 +0000 (13:57 -0400)]
behavior as file systems that use the monolithic /sbin/mount command.
See the MS_NOMTAB macro in utils-linux-ng/mount/mount.c.

Note that mount(8) has MS_USERS and MS_USER in the "nomtab" category
as well, but mount.nfs needs to record those values so that unmounting
a user-mounted NFS file system can work.

While we're here, fix some white space damage in fix_opts_string().

This is a partial fix for:

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=188

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoumount.nfs: Distinguish between nfs4 and nfs mounts
Chuck Lever [Wed, 13 Oct 2010 17:55:10 +0000 (13:55 -0400)]
umount.nfs: Distinguish between nfs4 and nfs mounts

Neil Brown reports that umount.nfs is still confused by "-t nfs -o
vers=4" mounts.

/etc/mtab can be confused.  /proc/mounts is authoritative on the
fstype of a mount.  Have umount.nfs consult it to determine which
mechanism to use for unmounting.  The code to read /proc/mounts was
lifted from the nfsstat command.

The code introduced by this patch may look like belt-n-suspenders, but
we have two use cases to consider:

  1.  Old kernels don't support the "vers=4" mount option, so
      umount.nfs must look for the "nfs4" fstype
  2.  Upcoming kernels may eliminate support the "nfs4" fstype, so
      umount.nfs must look for the "vers=4" mount option

Thus this logic checks for "nfs4" first then looks for the NFS version
setting.

Note that we could handle unmounting entirely in the kernel, but that
won't help older kernels that have this issue.

See:
  https://bugzilla.linux-nfs.org/show_bug.cgi?id=189

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: mountproto does not support RDMA
Chuck Lever [Wed, 13 Oct 2010 17:01:51 +0000 (13:01 -0400)]
mount.nfs: mountproto does not support RDMA

Clean up.  Our client does not support the MNT protocol on RDMA.

nfs_mount_protocol() isn't invoked for RDMA mounts (they are shunted
off before nfs_options2pmap() is invoked).  But in case it ever is,
it should return the expected response.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agogcc complained:
NeilBrown [Wed, 13 Oct 2010 16:08:41 +0000 (12:08 -0400)]
gcc complained:

client.c: In function 'init_netmask6':
client.c:181:1: warning: no return statement in function returning
non-void

and Suse' build system complained

I: Program returns random data in a function
E: nfs-utils no-return-in-nonvoid-function client.c:181

when I built without --enable-ipv6

Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warnings in utils/mount/network.c
Chuck Lever [Wed, 13 Oct 2010 16:02:32 +0000 (12:02 -0400)]
mount.nfs: Eliminate compiler warnings in utils/mount/network.c

Clean up.

network.c: In function get_socket:
network.c:431: warning: dereferencing type-punned pointer might break
    strict-aliasing rules

network.c: In function probe_bothports:
network.c:759: warning: dereferencing type-punned pointer might break
    strict-aliasing rules
network.c:762: warning: dereferencing type-punned pointer might break
    strict-aliasing rules

network.c: In function nfs_probe_statd:
network.c:775: warning: dereferencing type-punned pointer might break
    strict-aliasing rules

network.c: In function nfs_call_umount:
network.c:904: warning: dereferencing type-punned pointer might break
    strict-aliasing rules

network.c: In function nfs_ca_sockname:
network.c:1106: warning: dereferencing type-punned pointer might break
    strict-aliasing rules
network.c:1112: warning: dereferencing type-punned pointer might break
    strict-aliasing rules

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warning in utils/mount/parse_opt.c
Chuck Lever [Wed, 13 Oct 2010 15:59:30 +0000 (11:59 -0400)]
mount.nfs: Eliminate compiler warning in utils/mount/parse_opt.c

parse_opt.c: In function po_rightmost:
parse_opt.c:517: warning: conversion to int from unsigned int may
    change the sign of the result

"i" contains the function's result value, so it should be defined as
the same type as the function's return type.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warning in utils/mount/nfsumount.c
Chuck Lever [Wed, 13 Oct 2010 15:58:27 +0000 (11:58 -0400)]
mount.nfs: Eliminate compiler warning in utils/mount/nfsumount.c

Clean up.

nfsumount.c:374: warning: ISO C forbids omitting the middle term of
   a ?: expression

