xref: /freebsd/usr.sbin/cron/doc/MAIL (revision 97759ccc715c4b365432c16d763c50eecfcb1100)
1[ this is really old mail that came to me in response to my 1986 posting
2  to usenet asking for feature suggestions before releasing the first
3  version of cron.  it is presented here for its entertainment value.
4  --vix ]
5
6From ptsfa!lll-crg!ames!acornrc!bob Wed Dec 31 10:07:08 1986
7Date: Wed, 31 Dec 86 08:59:31 pst
8From: lll-crg!ames!acornrc!bob (Bob Weissman)
9To: ptsfa!vixie!paul
10Status: RO
11
12Sure, here's a suggestion: I'd like to be able to run a program, say,
13every two hours.  Current cron requires me to write
140,2,4,6,8,10,12,14,16,18,20,22 in the hours field.  How about a notation
15to handle this more elegantly?
16
17<<	Okay, I've allowed 0-22/2 as a means of handling this.
18	The time specification for my cron is as follows:
19		specification = range {"," range}
20		range = (start "-" finish ["/" step]) | single-unit
21	This allows "1,3,5-7", which the current cron doesn't (it won't
22	do a range inside a list), and handles your specific need.	>>
23
24From drw@mit-eddie Wed Dec 31 18:25:27 1986
25Date: Wed, 31 Dec 86 14:28:19 est
26From: drw@mit-eddie (Dale Worley)
27To: mit-eddie!vixie!paul
28Status: RO
29
30We have a lot of lines in our crontab of the form
31
32	00 12 * * * su user < /usr/users/user/script.file
33
34This barfs (silently!) on our system (Dec Ultrix 1.2 == 4.2bsd) if
35user's shell is csh.  This, I am told, is because csh requires that
36the environment be set up in certain ways, which cron doesn't do.
37(Actually, I believe, it is because /etc/rc, which runs cron, doesn't
38set up the environment enough for csh to run, and cron just inherits
39the situation.)  Anyway, the point is that if you find out what csh
40really needs in its environment, you might want to set up cron to
41provide some reasonable defaults (if it isn't supplied by cron's
42parent).  Also, could you tell me what csh needs, if you find out, so
43we can hack our /etc/rc?
44
45<<	well, the environment IS a problem. processes that cron forks
46	will inherit the environment of the person who ran the cron
47	daemon... I plan to edit out such useless things as TERMCAP,
48	TERM, and the like; supply correct values for HOME, USER, CWD,
49	and whatever else comes to mind.  I'll make sure csh works...	>>
50From ptsfa!ames!seismo!dgis!generous Thu Jan  1 07:33:17 1987
51Date: Thu Jan 1 10:29:20 1987
52From: ames!seismo!dgis!generous (Curtis Generous)
53To: nike!ptsfa!vixie!paul
54Status: RO
55
56Paul:
57
58One of the limitations of the present versions of cron is the lack
59of the capability of specifying a way to execute a command every
60n units of time.
61
62Here is a good example:
63
64# Present method to start up uucico
6502,12,22,32,42,52 * * * * 	exec /usr/lib/uucp/uucico -r1
66
67# New method ?? (the ':' here is just one possibility for syntax)
6802:10 * * * *			exec /usr/lib/uucp/uucico -r1
69
70This method would prove very helpful for those programs that get started
71every few minutes, making the entry long and not easily readable.  The first
72number would specify the base time, and the second number the repetition
73interval.
74
75<<	Good idea, but bob@acornrc beat you to it.  I used '/' instead of
76	':'.  This is my personal preference, and seems intuitive when you
77	think of the divide operator in C... Does anyone have a preference? >>
78
79From ptsfa!lll-lcc!seismo!decuac!c3pe!c3engr!charles Thu Jan  1 17:04:24 1987
80From: lll-lcc!seismo!c3pe!c3engr!charles (Charles Green)
81To: c3pe!decuac!dolqci!vrdxhq!seismo!lll-lcc!ptsfa!vixie!paul
82Date: Thu Jan  1 19:22:47 1987
83Status: RO
84
85Well, this isn't a compatible extension, but I have in times past wondered
86about a facility to let you start a process at intervals of, say, 17 minutes,
87instead of particular minutes out of each hour.
