aboutsummaryrefslogtreecommitdiff
path: root/usr.sbin/cron/doc/FEATURES
diff options
context:
space:
mode:
Diffstat (limited to 'usr.sbin/cron/doc/FEATURES')
-rw-r--r--usr.sbin/cron/doc/FEATURES83
1 files changed, 83 insertions, 0 deletions
diff --git a/usr.sbin/cron/doc/FEATURES b/usr.sbin/cron/doc/FEATURES
new file mode 100644
index 000000000000..71b1486a8e01
--- /dev/null
+++ b/usr.sbin/cron/doc/FEATURES
@@ -0,0 +1,83 @@
+
+Features of Vixie's cron relative to BSD 4.[23] and SysV crons:
+
+-- Environment variables can be set in each crontab. SHELL, USER,
+ LOGNAME, and HOME are set from the user's passwd entry; all except
+ USER can be changed in the crontab. PATH is especially useful to
+ set there. TZ can be set, but cron ignores it other than passing
+ it on through to the commands it runs. Format is
+
+ variable=value
+
+ Blanks surrounding the '=' will be eaten; other blanks in value are
+ okay. Leading or trailing blanks can be preserved by quoting, single
+ or double quotes are okay, just so they match.
+
+ PATH=.:/bin:/usr/bin
+ SHELL=/bin/sh
+ FOOBAR = this is a long blanky example
+
+ Above, FOOBAR would get "this is a long blanky example" as its value.
+
+ SHELL and HOME will be used when it's time to run a command; if
+ you don't set them, HOME defaults to your /etc/passwd entry
+ and SHELL defaults to /bin/sh.
+
+ MAILTO, if set to the login name of a user on your system, will be the
+ person that cron mails the output of commands in that crontab. This is
+ useful if you decide on BINMAIL when configuring cron.h, since binmail
+ doesn't know anything about aliasing.
+
+-- Weekdays can be specified by name. Case is not significant, but only
+ the first three letters should be specified.
+
+-- Months can likewise be specified by name. Three letters only.
+
+-- Ranges and lists can be mixed. Standard crons won't allow '1,3-5'.
+
+-- Ranges can specify 'step' values. '10-16/2' is like '10,12,14,16'.
+
+-- Sunday is both day 0 and day 7 -- apparently BSD and ATT disagree
+ about this.
+
+-- Each user gets their own crontab file. This is a win over BSD 4.2,
+ where only root has one, and over BSD 4.3, where they made the crontab
+ format incompatible and although the commands can be run by non-root
+ uid's, root is still the only one who can edit the crontab file. This
+ feature mimics the SysV cron.
+
+-- The 'crontab' command is loosely compatible with SysV, but has more
+ options which just generally make more sense. Running crontab with
+ no arguments will print a cute little summary of the command syntax.
+
+-- Comments and blank lines are allowed in the crontab file. Comments
+ must be on a line by themselves; leading whitespace is ignored, and
+ a '#' introduces the comment.
+
+-- (big win) If the `crontab' command changes anything in any crontab,
+ the 'cron' daemon will reload all the tables before running the
+ next iteration. In some crons, you have to kill and restart the
+ daemon whenever you change a crontab. In other crons, the crontab
+ file is reread and reparsed every minute even if it didn't change.
+
+-- In order to support the automatic reload, the crontab files are not
+ readable or writable except by 'crontab' or 'cron'. This is not a
+ problem, since 'crontab' will let you do pretty much whatever you
+ want to your own crontab, or if you are root, to anybody's crontab.
+
+-- If any output is generated by a command (on stdout OR stderr), it will
+ be mailed to the owner of the crontab that contained the command (or
+ MAILTO, see discussion of environment variables, above). The headers
+ of the mail message will include the command that was run, and a
+ complete list of the environment that was passed to it, which will
+ contain (at least) the USER (LOGNAME on SysV), HOME, and SHELL.
+
+-- the dom/dow situation is odd. '* * 1,15 * Sun' will run on the
+ first and fifteenth AND every Sunday; '* * * * Sun' will run *only*
+ on Sundays; '* * 1,15 * *' will run *only* the 1st and 15th. this
+ is why we keep 'e->dow_star' and 'e->dom_star'. I didn't think up
+ this behaviour; it's how cron has always worked but the documentation
+ hasn't been very clear. I have been told that some AT&T crons do not
+ act this way and do the more reasonable thing, which is (IMHO) to "or"
+ the various field-matches together. In that sense this cron may not
+ be completely similar to some AT&T crons.