diff options
| author | Laurent Bercot <ska-skaware@skarnet.org> | 2023-09-11 08:13:01 +0000 |
|---|---|---|
| committer | Laurent Bercot <ska@appnovation.com> | 2023-09-11 08:13:01 +0000 |
| commit | bc3863eaf3e4ae92eac8cd3ce0ca9dcb8915fc36 (patch) | |
| tree | 9c6d891677e1d48aa4219636bff1c103d9509a7d /doc/s6-setlock.html | |
| parent | ddc088fba6016ae839afaa208b0a441869dc936f (diff) | |
| download | s6-bc3863eaf3e4ae92eac8cd3ce0ca9dcb8915fc36.tar.xz | |
Better s6-setlock; delete the s6lock subsystem
Signed-off-by: Laurent Bercot <ska@appnovation.com>
Diffstat (limited to 'doc/s6-setlock.html')
| -rw-r--r-- | doc/s6-setlock.html | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/doc/s6-setlock.html b/doc/s6-setlock.html index c2e8fbf..91e2fcc 100644 --- a/doc/s6-setlock.html +++ b/doc/s6-setlock.html @@ -25,7 +25,7 @@ s6-setlock takes a lock on a file, then executes into another program. <h2> Interface </h2> <pre> - s6-setlock [ -n | -N | -t <em>timeout</em> ] [ -r | -w ] <em>file</em> <em>prog...</em> + s6-setlock [ -n | -N ] [ -t <em>timeout</em> ] [ -d <em>fd</em> ] [ -r | -w ] <em>file</em> <em>prog...</em> </pre> <ul> @@ -46,20 +46,20 @@ the lock after <em>timeout</em> milliseconds, it will exit 1. </li> <li> <tt>-r</tt> : shared lock. Other shared locks on the same file will not prevent the lock from being acquired (but an exclusive lock will). </li> <li> <tt>-w</tt> : exclusive lock. This is the default. </li> + <li> <tt>-d <em>fd</em></tt> : make the lock visible in <em>prog</em> +on file descriptor <em>fd</em>. </li> </ul> <h2> Notes </h2> <ul> - <li> s6-setlock leaks an open file descriptor into the <em>prog</em> + <li> +s6-setlock leaks an open file descriptor into the <em>prog</em> execution. This is intended: the fd holds the lock, which is released when <em>prog</em> exits. <em>prog</em> must not touch fds it does not -know about. </li> - <li> If the timed lock option is chosen, s6-setlock does not acquire the lock -itself. Instead, it spawns a <a href="libs6/s6lockd-helper.html">s6lockd-helper</a> -process that acquires the lock while s6-setlock controls the timeout; the -s6lockd-helper process then holds the lock and lives as long as -<em>prog</em>. </li> +know about; by default it has no need to know the descriptor that holds +the lock. But if you need to officialize the presence of the lock in +<em>prog</em>, that's where the <tt>-d</tt> option comes in. </li> </ul> </body> |