88
89<<	This was a popular request!	>>
90
91From seismo!uwvax!astroatc!nicmad!norvax!mann Sun Jan  4 13:04:01 1987
92Date: Fri, 2 Jan 87 09:23:53 cst
93From: lll-lcc!seismo!uwvax!astroatc!nicmad!norvax!mann (Tom Mann)
94To: ptsfa!vixie!paul
95Status: RO
96
97I'm not sure if it is in cron (either SysV or BSD ... if it is, I haven't
98figured it out ) but a comment feature would SURE BE NICE!.
99There are times when I want to comment out an entry
100for a period of time; it might also make it a lot more legible.
101
102<<	My cron allows blank lines and standard #-type comments.  I know
103	that one BSD4.2 cron I've used had it.  I don't know about SysV.  >>
104
105From ptsfa!hoptoad!hugh Mon Jan  5 10:26:46 1987
106Date: Mon, 5 Jan 87 01:22:17 PST
107From: hoptoad!hugh (Hugh Daniel)
108To: ptsfa!vixie!paul
109Status: RO
110
111  Hi, I do have a BIG one that I would like.  I want to log ALL output
112from command lines into a file for each line.  Thus I might have a chance
113of finding out why my crontab entry did not work.
114  This would seem to work best if done by cron, as it is now I have a google
115of shell scripts laying about just to put the error output where I can see
116it.
117
118<<	My cron (and the SysV cron) will send mail to the owner of the
119	particular crontab file if a command generates any output on stdout
120	or stderr.  This can be irritating, but if you write a script such
121	that any output means a problem occurred, you can avoid most logfile
122	needs, and not generate mail except in unforeseen circumstances.   >>
123
124From ptsfa!dual!ucbvax!ihnp4!anvil!es!Robert_Toxen Mon Jan  5 13:08:46 1987
125From: dual!ucbvax!ihnp4!anvil!es!Robert_Toxen
126Date: Fri,  2 Jan 87 14:25:29 EST
127To: anvil!ihnp4!ucbvax!dual!ptsfa!vixie!paul
128Status: RO
129
130Here are some suggestions:
1311. Run it through the C preprocessor via "/lib/<whatever>".
132
133<<	hmmm. this seems of limited utility, and if you really wanted
134	to do it that way, you could do it yourself (since users can
135	write to their own crontab files).  I'll add '-' (read stdin)
136	to the crontab installer program to facilitate this.		>>
137
1382. Allow specifying every Nth day of week, i.e., every second Wednesday.
139   I did this to calendar by separating the day of week (Wed=4, which one
140   to start on and N with slashes).  I took modulo the day of year as a
141   starting point so that someone with a desk calendar documenting such
142   things can easily determine the offset (second number).  I did this
143   while at SGI; alas I don't have a copy of the code.
144
145<<	I can see how this could be useful, but I'm not sure how I'd
146	implement it.  Cron currently doesn't keep track of the last time
147	a given command was run; whether the current Wednesday is the first
148	or second since the command was last run would be pretty hard to
149	figure out.  I'd have to keep a database of commands and their
150	execution around, and purge it when the crontab was overwritten.
151	This is too much work for me, but if someone adds it, let me know.  >>
152
153From ptsfa!ames!seismo!cbmvax!devon!paul Tue Jan  6 05:50:17 1987
154From: ames!seismo!cbmvax!devon!paul
155To: cbmvax!seismo!nike!ptsfa!vixie!paul
156Date: Mon Jan  5 09:29:57 1987
157Status: RO
158
159One problem that has always plagued me with cron is the assumed ORing.
160I'd like to see some type of ANDing implemented.  I guess I can best
161describe this by example.  Say I have the following line in my crontab
162file:
163
164*  *  4-31  *  1-6	/usr/bin/command
165
166What this does is run 'command' on the 4th thru 31st days of the
167month, AND on Monday thru Saturday; which probably means running it
168every day of the month (unless Sunday falls on days 1-3).  This
169happens because cron runs the command if the day-of-month OR the
170day-of-week is true.
171
172What I'd like to happen with the above line is to run the command ONLY
173on Monday thru Saturday any time after the 3rd of the month, e.g. if
174the day-of-month AND the day-of-week are true.
175
176My proposal to you is to implement some special chars for the first
177five fields.  Examples:
178
179*  *  !1-3  *  1-6	/usr/bin/command
180
181(run command Mon-Sat, but NOT [!] on the first 3 days of the month)
182
183*  *  &4-31 *  &1-6	/usr/bin/command
184
185(run command if day-of-month AND day-of-week are true)
186
187Get the picture?  This would be compatible with existing versions of
188cron (which wouldn't currently be using any special characters, so
189that old crontabs would be handled correctly).
