Martin Leisner [Fri, 25 Jul 2008 18:50:06 +0000 (14:50 -0400)]
showmount issues
The connect_nb() routne returns zero for success and a negative
value for failure which was not being interpreted correctly
by the getport() routine. This patch fixes that problem.
Steve Dickson [Fri, 25 Jul 2008 18:15:47 +0000 (14:15 -0400)]
sm-notify: perform DNS lookup in the background.
If an NFS server has no network connectivity when it reboots,
it will block in sm-notify waiting for DNS lookup for a potentially
large number of hosts. This is not helpful and just annoys the
sysadmin.
So do the DNS lookup in the backgrounded phase of sm-notify,
before sending off the NOTIFY requests.
Acked-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Neil Brown [Wed, 16 Jul 2008 17:28:52 +0000 (13:28 -0400)]
If portmap is not listening on UDP (as apparently happens with
MS-Windows-Server2003R2SP2), then nfs mounts have to be mounted
with -o mountproto=tcp to succeed.
In this case a umount will still try UDP and will fail to contact the
server. It will still succeed with the local unmount (after a
timeout) but exits with a non-zero exit status. This causes
/bin/mount to retry so we get a strange error about the filesystem
not being mounted.
So:
get umount to use tcp if "mountproto=tcp" appears in mtab
ignore any failure message from the server that would overwrite
a success message from the local umount syscall.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Neil Brown [Wed, 16 Jul 2008 17:15:46 +0000 (13:15 -0400)]
If an NFS server is only listening on TCP for portmap (as apparently
MS-Windows-Server2003R2SP2 does), mount doesn't cope. There is retry
logic in case the initial choice of version/etc doesn't work, but it
doesn't cope with mountd needing tcp.
So:
Fix probe_port so that a TIMEDOUT error doesn't simply abort
but probes with other protocols (e.g. tcp).
Fix rewrite_mount_options to extract the mountproto option before
doing a probe, then set mountproto (and mount prot) based
on the result.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Tue, 15 Jul 2008 18:43:00 +0000 (14:43 -0400)]
It appears that a recent glibc update now enforces the requirement for a mode
parameter for open calls with the O_CREAT flag set. nfs-utils support code
defines a function xflock used by exportfs and mountd that calls open with
O_CREAT but no mode parameter. This causes exportfs and mountd to dump core,
with the error message:
*** invalid open64 call: O_CREAT without mode ***:rpc.mountd terminated
Chuck Lever [Tue, 15 Jul 2008 17:51:07 +0000 (13:51 -0400)]
Traditionally the mount command has looked for a ":" to separate the
server's hostname from the export path in the mounted on device name,
like this:
mount server:/export /mounted/on/dir
The server's hostname is "server" and the export path is "/export".
You can also substitute a specific IPv4 network address for the server
hostname, like this:
mount 192.168.0.55:/export /mounted/on/dir
Raw IPv6 addresses present a problem, however, because they look something
like this:
fe80::200:5aff:fe00:30b
Note the use of colons.
To get around the presence of colons, copy the Solaris convention used for
raw NFS server IPv6 addresses, which is to wrap the raw IPv6 address with
square brackets. This is also suggested in RFC 4038.
Introduce a new device name parser that can support traditional device
names and square brackets. Place the parser in a separate source file
so both the mount and umount paths can derive the server's hostname and
export pathname the same way.
Bonus points: add a check for NFS URLs and display an appropriate error
message in that case. This is cleaner than failing with "unknown host:
nfs".
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 17:37:07 +0000 (13:37 -0400)]
Change the append_clientaddr_option() function to support sending either
IPv4 or IPv6 addresses to the kernel via the "clientaddr=" option.
If the mount.nfs4 command can't determine an appropriate callback address,
it used to fail the mount request. This new function simply sends an ANY
address instead, so the mount request succeeds, but delegation is disabled.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 17:33:32 +0000 (13:33 -0400)]
There are three helpers that convert sockaddr-style addresses to text
addresses, then construct mount options to pass these addresses to the
kernel. The tail of each of these helpers does exactly the same thing,
so introduce a helper that handles the common code.