This is also probably harmless, but let's make the code unambiguous.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warning in utils/mount/nfsumount.c
Chuck Lever [Wed, 13 Oct 2010 15:56:58 +0000 (11:56 -0400)]
mount.nfs: Eliminate compiler warning in utils/mount/nfsumount.c

Clean up.

nfsumount.c:265: warning: no previous prototype for nfsumount

It's also a good idea if the compiler can ensure that the prototype
in nfsmount.h matches the actual function defined in nfsumount.c.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warnings in utils/mount/mount.c
Chuck Lever [Wed, 13 Oct 2010 15:54:49 +0000 (11:54 -0400)]
mount.nfs: Eliminate compiler warnings in utils/mount/mount.c

Clean up.

mount.c: In function parse_opt:
mount.c:354: warning: conversion to size_t from int may change the
    sign of the result
mount.c:354: warning: conversion to int from size_t may change the
    sign of the result
mount.c:359: warning: conversion to size_t from int may change the
    sign of the result
mount.c:359: warning: conversion to int from size_t may change the
    sign of the result
mount.c: In function parse_opts:
mount.c:374: warning: conversion to int from size_t may change the
    sign of the result
mount.c:377: warning: conversion to size_t from int may change the
    sign of the result

Character string lengths are usually size_t anyway.  We can easily
avoid the implicit type cast here.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warning in utils/mount/mount.c
Chuck Lever [Wed, 13 Oct 2010 15:53:44 +0000 (11:53 -0400)]
mount.nfs: Eliminate compiler warning in utils/mount/mount.c

Clean up.

mount.c: At top level:
mount.c:324: warning: no previous prototype for ?mount_usage?

mount_usage() has no callers outside of utils/mount/mount.c and no
prototype is provided in a header file.  Make it static.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warnings in utils/mount/version.h
Chuck Lever [Wed, 13 Oct 2010 15:50:57 +0000 (11:50 -0400)]
mount.nfs: Eliminate compiler warnings in utils/mount/version.h

Clean up.

In file included from mount.c:50:
version.h: In function linux_version_code:
version.h:48: warning: conversion to unsigned int from int may
    change the sign of the result
version.h:48: warning: conversion to unsigned int from int may
    change the sign of the result
version.h:48: warning: conversion to unsigned int from int may
    change the sign of the result

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warning in utils/mount/mount.c
Chuck Lever [Wed, 13 Oct 2010 15:44:23 +0000 (11:44 -0400)]
mount.nfs: Eliminate compiler warning in utils/mount/mount.c

Clean up.

In file included from mount.c:41:
mount_config.h:35: warning: no previous prototype for mount_config_opts

Functions defined in include files are usually declared as "static
inline," eliminating the need for a forward declaration.

While I was there, I also fixed the macro that prevents including
mount_config.h multiple times, and fixed some white space damage.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Eliminate compiler warnings
Chuck Lever [Wed, 13 Oct 2010 15:38:22 +0000 (11:38 -0400)]
mount.nfs: Eliminate compiler warnings

Clean up.

fstab.c: In function ?lock_mtab?:
fstab.c:385: warning: declaration of ?errsv? shadows a previous local
fstab.c:367: warning: shadowed declaration is here
fstab.c:407: warning: declaration of ?errsv? shadows a previous local
fstab.c:367: warning: shadowed declaration is here
fstab.c:417: warning: declaration of ?tries? shadows a previous local
fstab.c:325: warning: shadowed declaration is here
fstab.c:422: warning: declaration of ?errsv? shadows a previous local
fstab.c:367: warning: shadowed declaration is here

These are probably harmless.  Reusing a variable name, however, is a
little confusing to follow when reading the code.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoexport: Ensure that we free struct exportent->e_uuid
Trond Myklebust [Wed, 13 Oct 2010 15:27:21 +0000 (11:27 -0400)]
export: Ensure that we free struct exportent->e_uuid

Currently, the exportent->e_uuid is initialised in
support/nfs/exports.c:parseopts(), but it is never freed.

Also ensure that exportent->e_uuid is duplicated correctly in
dupexportent().

Adjusted to account for the new export_free() helper.

Also, e_uuid points to memory that is always allocated with strdup(3),
not with xstrdup().  Thus it must be freed via free(3) and not via
xfree().

