| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit 76f8ce8c (statd: Update existing record if we receive SM_MON with
new cookie) added some logic to unconditionally delete some existing
on-disk monitor records. That works fine in an HA-NFS setup where
there's a good chance of monitor files being left around after service
failovers, but in the case where there isn't an existing monitor file
statd emits a scary looking message like this:
Jun 15 14:14:59 hostname rpc.statd[1368]: Failed to delete: could not
stat original file /var/lib/nfs/statd/sm/nfs.smayhew.test: No such file
or directory
That message can be suppressed.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
| |
This prevents rpc.statd's in-memory (and on-disk) monitor lists from
winding up with multiple records for the same peer with outdated
cookie values. This happens in some HA-NFS configurations where
rpc.statd is always running.
Signed-off-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
| |
Fix a debugging message to report correctly the count of hosts loaded
when statd starts up.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Clean up: The contents of NL_ADDR are fixed: they are always the IPv4
loopback address. Some time ago, the use of NL_ADDR() was stubbed out
of the NLM downcall forward path, replaced with a constant IPv4
loopback address.
Stub it out of the reply path as well, and then remove NL_ADDR
entirely.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
| |
|
|
|
|
|
| |
Replace deprecated gethostbyname(3) and gethostbyaddr(3) calls in
monitor.c, and address a couple of memory leaks.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
To support IPv6, statd must support multi-homed remote peers. For our
purposes, "multi-homed peer" means that more than one unique IP
address maps to the one canonical host name for that peer.
An SM_MON request from the local lockd has a "mon_name" argument that
statd reverse maps to a canonical hostname (ie the A record for that
host). statd assumes the canonical hostname is unique enough that
it stores the callback data for this mon_name in a file named after
that canonical hostname.
Because lockd can't distinguish between two unique IP addresses that
may be from the same physical host, the kernel can hand statd a
mon_name that maps to the same canonical hostname as some previous
mon_name. So that the kernel can keep this instance of the mon_name
unique, it creates a fresh priv cookie for each new address.
Note that a mon_name can be a presentation address string, or the
caller_name string sent in each NLMPROC_LOCK request. There's
nothing that requires the caller_name to be a fully-qualified
hostname, thus it's uniqueness is not guaranteed. The current
design of statd assumes that canonical hostnames will be unique
enough.
When a mon_name for a fresh SM_MON request maps to the same canonical
hostname as an existing monitored peer, but the priv cookie is new,
statd will try to write the information for the fresh request into an
existing monitor record file, wiping out the contents of the file.
This is because the mon_name/cookie combination won't match any record
statd already has.
Currently, statd doesn't check if a record file already exists before
writing into it. statd's logic assumes that the svc routine has
already checked that no matching record exists in the in-core monitor
list. And, it doesn't use O_EXCL when opening the record file. Not
only is the old data in that file wiped out, but statd's in-core
monitor list will no longer match what's in the on-disk monitor list.
Note that IPv6 isn't needed to exercise multi-homed peer support.
Any IPv4 peer that has multiple addresses that map to its canonical
hostname will trigger this behavior. However, this scenario will
become quite common when all hosts on a network automatically get both
an IPv4 address and an IPv6 address.
I can think of a few ways to address this:
1. Replace the current on-disk format with a database that has a
uniqueness constraint on the monitor records
2. Create a new file naming scheme; eg. one that uses a truly
unique name such as a hash generated from the mon_name, my_name, and
priv cookie
3. Support multiple lines in each monitor record file
Since statd's on-disk format constitutes a formal API, options 1 and 2
are right out. This patch implements option 3. There are two parts:
adding a new line to an existing file; and deleting a line from a file
with more than one line. Interestingly, the existing code already
supports reading more than one line from these files, so we don't need
to add extra code here to do that.
One file may contain a line for every unique mon_name / priv cookie
where the mon_name reverse maps to the same canonical hostname. We
use the atomic write facility added by a previous patch to ensure the
on-disk monitor record list is updated atomically.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
For the time being, statd is not going to support receiving SM_MON
calls from the local lockd via IPv6.
However, the upcalls (SM_MON, etc.) from the local lockd arrive on the
same socket that receives calls from remote peers. Thus
caller_is_localhost() at least has to be smart enough to notice that
the caller is not AF_INET, and to display non-AF_INET addresses
appropriately.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For the near future, statd will support IPv6 but exportfs will not.
Thus statd will need a version of matchhostname() that can deal
properly with IPv6 remotes. To reduce the risk of breaking exportfs,
introduce a separate version of matchhostname() for statd to use while
exportfs continues to use the existing AF_INET-only implementation.
Note that statd will never send matchhostname() a hostname string
containing export wildcards, so is_hostame() is not needed in the
statd version of matchhostname(). This saves some computational
expense when comparing hostnames.
A separate statd-specific implementation of matchhostname() allows
some flexibility in the long term, as well. We might want to enrich
the matching heuristics of our SM_NOTIFY, for example, or replace
them entirely with a heuristic that is not dependent upon DNS.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
|
| |
|
|
|
|
|
|
|
|
| |
Replace open-coded accesses to on-disk NSM information in rpc.statd
with calls to the new API.