Magically, the new helper supports IPv6 as well as IPv4.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 17:31:17 +0000 (13:31 -0400)]
Introduce IPv6-enabled version of get_client_address. The legacy mount
command could use this eventually as well.
If this new function fails to discover an appropriate callback address, it
fills in an ANY address to indicate to the server that it should not call the
client back (ie delegations are disabled in this case).
The user can specify a callback address via the clientaddr= mount option in
this case to enable delegation.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 17:23:58 +0000 (13:23 -0400)]
Add #include directives for additional header files needed to support IPv6
networking. This is a separate patch so subsequent
patches can be reordered without collision.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 17:20:01 +0000 (13:20 -0400)]
We want to continue to support building nfs-utils on systems that do not
have IPv6-enabled RPC libraries and headers installed, so add a
./configure switch that allows distros to disable IPv6 functionality.
This patch introduces the nfs-utils autotools configuration to the library
and header dependencies that will be required in subsequent patches.
Later patches can then be reordered more easily if these new dependencies
are added in one heap.
For now, --enable-ipv6 defaults to "no", so this patch should not result in
any behavioral changes to the nfs-utils build process, by default.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 16:17:19 +0000 (12:17 -0400)]
Currently the "-s" option is ignored by the text-based mount interface. To
notify the kernel that sloppy mount option parsing is needed, add "sloppy"
to the string of mount options passed to the kernel.
The 2.6.23 - 2.6.26 kernels will fail the mount if "sloppy" is present, as
they won't recognize it. To prevent them from ever seeing this option,
have the mount command check the kernel version before appending the option.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 16:15:29 +0000 (12:15 -0400)]
Lots of parts of nfs-utils already depend on getaddrinfo(3).
We could find each instance where getaddrinfo(3) is invoked, wrap it with
'#ifdef HAVE_GETADDRINFO', and provide equivalent logic without it, but that's
a whole lot of work... and no-one has complained about this so far.
So as a clean-up, let's simply add a hard dependency for it in configure.ac,
and call it a day.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 15 Jul 2008 15:59:03 +0000 (11:59 -0400)]
The text-based mount command displays the rather inexplicable "mount:
internal error" whenever it encounters a problem that is entirely
unexpected by its designers.
Let's beef that error message up to include instructions about reporting
the problem, and fix the error code returned by the mount option rewriting
logic so that also will no longer report "internal error". An error in there
should generally only occur if there was an invalid mount option specified.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Ported the create_mtab() routine from util-linux-ng as well
some add_mtab() updates to better hand the instances where
/etc/mtab does not exist or is not writable
Signed-off-by: Christiaan Welvaart <cjw@daneel.dyndns.org> Signed-off-by: Steve Dickson <steved@redhat.com>
The rpc.gssd scans for any suitable kerberos ticket. In cross-realm
environment this may not be the desired behaviour. Therefore a new
option, -R preferred realm, is presented so that the rpc.gssd prefers tickets
from this realm. By default, the default realm is preferred.
Signed-off-by: Lukas Hejtmanek <xhejtman@ics.muni.cz> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Steve Dickson <steved@redhat.com>
The default expiration of kernel gss contexts is the expiration
of the Kerberos ticket used in its creation. (For contexts
created using the Kerberos mechanism.) Thus kdestroy has
no effect in nullifying the kernel context.
This patch adds -t <timeout> option to rpc.gssd so that the client's
administrator may specify a timeout for expiration of contexts in kernel.
After this timeout, rpc.gssd is consulted to create a new context.
By default, timeout is 0 (i.e., no timeout at all) which follows the
previous behavior.
Signed-off-by: Lukas Hejtmanek <xhejtman@ics.muni.cz> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Steve Dickson <steved@redhat.com>
gssd_setup_krb5_user_gss_ccache must return an error if no usable cache is
found. Trying to use invalid default cache and continue is not good idea at all.
Signed-off-by: Lukas Hejtmanek <xhejtman@ics.muni.cz> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Wed, 25 Jun 2008 13:23:45 +0000 (09:23 -0400)]
When a FQDN exists in /var/lib/nfs/rmtab it causes
the exportfs command to seg fault due to the nfs_export pointer
not being allocated. Reworking the parentheses in rmtab_read()
so the htype variable is evaluated correctly fix the problem.