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnfs.a: Allow multiple RPC listeners to share listener port number
Chuck Lever [Wed, 13 Oct 2010 15:22:07 +0000 (11:22 -0400)]
libnfs.a: Allow multiple RPC listeners to share listener port number

Normally, when "-p" is not specified on the mountd command line, the
TI-RPC library chooses random port numbers for each listener.  If a
port number _is_ specified on the command line, all the listeners
will get the same port number, so SO_REUSEADDR needs to be set on
each socket.

Thus we can't let TI-RPC create the listener sockets for us in this
case; we must create them ourselves and then set SO_REUSEADDR (and
other socket options) by hand.

Different versions of the same RPC program have to share the same
listener and SVCXPRT, so we have to cache xprts we create, and re-use
them when additional requests for registration come from the
application.

Though it doesn't look like it, this fix was "copied" from the legacy
rpc_init() function.  It's more complicated for TI-RPC, of course,
since a TI-RPC application can set up listeners with a nearly
arbitrary number of address families and socket types, not just the
two listeners that legacy RPC applications can set up (one for AF_INET
UDP and one for AF_INET TCP).

See:
  https://bugzilla.linux-nfs.org/show_bug.cgi?id=190

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Fixed typo in NFS man page
Steve Dickson [Wed, 13 Oct 2010 14:17:58 +0000 (10:17 -0400)]
nfs-utils: Fixed typo in NFS man page

Chuck pointed out there was a grammar typo in addition to the spelling
typo.  Here is a revised version of the patch.

Signed-off-by: Jim Rees <rees@umich.edu>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoFix style nits in atomicio.c
Steve Dickson [Wed, 13 Oct 2010 14:15:12 +0000 (10:15 -0400)]
Fix style nits in atomicio.c

Signed-off-by: Jim Rees <rees@umich.edu>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Move common code into support
Steve Dickson [Wed, 13 Oct 2010 14:09:53 +0000 (10:09 -0400)]
nfs-utils: Move common code into support

There are several source files and headers present in the ./utils/idmapd
directory which are also usable in a doimapd daemon. Because of this we
move that support into the support directory such that it can be shared by
both daemons.

Signed-off-by: Jim Rees <rees@umich.edu>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoThe kernel 2.6.37 has a add new mount option: local_lock.
Suresh Jayaraman [Wed, 29 Sep 2010 11:14:14 +0000 (07:14 -0400)]
The kernel 2.6.37 has a add new mount option: local_lock.
Document the new option in the nfs(5) man page.

Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoRevert "nfs-iostat.py: don't wait for an extra interval when given a count" nfs-utils-1-2-3
Steve Dickson [Tue, 28 Sep 2010 12:24:16 +0000 (08:24 -0400)]
Revert "nfs-iostat.py: don't wait for an extra interval when given a count"

This reverts commit 837796686ad8f9178c7b6855ada728a53ae511e3.

9 years agoRelease 1.2.3
Steve Dickson [Tue, 28 Sep 2010 11:38:27 +0000 (07:38 -0400)]
Release 1.2.3

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoAdded the -p <principal> flag to the svcgssd manpage
Eberhard Kuemmerle [Mon, 27 Sep 2010 17:24:48 +0000 (13:24 -0400)]
Added the -p <principal> flag to the svcgssd manpage

Signed-off-by: Eberhard Kuemmerle <E.Kuemmerle@fz-juelich.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agosvcgssd: Adding a <-p principal> flag
Eberhard Kuemmerle [Mon, 27 Sep 2010 17:16:23 +0000 (13:16 -0400)]
svcgssd: Adding a <-p principal> flag

Allow the principal that is used to get the machines creds definable
on the command like with the new '-p <principal>'. This is useful
in cluster environments.

Signed-off-by: Eberhard Kuemmerle <E.Kuemmerle@fz-juelich.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-iostat.py: don't wait for an extra interval when given a count
David Lecorfe [Mon, 27 Sep 2010 17:29:31 +0000 (13:29 -0400)]
nfs-iostat.py: don't wait for an extra interval when given a count

If I invoke the tool with an interval of 10 and a count of 2, it will:
- show the summary
- sleep 10s
- show the stats for the last 10s
- sleep 10s
- exit

Signed-off-by: David Lecorfe <dlecorfec@gmail.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfsd: Enable IPv6 support in rpc.nfsd again.
Chuck Lever [Mon, 27 Sep 2010 14:16:34 +0000 (10:16 -0400)]
nfsd: Enable IPv6 support in rpc.nfsd again.

