LOCKFILE_CREATE(3) Linux Programmer's Manual LOCKFILE_CREATE(3)

lockfile_create, lockfile_remove, lockfile_touch, lockfile_check - manage lockfiles

#include <lockfile.h>

cc [ flag ... ] file ... -llockfile [ library ]

int lockfile_create( const char *lockfile, int retrycnt, int flags [, struct lockargs args ] );
int lockfile_remove( const char *lockfile );
int lockfile_touch( const char *lockfile );
int lockfile_check( const char *lockfile, int flags );

Functions to handle lockfiles in an NFS safe way.

The lockfile_create function creates a lockfile in an NFS safe way.

If flags is set to L_PID or L_PPID then lockfile_create will not only check for an existing lockfile, but it will read the contents as well to see if it contains a process id in ASCII. If so, the lockfile is only valid if that process still exists. Otherwise, a lockfile older than 5 minutes is considered to be stale.

When creating a lockfile, if L_PID is set in flags, then the current process' PID will be written to the lockfile. Sometimes it can be useful to use the parent's PID instead (for example, the dotlockfile command uses that). In such cases you can use the L_PPID flag instead.

If the lockfile is on a shared filesystem, it might have been created by a process on a remote host. So the L_PID or L_PPID method of deciding if a lockfile is still valid or stale is incorrect and must not be used. If you are holding a lock longer than 5 minutes, a call to lockfile_create by another process will consider the lock stale and remove it. To prevent this, call lockfile_touch to refresh the lockfile at a regular interval (every minute or so).

This function checks if a valid lockfile is already present without trying to create a new lockfile.

Removes the lockfile.

lockfile_create returns one of the following status codes:

   #define L_SUCCESS   0    /* Lockfile created                      */
   #define L_TMPLOCK   2    /* Error creating tmp lockfile           */
   #define L_TMPWRITE  3    /* Can't write pid int tmp lockfile      */
   #define L_MAXTRYS   4    /* Failed after max. number of attempts  */
   #define L_ERROR     5    /* Unknown error; check errno            */
   #define L_ORPHANED  7    /* Called with L_PPID but parent is gone */
   #define L_RMSTALE   8    /* Failed to remove stale lockfile       */

lockfile_check returns 0 if a valid lockfile is present. If no lockfile or no valid lockfile is present, -1 is returned.

lockfile_touch and lockfile_remove return 0 on success. On failure -1 is returned and errno is set appropriately. It is not an error to lockfile_remove() a non-existing lockfile.

The algorithm that is used to create a lockfile in an atomic way, even over NFS, is as follows:

1
A unique file is created. In printf format, the name of the file is .lk%05d%x%s. The first argument (%05d) is the current process id. The second argument (%x) consists of the 4 minor bits of the value returned by time(2). The last argument is the system hostname.
2
Then the lockfile is created using link(2). The return value of link is ignored.
3
Now the lockfile is stat()ed. If the stat fails, we go to step 6.
4
The stat value of the lockfile is compared with that of the temporary file. If they are the same, we have the lock. The temporary file is deleted and a value of 0 (success) is returned to the caller.
5
A check is made to see if the existing lockfile is a valid one. If it isn't valid, the stale lockfile is deleted.
6
Before retrying, we sleep for n seconds. n is initially 5 seconds, but after every retry 5 extra seconds is added up to a maximum of 60 seconds (an incremental backoff). Then we go to step 2 up to retries times.

These functions do not lock a file - they generate a lockfile. However in a lot of cases, such as Unix mailboxes, all concerned programs accessing the mailboxes agree on the fact that the presence of <filename>.lock means that <filename> is locked.

If you are using lockfile_create to create a lock on a file that resides on a remote server, and you already have that file open, you need to flush the NFS attribute cache after locking. This is needed to prevent the following scenario:

open /var/mail/USERNAME
attributes, such as size, inode, etc are now cached in the kernel!
meanwhile, another remote system appends data to /var/mail/USERNAME
grab lock using lockfile_create()
seek to end of file
write data

Now the end of the file really isn't the end of the file - the kernel cached the attributes on open, and st_size is not the end of the file anymore. So after locking the file, you need to tell the kernel to flush the NFS file attribute cache.

The only portable way to do this is the POSIX fcntl() file locking primitives - locking a file using fcntl() has the fortunate side-effect of invalidating the NFS file attribute cache of the kernel.

lockfile_create() cannot do this for you for two reasons. One, it just creates a lockfile- it doesn't know which file you are actually trying to lock! Two, even if it could deduce the file you're locking from the filename, by just opening and closing it, it would invalidate any existing POSIX locks the program might already have on that file (yes, POSIX locking semantics are insane!).

So basically what you need to do is something like this:

  fd = open("/var/mail/USER");
  .. program code ..
  lockfile_create("/var/mail/USER.lock", x, y);
  /* Invalidate NFS attribute cache using POSIX locks */
  if (lockf(fd, F_TLOCK, 0) == 0) lockf(fd, F_ULOCK, 0);

You have to be careful with this if you're putting this in an existing program that might already be using fcntl(), flock() or lockf() locking- you might invalidate existing locks.

There is also a non-portable way. A lot of NFS operations return the updated attributes - and the Linux kernel actually uses these to update the attribute cache. One of these operations is chmod(2).

So stat()ing a file and then chmod()ing it to st.st_mode will not actually change the file, nor will it interfere with any locks on the file, but it will invalidate the attribute cache. The equivalent to use from a shell script would be

  chmod u=u /var/mail/USER

If you are on a system that has a mail spool directory that is only writable by a special group (usually "mail") you cannot create a lockfile directly in the mailspool directory without special permissions.

Lockfile_create and lockfile_remove check if the lockfile ends in $USERNAME.lock, and if the directory the lockfile is writable by group "mail". If so, an external set group-id mail executable (dotlockfile(1) ) is spawned to do the actual locking / unlocking.

/usr/lib/liblockfile.so.1

Miquel van Smoorenburg

dotlockfile(1), maillock(3), touchlock (3), mailunlock(3)

27 Januari 2021 Linux Manpage