Chuck Lever [Mon, 23 Jun 2008 16:52:33 +0000 (12:52 -0400)]
The "mountstats" utility is a Python program that extracts and displays NFS
client performance information from /proc/self/mountstats.
Note that if mountstats is named 'ms-nfsstat' or 'ms-iostat' it offers
slightly different functionality. It needs two man pages and the install
script should provide both commands by installing the script and providing the
other command via a symlink.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Jeff Layton [Mon, 23 Jun 2008 11:21:52 +0000 (07:21 -0400)]
The nfsstat program reads /proc/net/rpc/* files to gets info about
calls. This info is output as unsigned numbers (at least on any
relatively recent kernel). When nfsstat prints these numbers, they are
printed as signed integers. When the call counters reach 2^31, things
start being printed as negative numbers.
This patch changes nfsstat to read and print all counters as unsigned
integers. Tested by hacking up a kernel to initialize call counters to
2^31+1.
Thanks to Takafumi Miki for the initial version of this patch.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Neil Brown [Fri, 6 Jun 2008 19:17:55 +0000 (15:17 -0400)]
nfsstat -m lists all current nfs mounts, with the mount options.
It does this by reading /proc/mounts and looking for mounts of type
"nfs". It really should check for "nfs4" as well.
For simplicity, just check the first 3 characters of the type.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Fri, 6 Jun 2008 19:06:21 +0000 (15:06 -0400)]
Clean up: instead of passing so many arguments to all the helpers, have
nfsmount_string build a data structure that contains all the arguments, and
pass a pointer to that instead.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Fri, 6 Jun 2008 19:02:18 +0000 (15:02 -0400)]
Steinar Gunderson reports:
"It seems retry= is now additive with the text-based mount interface. In
particular, "mount -o retry=0" still gives a two-minute timeout."
Correct the bug and make retry= option parsing more robust. If parsing
the retry option fails, the option is ignored and a default timeout is
used.
Note that currently the kernel parser ignores the "retry=" option if the
value is a number. If the value contains other characters, the kernel will
choke. A subsequent patch to the kernel will allow any characters as the
value of the retry option (excepting of course ",").
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
NeilBrown [Thu, 8 May 2008 09:18:25 +0000 (05:18 -0400)]
If mount.nfs is not installed setuid, an attempt to perform a "user"
or "users" mount will fail with a fairly obscure error message,
typically about getting "permission denied" from the server.
This patch gives a more helpful message in that case.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Check the info file nfs/rpc_pipefs/nfs/clnt?/info to
see if a port number was supplied. If so, use it rather
than the default port number.
Signed-off-by: Olga Kornievskaia <aglo@citi.umich.edu> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Steve Dickson <steved@redhat.com>
Jeff Layton [Wed, 7 May 2008 14:37:40 +0000 (10:37 -0400)]
The prev_bg_host stuff made sense when NFS didn't have its own mount
handler. Now though, each mount.nfs invocation is really a one-shot
affair, and this check no longer works. It also leaked memory. Remove
it.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Jeff Layton [Wed, 7 May 2008 14:35:30 +0000 (10:35 -0400)]
The bg option is essentially ignored with nfs4 currently. nfs4mount()
will never exit with EX_BG, so the mount will never be backgrounded.
Fix it so that when bg is specified that we error out with EX_BG as
soon as possible after the first failed mount attempt.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Jeff Layton [Wed, 7 May 2008 14:27:53 +0000 (10:27 -0400)]
Currently nfs4mount() sets the retry value to 10000 on both fg and bg
mounts. It should be 2 for fg and 10000 for bg. nfsmount() sets it
properly, but there is a potential corner case. If someone explicitly
sets retry=10000 on a fg mount, then it will be reset to 2.
Fix this by having retry default to -1 for both flavors, and then reset if
needed after the mount options have been parsed.