Revert commit b2a3cd59 so that rpc.nfsd can create IPv6 listener
sockets for the kernel.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Update mountd/exportfs man pages to reflect IPv6 changes
Chuck Lever [Mon, 27 Sep 2010 14:14:34 +0000 (10:14 -0400)]
mountd: Update mountd/exportfs man pages to reflect IPv6 changes

Document IPv6 support in rpc.mountd and exportfs, and clarify existing
language in the man page.

Clean up: Use bold consistently for program names, and italics
consistently for file names.  Use "rpc.mountd" consistently as the
name of the mountd daemon.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Use MNT status values instead of NFSERR
Chuck Lever [Mon, 27 Sep 2010 14:13:39 +0000 (10:13 -0400)]
mountd: Use MNT status values instead of NFSERR

Clean up:  The MNT protocol has its own enum type defining error
status values.  While the values can be the same as the NFSERR enum
type on some systems, it's not guaranteed to be true everywhere.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Fix up version and usage messages
Chuck Lever [Mon, 27 Sep 2010 14:11:18 +0000 (10:11 -0400)]
mountd: Fix up version and usage messages

Clean up: rpc.mountd is no longer known as kmountd.  Use the program's
basename rather than the full pathname for the usage message.  Display
a version message at start up similar to statd's.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Unregister mountd if my_svc_run() returns
Chuck Lever [Mon, 27 Sep 2010 14:09:49 +0000 (10:09 -0400)]
mountd: Unregister mountd if my_svc_run() returns

Fix a long standing bug: when my_svc_run() returns, mountd should
unregister itself with the local rpcbind so that it can subsequently
start cleanly.

Log a more helpful error message in this case.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Support TI-RPC mountd listener
Chuck Lever [Mon, 27 Sep 2010 14:09:10 +0000 (10:09 -0400)]
mountd: Support TI-RPC mountd listener

If TI-RPC is available, use it to create mountd's svc listener.  If
not, use the old function, rpc_init(), to create mountd's listener.

IPv6 can be supported if TI-RPC is available.  In this case,
/etc/netconfig is searched to determine which transports to advertise.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Make NFS version checks more strict
Chuck Lever [Mon, 27 Sep 2010 14:06:35 +0000 (10:06 -0400)]
mountd: Make NFS version checks more strict

Ensure users and programmers specify NFS version numbers correctly.
This also makes the next patch more clean.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agonfs-utils: Fix source code character encoding
Chuck Lever [Mon, 27 Sep 2010 13:50:11 +0000 (09:50 -0400)]
nfs-utils: Fix source code character encoding

Minor clean up.

Most modern Linux distributions set UTF-8 locales.  Standardize the
character encoding of source files on UTF-8, to squelch vim com-
plaints.

I probably missed a few spots.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnfs.a: Remove support/nfs/fstab.c
Steve Dickson [Mon, 27 Sep 2010 13:35:26 +0000 (09:35 -0400)]
libnfs.a: Remove support/nfs/fstab.c

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agorpc.nfsd: mount up nfsdfs is it doesn't appear to be mounted yet nfs-utils-1-2-3-rc6
Jeff Layton [Thu, 16 Sep 2010 18:34:39 +0000 (14:34 -0400)]
rpc.nfsd: mount up nfsdfs is it doesn't appear to be mounted yet

There's a bit of a chicken and egg problem when nfsd is run the first
time. On Fedora/RHEL at least, /proc/fs/nfsd is mounted up whenever nfsd
is plugged in via a modprobe.conf "install" directive.

If someone runs rpc.nfsd without plugging in nfsd.ko first,
/proc/fs/nfsd won't be mounted and rpc.nfsd will end up using the legacy
nfsctl interface. After that, nfsd will be plugged in and subsequent
rpc.nfsd invocations will use that instead.

This is a problem as some nfsd command-line options are ignored when the
legacy interface is used. It'll also be a problem for people who want
IPv6 enabled servers. The upshot is that we really don't want to use the
legacy interface unless there is no other option.

