statd (8) - Linux Man Pages
statd: NSM service daemon
NAMErpc.statd - NSM service daemon
SYNOPSISrpc.statd [-dh?FLNvV] [-H prog] [-n my-name] [-o outgoing-port]
DESCRIPTIONFile locks are not part of persistent file system state. Lock state is thus lost when a host reboots.
Network file systems must also detect when lock state is lost because a remote host has rebooted. After an NFS client reboots, an NFS server must release all file locks held by applications that were running on that client. After a server reboots, a client must remind the server of file locks held by applications running on that client.
For NFS version 2 [RFC1094] and NFS version 3 [RFC1813], the Network Status Monitor protocol (or NSM for short) is used to notify NFS peers of reboots. On Linux, two separate user-space components constitute the NSM service:
- A daemon that listens for reboot notifications from other hosts, and manages the list of hosts to be notified when the local system reboots
- A helper program that notifies NFS peers after the local system reboots
The local NFS lock manager alerts its local rpc.statd of each remote peer that should be monitored. When the local system reboots, the sm-notify command notifies the NSM service on monitored peers of the reboot. When a remote reboots, that peer notifies the local rpc.statd, which in turn passes the reboot notification back to the local NFS lock manager.
NSM OPERATION IN DETAILThe first file locking interaction between an NFS client and server causes the NFS lock managers on both peers to contact their local NSM service to store information about the opposite peer. On Linux, the local lock manager contacts rpc.statd.
rpc.statd records information about each monitored NFS peer on persistent storage. This information describes how to contact a remote peer in case the local system reboots, how to recognize which monitored peer is reporting a reboot, and how to notify the local lock manager when a monitored peer indicates it has rebooted.
An NFS client sends a hostname, known as the client's caller_name, in each file lock request. An NFS server can use this hostname to send asynchronous GRANT calls to a client, or to notify the client it has rebooted.
The Linux NFS server can provide the client's caller_name or the client's network address to rpc.statd. For the purposes of the NSM protocol, this name or address is known as the monitored peer's mon_name. In addition, the local lock manager tells rpc.statd what it thinks its own hostname is. For the purposes of the NSM protocol, this hostname is known as my_name.
There is no equivalent interaction between an NFS server and a client to inform the client of the server's caller_name. Therefore NFS clients do not actually know what mon_name an NFS server might use in an SM_NOTIFY request. The Linux NFS client uses the server hostname from the mount command to identify rebooting NFS servers.
Reboot notificationWhen the local system reboots, the sm-notify command reads the list of monitored peers from persistent storage and sends an SM_NOTIFY request to the NSM service on each listed remote peer. It uses the mon_name string as the destination. To identify which host has rebooted, the sm-notify command sends the my_name string recorded when that remote was monitored. The remote rpc.statd matches incoming SM_NOTIFY requests using this string, or the caller's network address, to one or more peers on its own monitor list.
If rpc.statd does not find a peer on its monitor list that matches an incoming SM_NOTIFY request, the notification is not forwarded to the local lock manager. In addition, each peer has its own NSM state number, a 32-bit integer that is bumped after each reboot by the sm-notify command. rpc.statd uses this number to distinguish between actual reboots and replayed notifications.
- -d, --no-syslog
- Causes rpc.statd to write log messages on stderr instead of to the system log, if the -F option was also specified.
- -F, --foreground
- Keeps rpc.statd attached to its controlling terminal so that NSM operation can be monitored directly or run under a debugger. If this option is not specified, rpc.statd backgrounds itself soon after it starts.
- -h, -?, --help
- Causes rpc.statd to display usage information on stderr and then exit.
- -H, --ha-callout prog
- Specifies a high availability callout program. If this option is not specified, no callouts are performed. See the High-availability callouts section below for details.
- -L, --no-notify
- Prevents rpc.statd from running the sm-notify command when it starts up, preserving the existing NSM state number and monitor list.
- Note: the sm-notify command contains a check to ensure it runs only once after each system reboot. This prevents spurious reboot notification if rpc.statd restarts without the -L option.
- -n, --name ipaddr | hostname
- Specifies the bind address used for RPC listener sockets. The ipaddr form can be expressed as either an IPv4 or an IPv6 presentation address. If this option is not specified, rpc.statd uses a wildcard address as the transport bind address.
- This string is also passed to the sm-notify command to be used as the source address from which to send reboot notification requests. See sm-notify(8) for details.
- Causes rpc.statd to run the sm-notify command, and then exit. Since the sm-notify command can also be run directly, this option is deprecated.
- -o, --outgoing-port port
- Specifies the source port number the sm-notify command should use when sending reboot notifications. See sm-notify(8) for details.