Signed-off-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Jeff Layton [Mon, 14 Apr 2008 13:03:13 +0000 (09:03 -0400)]
Change how mount.nfs handles EACCES errors. Currently,
EACCES is a non-fatal error which means the mount will be
retied. This caused mounts to hang for 2mins when the client
does not have permission to access the export. In a strict
interpretation, the error that should be returned is EPERM, but
this is not always the case. So due to the fuzzy interpretation,
of EPERM and EACCES, EACCESS is now a fatal error
bc Wong [Tue, 18 Mar 2008 13:30:44 +0000 (09:30 -0400)]
There were 2 things wrong with auth flavour ordering:
- Mountd used to advertise AUTH_NULL as the first flavour on
the list, which means that it prefers AUTH_NULL to anything
else (as per RFC 2623 section 2.7).
- Mount.nfs used to scan the returned list in reverse order,
and stopping at the first AUTH_NULL or AUTH_SYS encountered.
If a server advertises (AUTH_SYS, AUTH_NULL), it will by
default choose AUTH_NULL and have degraded access.
I've fixed mount.nfs to scan from the beginning. For mountd,
it does not advertise AUTH_NULL anymore. This is necessary
to avoid backward compatibility issue. If AUTH_NULL appears
in the list, either the new or the old client will choose
that over AUTH_SYS.
Tested the server/client combination against the previous
versions, as well as Solaris and FreeBSD.
Signed-off-by: bc Wong <bcwong@cisco.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Tue, 11 Mar 2008 17:11:50 +0000 (13:11 -0400)]
Recently #include directives for autoconf's config.h file were added in
utils/mount/error.c and utils/mount/mount.c, but appropriate HAVE_CONFIG_H
checks were not added at the same time.
In addition, several other .c files under utils/mount reference
autoconf-generated HAVE_ macros, but don't appear to include config.h
Also, Heinz-Ado Arnolds <arnolds@MPA-Garching.MPG.DE> reports that this
patch is needed to ensure START_STATD is properly defined in
utils/mount/network.c. Otherwise start_statd() is always a no-op, even if
the configure script defines an appropriate statd start-up script.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Cc: Heinz-Ado Arnolds <arnolds@MPA-Garching.MPG.DE> Signed-off-by: Steve Dickson <steved@dickson.boston.devel.redhat.com>
The mount(5) man page states that the noquota, quota, usrquota and
grpquota options are ignored. (They are, however, used by the quota
tools, so having them in fstab can be useful.) Make mount.nfs ignore
them properly, matching the man page. There are a few aliases (like
usrjquota) that are parsed by quota, but as these are not documented
nor seem to be widely used, they are not included.
Signed-off-by: Steinar H. Gunderson <sgunderson@bigfoot.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Chuck Lever [Wed, 5 Mar 2008 15:07:11 +0000 (10:07 -0500)]
As part of migrating from nfs@lists.sf.net to linux-nfs@vger.kernel.org,
update the mailing list address used to report bugs in nfs-utils.
Removed the BUGS section in the mount.nfs and umount.nfs man pages since
they weren't consistent with the contents of the BUGS sections in others
in nfs-utils.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
NeilBrown [Tue, 26 Feb 2008 18:57:39 +0000 (13:57 -0500)]
If validateascii is passed a string containing only non-zero 7bit
values, then the loop with exit with i == len, and the following
test will access beyond the end of the array.
So add an extra test to fix this.
Found by Marcus Meissner <meissner@novell.com>.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
In mountd, if get_exportlist() (utils/mountd/mountd.c) returns NULL it
should not be considered a failure. It just means that there are no
exports on the system.
The practical problem with the current code is that a showmount -e
results in a syslog message from mountd that looks like:
rpc.mountd: export request from 10.250.100.2 failed.
Steve Dickson [Sat, 19 Jan 2008 12:59:26 +0000 (07:59 -0500)]
Fix bug when both crossmnt and fsid are set.
When exported a filesystems with option inherited (by the crossmnt
option) from a higherlevel filesystem, ignore filesystem specific
options like FSID and explicit UUID.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Thu, 17 Jan 2008 16:25:37 +0000 (11:25 -0500)]
Stop segfaults on amd64 during warnings messages by creating
a second va_list in xlog_backend() and then use that va_list
to print messages on stderr.