190
191<<	This message made me aware of the actual boolean expression involved
192	in a crontab entry.  I'd assumed that it was
193		(minute && hour && DoM && month && DoW)
194	But it's really
195		(minute && hour && month && (DoM || DoW))
196
197	I can see some value in changing this, but with a fixed order of
198	fields, operators get to be kindof unary, which && and || really
199	aren't.  If someone has an idea on a syntax that allows useful
200	variations to the standard (&& && && (||)) default, please suggest. >>
201
202From bobkat!pedz Tue Jan  6 20:02:10 1987
203From: pedz@bobkat.UUCP (Pedz Thing)
204Date: 2 Jan 87 17:34:44 GMT
205Status: RO
206
207Log files!  It would be nice to be able to specify a log for cron
208itself and also a log for each program's stdout and stderr to go to.
209The latter can of course be done with > and 2> but it would be nice if
210there could be a single line with some sort of pattern like
211`> /usr/spool/log/%' and the command would be substituted for the %.
212Another thing which would be nice is to be able to specify which shell
213to call to give the command to.
214
215<<	Log files are done with mail.  The '%' idea could be useful if
216	a different character were used (% is special to cron, see man
217	page); a different directory would have to be chosen, since each
218	user has their own crontab file; and something intelligent would
219	have to be done in the file naming, since the first word of the
220	command might be ambiguous (with other commands).  In short, it's
221	too much work.  Sorry.						  >>
222
223From guy%gorodish@sun Tue Jan  6 20:03:13 1987
224From: guy%gorodish@sun (Guy Harris)
225Message-ID: <10944@sun.uucp>
226Date: 5 Jan 87 12:09:09 GMT
227References: <429@vixie.UUCP> <359@bobkat.UUCP>
228Sender: news@sun.uucp
229Status: RO
230
231> Another thing which would be nice is to be able to specify which shell
232> to call to give the command to.
233
234Well, the obvious choice would be the user's shell, but this wouldn't work
235for accounts like "uucico".
236
237<<	I use the owning user's shell, and to handle "uucico" I check a
238	list of "acceptable shells" (currently compiled in, does anybody
239	mind?), substituting a default (compiled in) shell if the user's
240	shell isn't on the list.
241
242	BTW, "compiled in" means that it's in a .h file, easily changed
243	during installation, but requiring recompilation to modify.  You
244	don't have to go digging through the code to find it...		  >>
245
246From qantel!hplabs!ucbvax!mwm@violet.berkeley.edu Tue Jan  6 21:24:48 1987
247To: hplabs!qantel!vixie!paul (Paul Vixie Esq)
248Date: 04 Jan 87 00:42:35 PST (Sun)
249From: Mike Meyer <mwm@violet.berkeley.edu>
250Status: RO
251
252<<[Discussion of RMS/FSF, and mwm's GNU Cron deleted]>>
253
254Oh, yeah - here are the extensions on my cron:
255
2561) Sunday is both day 0 and day 7, so it complies with both SysV and
257BSD cron.
258
259<<	Good idea. I did it too, thanks for informing me.	>>
260
2612) At is integrated into the cron. Instead of atrun to scan the
262/usr/spool/at directory, at files are put into the /usr/lib/cron
263directory along with users cron files, and cron fabricates a line from
264a crontab file to run them. This is considered a major win by all who
265use it.
266
267<<	I don't use 'at', and my cron doesn't do anything with it.  To run
268	'at', I use 'atrun' the same way the current BSD cron does.  My
269	crontab files are in /usr/spool/cron/crontabs, in the SysV
270	tradition -- not in /usr/lib/cron.  This is a configuration
271	parameter, of course.						>>
272
273There are two known restrictions:
274
2751) I don't support any of the SysV security hooks. I don't have a use
276for them, and RMS didn't like the idea at all :-).
277
278<<	This means cron.allow and cron.deny.  I plan to support them, as
279	they've been quite helpful at various HPUX sites I've administered. >>
280
2812) Cron expects to be able to create files with names longer than 14
282characters, which makes it hard to run on SysV. At least one person
283was working on a port, but I don't know how it's going. That might
284make for a good reason for releasing yours, right there.
285
286<<	If someone has SysV (with the 14-character limit), they probably
287	won't want my cron, since it doesn't add much to the standard
288	version (which they may have support for).  My cron is not currently
289	portable to non-BSD systems, since it relies on interval timers (I
290	needed to sleep for intervals more granular than seconds alone would
291	allow).  The port would be trivial, and I will do it if a lot of
292	people ask for it...						>>
293
294Oh, yeah - I'm going to see about getting this cron integrated into
295the next 4BSD release.