- -p, --port port
- Specifies the port number used for RPC listener sockets. If this option is not specified, rpc.statd will try to consult /etc/services, if gets port succeed, set the same port for all listener socket, otherwise chooses a random ephemeral port for each listener socket.
- This option can be used to fix the port value of its listeners when SM_NOTIFY requests must traverse a firewall between clients and servers.
- -T, --nlm-port port
- Specifies the port number that lockd should listen on for NLM requests. This sets both the TCP and UDP ports unless the UDP port is set separately.
- -U, --nlm-udp-port port
- Specifies the UDP port number that lockd should listen on for NLM requests.
- -P, --state-directory-path pathname
- Specifies the pathname of the parent directory where NSM state information resides. If this option is not specified, rpc.statd uses /var/lib/nfs/statd by default.
- After starting, rpc.statd attempts to set its effective UID and GID to the owner and group of this directory.
- -v, -V, --version
- Causes rpc.statd to display version information on stderr and then exit.
CONFIGURATION FILEMany of the options that can be set on the command line can also be controlled through values set in the [statd] or, in some cases, the [lockd] sections of the /etc/nfs.conf configuration file. Values recognized in the [statd] section include port, outgoing-port, name, state-directory-path, and ha-callout which each have the same effect as the option with the same name.
The values recognized in the [lockd] section include port and udp-port which have the same effect as the --nlm-port and --nlm-udp-port options, respectively.
SECURITYThe rpc.statd daemon must be started as root to acquire privileges needed to create sockets with privileged source ports, and to access the state information database. Because rpc.statd maintains a long-running network service, however, it drops root privileges as soon as it starts up to reduce the risk of a privilege escalation attack.
During normal operation, the effective user ID it chooses is the owner of the state directory. This allows it to continue to access files in that directory after it has dropped its root privileges. To control which user ID rpc.statd chooses, simply use chown(1) to set the owner of the state directory.
You can also protect your rpc.statd listeners using the tcp_wrapper library or iptables(8). To use the tcp_wrapper library, add the hostnames of peers that should be allowed access to /etc/hosts.allow. Use the daemon name statd even if the rpc.statd binary has a different filename.
ADDITIONAL NOTESLock recovery after a reboot is critical to maintaining data integrity and preventing unnecessary application hangs. To help rpc.statd match SM_NOTIFY requests to NLM requests, a number of best practices should be observed, including:
- The UTS nodename of your systems should match the DNS names that NFS peers use to contact them
- The UTS nodenames of your systems should always be fully qualified domain names
- The forward and reverse DNS mapping of the UTS nodenames should be consistent
- The hostname the client uses to mount the server should match the server's mon_name in SM_NOTIFY requests it sends
Unmounting an NFS file system does not necessarily stop either the NFS client or server from monitoring each other. Both may continue monitoring each other for a time in case subsequent NFS traffic between the two results in fresh mounts and additional file locking.
On Linux, if the lockd kernel module is unloaded during normal operation, all remote NFS peers are unmonitored. This can happen on an NFS client, for example, if an automounter removes all NFS mount points due to inactivity.
High-availability calloutsrpc.statd can exec a special callout program during processing of successful SM_MON, SM_UNMON, and SM_UNMON_ALL requests, or when it receives SM_NOTIFY. Such a program may be used in High Availability NFS (HA-NFS) environments to track lock state that may need to be migrated after a system reboot.
The name of the callout program is specified with the -H option. The program is run with 3 arguments: The first is either add-client del-client or sm-notify depending on the reason for the callout. The second is the mon_name of the monitored peer. The third is the caller_name of the requesting lock manager for add-client or del-client , otherwise it is IP_address of the caller sending SM_NOTIFY. The forth is the state_value in the SM_NOTIFY request.
IPv6 and TI-RPC supportTI-RPC is a pre-requisite for supporting NFS on IPv6. If TI-RPC support is built into rpc.statd, it attempts to start listeners on network transports marked 'visible' in /etc/netconfig. As long as at least one network transport listener starts successfully, rpc.statd will operate.
- If set to a positive integer, has the same effect as --no-notify.
- directory containing monitor list
- directory containing notify list
- NSM state number for this host
- pid file
- network transport capability database
AUTHORSJeff Uphoff <juphoff [at] users.sourceforge.net>
Olaf Kirch <okir [at] monad.swb.de>
H.J. Lu <hjl [at] gnu.org>
Lon Hohberger <hohberger [at] missioncriticallinux.com>
Paul Clements <paul.clements [at] steeleye.com>
Chuck Lever <chuck.lever [at] oracle.com>