Signed-off-by: Steinar H. Gunderson <sesse@debian.org> Signed-off-by: Steve Langasek <vorlon@debian.org> Signed-off-by: Steve Dickson <steved@dickson.boston.devel.redhat.com>
Steve Dickson [Wed, 16 Jan 2008 16:53:41 +0000 (11:53 -0500)]
Address an inconsistency: the mount.nfs command uses the glibc routines
to manipulate /etc/mtab (setmntent) but, everything else in nfs-utils
uses a local private version (nfs_setmntent). The local version does
some extra mangling of the mtab entries.
We should check what util-linux does these days to be sure, but for now,
let's make the mount.nfs command use the nfs_ variants of setmntent().
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Sat, 3 Nov 2007 14:40:36 +0000 (10:40 -0400)]
When mountd gets a request to export a mountpoint which is not
explicitly exported, but is below an export point that is flagged as
"crossmnt", it passes the wrong path name to the kernel for the
"filehandle -> directory"
mapping.
This can badly confuse the NFS client, and is certainly wrong.
So use the correct path names.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Sat, 3 Nov 2007 14:36:36 +0000 (10:36 -0400)]
When following a list of mount versions to probe -
e.g. probe_mnt1_first or probe_mnt3_first - probe_both will first
probe the appropriate NFS version and then, if that succeeds, probe
the actual mount version. However instead of probing the target mount
version, it probes the "most appropriate" mount version for the given NFS version.
This results in it probing:
NFSv2, MOUNTv1
twice rather than
NFSv2, MOUNTv1
NFSv2, MOUNTv2
as would be more correct.
This patch removes the "choose most correct" step and just use the
current mouint version for the probe_vers array.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Steve Dickson [Sat, 3 Nov 2007 13:35:05 +0000 (09:35 -0400)]
This means that if mountd is run with "--no-nfs-version 3",
It will first probe for NFS version 3, which will succeed (assuming the
kernel supported NFSv3), then it will check the matching mountd version (3)
and probe_port on discovering that isn't supported will try other versions,
find "1" is supported will succeed.
This leaves up using mount version 1 for an NFSv3 mount, which doesn't work
and leads to a SIGSEGV
There is no case where trying other versions is needed the request one is
not supported, so simply remove that code.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Steve Dickson <steved@redhat.com>
Kevin Coffman [Fri, 12 Oct 2007 20:35:25 +0000 (16:35 -0400)]
Remove unnecessary code from idmapd.
This patch removes unnecessary code from idmapd. setproctitle is not used
anywhere and it can be removed. In addition the kernel section of the
nfs_idmap.h header is not used and is out of date and thus is removed.
Signed-off-by: David P. Quigley <dpquigl@tycho.nsa.gov> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de>
Kevin Coffman [Fri, 12 Oct 2007 20:35:20 +0000 (16:35 -0400)]
Remove old logging implementation for idmapd and rework gssd and idmapd to use the new xlog logging infrastructure.
This patch removes all of the old idmap_* logging functions and replaced them
with the corresponding xlog functions. In addition that that it also reworks
the gssd logging wrappers to use the new xlog_backend. Finally it makes
necessary changes to the build files to get the project compiling again.
Signed-off-by: David P. Quigley <dpquigl@tycho.nsa.gov> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de>
Kevin Coffman [Fri, 12 Oct 2007 20:35:15 +0000 (16:35 -0400)]
Cleanup xlog logging code to be safe and usable for all
This patch reworks the xlog logging code to avoid rebuilding the message into a
fixed size buffer. It also adds two new logging functions xlog_warn and
xlog_err which are replacements for idmap_warn and idmap_err. There use to be
two different variates of these functions with the only difference being that
one flavor tacked on the error string to the end of the message. This
responsibility has been pushed to the called of the function since it
needlessly complicated the function and required us to rebuild the message
strings.
Signed-off-by: David P. Quigley <dpquigl@tycho.nsa.gov> Signed-off-by: Kevin Coffman <kwc@citi.umich.edu> Signed-off-by: Neil Brown <neilb@suse.de>
projects / nfs-utils.git / log