296
297<<	How does one go about this?  I have a few nifty gadgets I'd like
298	to contribute, this cron being one of them...			>>
299
300<<[more FSF/GNU discussion deleted]>>
301
302From qantel!hplabs!ames!ut-sally!ut-ngp!melpad!bigtex!james Tue Jan  6 21:24:57 1987
303Posted-Date: Fri, 2 Jan 87 19:26:16 est
304Date: Fri, 2 Jan 87 19:26:16 est
305From: hplabs!ames!ut-sally!ut-ngp!bigtex!james
306To: vixie!paul
307Status: RO
308
309Yes!!!  There are several critical failures in System V cron...
310
3111. Pass all variables in cron's environment into the environment of things
312   cron starts up, or at least into the crontab entries started up (at jobs
313   will inherit the environment of the user).  If nothing else it is critically
314   important that the TZ variable be passed on.  PATH should be passed on too.
315   Basically, passing environment values allows one to design a standard
316   environment with TZ and PATH and have that run by everything.  If anyone
317   tells you this is no big deal, consider what happens when uucico is
318   started by cron in CA to make a long distance phone link...  Unless the
319   administrator is really on his/her toes, calls scheduled at 5pm will really
320   go at two in the afternoon, needlessly incurring huge phone bills, all
321   because System V refuses to pass the TZ from its environment down.  There
322   are work arounds, but only putting it in cron will really work.  This is
323   not a security hole.
324
325<<	delete TERM and TERMCAP; modify HOME, USER, and CWD; pass TZ and
326	PATH through undisturbed.  any other requests out there?
327
328	BSD doesn't have this problem -- TZ is passed right on through if
329	you define it in the shell before starting my cron daemon.  However,
330	the BSD I'm running this on doesn't need TZ to be defined anyway...
331	The default in the kernel has been just fine so far...  But just the
332	same, if/when I port to SysV (I guess I really should), I'll make
333	sure this works right.
334
335	I guess I've been spoiled.  HPUX is SysV-based, and I never had a
336	problem with cron and TZ when I used it.			  >>
337
3382. A way to avoid logging stuff in /usr/lib/cron/log.  I have a cron entry
339   run uudemon.hr every 10 minutes.  This is 144 times/day.  Each run generates
340   three lines of text, for a total of 432 lines of text I don't want to see.
341   Obviously this should be optional, but it would be nice if there were a
342   way to flag an entry so that it wasn't logged at all unless there was an
343   error.
344
345<<	I don't know nothin' 'bout no /usr/lib/cron/log.  What is this file?
346	I don't see any reason to create log entries, given the mail-the-
347	output behaviour.  Opinions, anyone?				>>
348
349I will come up with other ideas no doubt, but I can always implement them
350myself.
351
352<<	That's what I like about PD software.  Please send me the diffs!  >>
353
354The other problem you have is making sure you can run standard
355crontabs.  I would suggest something like this: if the command part of the
356entry starts with an unescaped -, then flags and options follow immediately
357thereafter.  As in:
358
3592,12,22,32,42,52 * * * * -q /usr/lib/uucp/uudemon.hr
360
361This could mean do not log the uudemon.hr run unless there is a problem of
362some kind.  This is probably safe as not many filenames start with "-", and
363those that do are already a problem for people.
364
365<<	Since I don't plan on supporting /usr/lib/cron/log in ANY form unless
366	many people request it, I won't be needing -q as you've defined it.
367	I could use something like this to avoid sending mail on errors, for
368	the occasional script that you don't want to bullet-proof.
369
370	The compatibility issue is CRITICAL.  The 4.3BSD crontab format is
371	a crime against the whole philosophy of Unix(TM), in my opinion.   >>
372
373One other minor thing to consider is the ulimit: can different users get
374different ulimits for their crontab entries?
375
376<<	Boy I'm ignorant today.  What's a ulimit, and what should I do with
377	it in a crontab?  Suggestions, enlightenment, etc ??		   >>
378
379From qantel!lll-crg!ames!uw-beaver!uw-nsr!john Tue Jan  6 23:32:44 1987
380Date: Thu, 1 Jan 87 10:53:05 pst
381From: lll-crg!ames!uw-beaver!uw-nsr!john (John Sambrook 5-7433)
382To: vixie!paul
383Status: RO
384
385How about not hardwiring the default environment that cron builds for its
386children in the cron program itself?  Our cron does this and it's the pits
387because we are TZ=PST8PDT not TZ=EST5EDT !