To avoid this situation, have rpc.nfsd check to see if the "threads"
file is already present. If it's not, then make an attempt to mount
/proc/fs/nfsd.  This is a "best-effort" sort of thing, however so we
just ignore the return code from the mount attempt and fall back to
using nfsctl() if it fails.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibexport.a: Enable IPv6 support in hostname.c
Chuck Lever [Thu, 16 Sep 2010 18:25:52 +0000 (14:25 -0400)]
libexport.a: Enable IPv6 support in hostname.c

If --enable-ipv6 is specified when building nfs-utils, libexport's
host_foo() helpers can now return both IPv4 and IPv6 addresses.

This means IPv6 presentation addresses and IPv6 DNS resolution
results are handled properly in the mountd cache and /etc/exports,
but does not yet enable IPv6 mountd listeners.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Ensure cache downcall can handle IPv6 addresses
Chuck Lever [Thu, 16 Sep 2010 18:21:48 +0000 (14:21 -0400)]
mountd: Ensure cache downcall can handle IPv6 addresses

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Handle IPv6 addresses in kernel auth_unix_ip upcalls
Chuck Lever [Thu, 16 Sep 2010 18:21:08 +0000 (14:21 -0400)]
mountd: Handle IPv6 addresses in kernel auth_unix_ip upcalls

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: clean up cache API
Chuck Lever [Thu, 16 Sep 2010 18:19:19 +0000 (14:19 -0400)]
mountd: clean up cache API

Clean up: Squelch compiler warnings and document public parts of
cache API.

cache.c: At top level:
cache.c:67: warning: no previous prototype for auth_unix_ip
cache.c:123: warning: no previous prototype for auth_unix_gid
cache.c:217: warning: no previous prototype for get_uuid
cache.c:247: warning: no previous prototype for uuid_by_path
cache.c:326: warning: no previous prototype for nfsd_fh
cache.c:745: warning: no previous prototype for nfsd_export
cache.c:820: warning: no previous prototype for cache_open
cache.c:832: warning: no previous prototype for cache_set_fd
cache.c:841: warning: no previous prototype for
cache_process_req
cache.c:921: warning: no previous prototype for cache_export
cache.c:953: warning: no previous prototype for
cache_get_filehandle

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agoexportfs: Enable IPv6 support in matchhostname()
Chuck Lever [Thu, 16 Sep 2010 18:17:08 +0000 (14:17 -0400)]
exportfs: Enable IPv6 support in matchhostname()

To gain IPv6 support in matchhostname(), simply replace the socket
address comparison helpers with the generic versions that can handle
IPv4 and IPv6.

host_addrinfo() (called by matchhostname()) returns IPv6 addresses
only if IPv6 support is enabled.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Support IPv6 in mountlist_list()
Chuck Lever [Thu, 16 Sep 2010 17:54:21 +0000 (13:54 -0400)]
mountd: Support IPv6 in mountlist_list()

Replace inet_aton(3) and gethostbyaddr(3) calls in mountlist_list()
with calls to the new host_foo() DNS helpers.

The new functions will support IPv6 without additional changes, once
IPv6 is enabled in the generic hostname helpers.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Handle memory exhaustion in mountlist_list()
Chuck Lever [Thu, 16 Sep 2010 17:51:51 +0000 (13:51 -0400)]
mountd: Handle memory exhaustion in mountlist_list()

I'm about to replace inet_aton(3)/gethostbyaddr(3) with
host_pton()/host_canonname() in mountlist_list().

Since host_canonname() returns a string allocated with strdup(3)
instead of xstrdup(), mountlist_list() must now deal with memory
exhaustion properly.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Add mountlist_freeall()
Chuck Lever [Thu, 16 Sep 2010 17:48:38 +0000 (13:48 -0400)]
mountd: Add mountlist_freeall()

I'm about to add a second bit of logic that needs to free all
mountlist records, so introduce a helper for freeing them.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: support IPv6 in mountlist_del_all()
Chuck Lever [Thu, 16 Sep 2010 17:46:26 +0000 (13:46 -0400)]
mountd: support IPv6 in mountlist_del_all()

Replace IPv4-specific code in the mountlist_del_all() path with code
that is address family agnostic.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: Support IPv6 in mountd's svc routines
Chuck Lever [Thu, 16 Sep 2010 17:44:02 +0000 (13:44 -0400)]
mountd: Support IPv6 in mountd's svc routines

Replace IPv4-specific code with use of our generic hostname helpers
in the routines that handle incoming MNT RPC requests.

These functions will support IPv6 without additional changes, once
IPv6 is enabled in the generic hostname helpers.