Behavior should be much the same as it was before.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
| |
To facilitate code sharing between statd and sm-notify (and with other
components of nfs-utils), replace sm-notify's nsm_log() with xlog().
Since opt_quiet is used in only a handful of insignificant cases, it
is removed.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
| |
flag has been set. This cause warnings to be generated when
return values from reads/writes (and other calls) are not
checked. The patch address those warnings.
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
TI-RPC's version of the svc_getcaller() macro points to a sockaddr_in6,
not a sockaddr_in, though for AF_INET callers, an AF_INET address
resides there. To squelch compiler warnings when the TI-RPC version of
the svc_req structure is used, add inline helpers with appropriate
type casting.
Note that tcp_wrappers support only AF_INET addresses for now.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Statd is not unlinking host files during SM_UNMON and
SM_UNMON_ALL calls because the given host is still on the run-time
notify list (rtnl) and the check flag is set when xunlink() is
called. But the next thing the caller of xunlink() does is
remove the host from the rtnl list which means the
unlink will never happen.
So this patch removes the check flag from xunlink() since
its not needed and correctly allocates and frees memory
used by xunlink().
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Static code checkers flag this kind of thing because it's easy to
confuse with "if (!(foo == rtnl))". In one of these cases, the
combination of evaluation and assignment isn't even necessary.
While we are in the neighborhood, remove an extra argument to note() that is
not called for in the passed-in format string.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Clean up: Remove RESTRICTED_STATD to help make IPv6 changes simpler.
We keep the code behind RESTRICTED_STATD, and toss anything that is
compiled out when it is set.
RESTRICTED_STATD was added almost 10 years ago in response to CERT
CERT CA-99.05, which addresses exposures in rpc.statd that might allow
an attacker to take advantage of buffer overflows in rpc.statd while it
is running in privileged mode.
These days, I can't think of a reason why anyone would want to run
rpc.statd without setting RESTRICTED_STATD. In addition, I don't
think rpc.statd is ever tested without it.
Removing RESTRICTED_STATD will get rid of some address storage and
comparison issues that will make IPv6 support simpler. Plus it will
make our test matrix smaller!
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Neil Brown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Refactor common logic to check if SM_FOO request is from loopback
address.
We'll have to do something about this for IPv6. On IPv6-capable
systems, there will be only one AF_INET6 listener. The loopback caller
will get either an IPv6 loopback address, or a mapped IPv4 loopback --
either way this will be an AF_INET6 address.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: Neil Brown <neilb@suse.de>
Signed-off-by: Steve Dickson <steved@redhat.com>
|
| | |
|
| |
|
|
| |
Also free dns_name when freeing an 'nlist', so do the unlink before the free.
|
| |
|
|
|
| |
statd now passes the 'my_name' from the SM_MON call faithfully to the
ha-callout and records it in the sm/ files.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Both SM_STAT and SM_MON can return the state of an NSM, but it is
unclear which NSM they return the state of, so the value cannot be
used, and lockd doesn't use it.
Document this confusion, and give the current state to the kernel
via a sysctl if that sysctl is available (since about 2.6.19).
This should make is possible for the NFS server to detect a small
class of bad SM_NOTIFY packets and not flush locks in that case.
Signed-off-by: Neil Brown <neilb@suse.de>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When lockd asks to monitor a host, we find the FQDN from the DNS
and remember that, both internally and in the /var/lib/nfs/sm/*
file.
When we receive an SM_NOTIFY request, we compare both the
mon_name and the source IP address against that DNS name to find
a match.
If a DNS name is not available, we fall back to the name provided by
lockd, which at least is known to map to an IP address via
gethostbyname.
Signed-off-by: Neil Brown <neilb@suse.de>
|
| |
|
|
|
|
|
|
| |
From the point of view of the client (lockd), the 'priv' blob is probably
the most important key, so make sure to not throw away requests with
new 'priv' information.
Signed-off-by: Neil Brown <neilb@suse.de>
|
| |
|
|
|
|
| |
The if contains a while with essentially the same condition.
Signed-off-by: Neil Brown <neilb@suse.de>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
If statd dies and is restarted, it forgets what peers the kernel
is interested in monitoring, and so will not forward NOTIFY
requests properly.
With this patch the required information is recorded in the files
in /var/lib/nfs/sm/* so that a kill/restart does what you might
hope.
Signed-off-by: Neil Brown <neilb@suse.de>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
i.e. you now need --disable-secure-statd if you want any client
other than lockd to talk to statd.
Also relax the RESTRICTED_STATD checks so that a recent kernel
with /proc/sys/fs/nfs/nsm_use_hostnames set can still talk to
statd.
Finally, restrict access to simulate_crash so that only privileged
processes on localhost can call it. Having it accessible by the
whole world is probably not much more than a minor inconvenience,
but it really should be kept closed.
|
| |
|
|
| |
unused labels, constness, signedness.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
* utils/statd/monitor.c (sm_mon_1_svc): Fix buggy check for
program and procedure numbers of kernel lockd's callback. Also,
besides the old (and broken) procedure #24, allow #16 per Trond.
|
| |
|
|
| |
port 100021.
|
| |
|