388
389<<	yeachk.  I assure you, I will not hardwire the TZ!		>>
390From ptsfa!well!dv Fri Jan  9 04:01:50 1987
391Date: Thu, 8 Jan 87 23:50:40 pst
392From: well!dv (David W. Vezie)
393To: ptsfa!vixie!paul
394Status: RO
395
3966, have a special notation called 'H' which would expand to weekends
397   and holidays (you'd have to keep a database somewhere of real
398   holidays), and also 'W' for workdays (neither weekend or holiday).
399
400<<	Too much work.  There should be a standard way to define and
401	detect holidays under Unix(TM); if there were, I'd use it.  As
402	it is, I'll leave this for someone else to add.
403
404	I can see the usefulness; it just doesn't quite seem worth it.    >>
405From qantel!gatech!akgua!blnt1!jat Wed Jan 14 20:00:40 1987
406Date: Tue, 13 Jan 87 16:39:38 EST
407From: gatech!akgua!blnt1!jat
408Status: RO
409
4101) Add some way to tell cron to reread the files, say kill -1 <pid>
411
412<<	whenever the 'crontab' program is run and updates a crontab file,
413	a file /usr/spool/cron/POKECRON is created; next time the cron
414	daemon wakes up, it sees it, and re-reads the crontab files.
415
416	I thought of handling the signal; even implemented it.  Then this
417	clever idea hit me and I ripped it all out and added a single
418	IF-statement to handle the POKECRON file.			>>
419
4202) Have some kind of retry time so that if a command fails, cron will try to
421   execute it again after a certain period.  This is useful if you have some
422   type of cleanup program that can run at the scheduled time for some reason
423   (such as locked device, unmounted filesystem, etc).
424
425<<	Hmmm, sounds useful.  I could do this by submitting an 'at' job...
426	I'll think about it.						>>
427From ptsfa!dual!ucbvax!ihnp4!mtuxo!ender Sat Jan  3 16:54:00 1987
428From: dual!ucbvax!ihnp4!mtuxo!ender
429Date: Sat, 3 Jan 87 14:05:13 PST
430To: ucbvax!dual!ptsfa!vixie!paul
431Status: RO
432
433It would be nice if nonprivileged users can setup personal crontab files
434(~/.cronrc, say) and be able to run personal jobs at regular intervals.
435
436<<	this is done, but in the SysV style: the 'crontab' program installs
437	a new crontab file for the executing user (can be overridden by root
438	for setup of uucp and news).  the advantage of this is that (1) when
439	a crontab is changed, the daemon can be informed automatically; and
440	(2) the file can be syntax-checked before installation.		>>
441From ptsfa!ames!seismo!ihnp4!lcc!richard Fri Jan 16 04:47:33 1987
442Date: Fri, 16 Jan 87 07:44:57 EST
443To: nike!ptsfa!vixie!paul
444Status: RO
445
446The System V cron is nice, but it has a few annoying features.  One is that
447its mail files will say that the previous message is the output of "one of your
448cron commands."  I wish it would say WHICH cron command.
449
450<<	Done.  Also which shell, which user (useful when the mail gets
451	forwarded), which home directory, and other useful crud.	>>
452
453Another problem is with timezones.  It is necessary to specify TZ=PST8PDT (or
454whatever) when you invoke cron (from inittab, or /etc/rc) and it is also
455necessary to add TZ=PST8PDT to each crontab line which might need it.  Cron
456should automatically export its idea of the "TZ" to each invoked command, and
457it should be possible to put a line in the crontab file which overrides that
458for every command in the file (e.g., most users are on EST, so cron is run
459with TZ=EST5EDT; but one user is usually on PST and wants all of his cron
460commands to run with TZ=PST8PDT).  This might be extended to allow any
461environment variable to be specified once for the whole crontab file (e.g.,
462PATH).
463
464<<	Well, since I run the user's shell, you could put this into .cshrc.
465	generic environment-variable setting could be useful, though.  Since
466	I have to modify the environment anyway, I'll consider this.	  >>
467
468A log file might be a nice idea, but the System V cron log is too verbose.
469I seem to remember that cron keeps it open, too; so you can't even have
470something go and periodically clean it out.
471
472<<	I don't do /usr/lib/cron/log.  I wasn't aware of this file until I
473	got all these suggestions.  Do people want this file?  Tell me!    >>
474