As part of this update, I've modified all of mountd's _svc routines
to use a debug message format that is consistent with statd.  It may
be overkill for some of these; if so we can pull them out later.

Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomountd: add IPv6 support in auth_authenticate()
Chuck Lever [Thu, 16 Sep 2010 17:39:39 +0000 (13:39 -0400)]
mountd: add IPv6 support in auth_authenticate()

Make the entire auth_authenticate() code path address-family agnostic.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agolibnfs.a: Fix API for getfh() & friends
Chuck Lever [Thu, 16 Sep 2010 13:32:52 +0000 (09:32 -0400)]
libnfs.a: Fix API for getfh() & friends

This is more of a clean-up than a behavioral change.

POSIX requires that a "struct sockaddr" is the same size as a "struct
sockaddr_in".  Therefore, a variable or field of type "struct sockaddr"
cannot contain an AF_INET6 address.  However, "struct sockaddr *" is
often used to reference a generic (ie non-address family specific)
socket address, generating some confusion about this.

The nfsctl_arg struct uses a struct sockaddr (not a pointer) to pass
the client's IP address to the kernel.  This means the legacy nfsctl()
kernel API can never support IPv6.  Fortunately for us, this legacy
interface was replaced by a text-based cache interface a few years
back.  We don't need to support non-AF_INET addresses here.

The getfh() functions in nfs-utils provide a handy C API for the
kernel's nfsctl interface.  The getfh() functions still take a struct
sockaddr *, though, and that can imply that a non-IPv4 address can be
passed via this API.  To make it abundantly clear that only IPv4
addresses can be used with this interface, change the synopses of
getfh() and friends to take a struct sockaddr_in * instead of a struct
sockaddr * .

This makes these functions conform with other places in mountd and
exportfs that already grok the difference between a struct sockaddr
and a struct sockaddr_in.

While we're here...

Introduce some nice documenting comments for the get_fh() functions,
and...

Since mountd will support IPv6 in the near future, assert that the
family of client addresses passed to this API is indeed AF_INET, in
order to prevent non-AF_INET addresses from ever being passed to the
legacy nfsctl() interface.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Don't do anything fancy if this is a remount
Chuck Lever [Thu, 16 Sep 2010 11:28:18 +0000 (07:28 -0400)]
mount.nfs: Don't do anything fancy if this is a remount

We don't want to append "vers=4" or perform any negotiation if the
"remount" mount option was specified.  It will just end in tears.

This attempts to address

  https://qa.mandriva.com/show_bug.cgi?id=60311

and

  https://bugzilla.linux-nfs.org/show_bug.cgi?id=187

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agomount.nfs: Refactor mount version and protocol autonegotiation
Chuck Lever [Thu, 16 Sep 2010 11:26:07 +0000 (07:26 -0400)]
mount.nfs: Refactor mount version and protocol autonegotiation

Clean up.

I'm beginning to agree with Bruce and Steve's assessment that the
fallthrough switch case in nfs_try_mount() is more difficult to read
and understand than it needs to be.  The logic that manages
negotiating NFS version and protocol settings is getting more complex
over time anyway.

So let's split the autonegotiation piece out of nfs_try_mount().

We can reduce indenting, and use cleaner switch-based logic.  Also,
adding more comments can only help.

Neil also suggested replacing the pre-call "errno = 0" trick.  The
lower-level functions may try to mount several times (given a list of
addresses to try).  errno could be set by any of those.  The mount
request will succeed at some point, and "success" is returned, but
errno is still set to some non-zero value.

The kernel version check in nfs_try_mount() is more or less loop
invariant: it's impossible for the result of that test to change
between retries.  So we should be able to safely move it to the logic
that sets the initial value of mi->version.

This patch is not supposed to cause a behavioral change.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
9 years agorpc.idmapd: Type of idmap client should be defined by ic_id not ic_clid.
Bian Naimeng [Thu, 16 Sep 2010 11:12:47 +0000 (07:12 -0400)]
rpc.idmapd: Type of idmap client should be defined by ic_id not ic_clid.

The type of idmap_client is defined by idmap_client.ic_id for nfs,
so nfsd should have the same style.

Signed-off-by: Bian Naimeng <biannm@cn.fujitsu.com>
Signed-off-by: Steve Dickson <steved@redhat.com>