/usr/share/doc/coop-computing-tools/manual/ftsh.html is in coop-computing-tools-doc 4.0-1.1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 | <html>
<head>
<title>Fault-Tolerant Shell (ftsh) Technical Manual</title>
</head>
<body bgcolor=white>
<h1>Fault-Tolerant Shell (ftsh) Technical Manual</h1>
<b>27 August 2004</b>
<p>
The Fault Tolerant Shell (ftsh) is Copyright (c) 2003-2004 Douglas Thain and
Copyright (c) 2005 The University of Notre Dame. All rights reserved.
This software is distributed under the GNU General Public License.
Please see the file COPYING for details.
<p>
<b>Please use the following citation for FTSH:</b>
<dir>
<li> Douglas Thain and Miron Livny, <a href=http://www.cse.nd.edu/~dthain/papers/ethernet-hpdc12.pdf>The Ethernet Approach to Grid Computing</a>, IEEE High Performance Distributed Computing, August 2003.
</dir>
<a name=introduction></a>
<h2>Introduction</h2>
Shell scripts are a vital tool for integrating software.
They are indispensable for rapid prototyping and system assembly.
Yet, they are extraordinarily sensitive to errors.
A missing file, a broken network, or a sick file server can
cause a script to barrel through its actions with surprising results.
It is possible to write error-safe scripts,
but only with extraordinary discipline and complexity.
<p>
The Fault Tolerant Shell (ftsh) aims to solve this problem
by combining the ease of scripting with precise error semantics.
Ftsh is a balance between the flexibility and power of script languages
and the precision of most compiled languages. Ftsh parses complete
programs in order to eliminate run-time errors.
An exception-like structure allows scripts to be both succinct and safe.
A focus on timed repetition simplifies the most common form of
recovery in a distributed system. A carefully-vetted set of language
features limits the "surprises" that haunt system programmers.
<p>
As an example, consider this innocuous script written in Bourne Shell:
<pre>
#!/bin/sh
cd /work/foo
rm -rf bar
cp -r /fresh/data .
</pre>
Suppose that the <tt>/work</tt> filesystem is temporarily unavailable,
perhaps due to an NFS failure.
The <tt>cd</tt> command will fail and print a message on the console.
The shell will ignore this error result -- it is primarily designed
as a user interface tool -- and proceed to execute the <tt>rm</tt> and
<tt>cp</tt> in the directory it happened to be before.
<p>
Naturally, we may attempt to head off these cases with code that
checks error codes, attempts to recover, and so on. However,
even the disciplined programmer that leaves no value unturned must
admit that this makes shell scripts incomprehensible:
<pre>
#!/bin/sh
for attempt in 1 2 3
cd /work/foo
if [ ! $? ]
then
echo "cd failed, trying again..."
sleep 5
else
break
fi
done
if [ ! $? ]
then
echo "couldn't cd, giving up..."
return 1
fi
</pre>
And that's just the first line!
<p>
If we accept that failure, looping, timeouts, and job cancellation
are fundamental concerns in distributed systems, we may both simplify and
strengthen programs by making them fundamental expressions in a programming
language. These concepts are embodied in the simple <tt>try</tt> command:
<pre>
#!/usr/bin/ftsh
try for 5 minutes every 30 seconds
cd /work/foo
rm -rf bar
cp -r /fresh/data .
end
</pre>
<p>
Ftsh provides simple structures that encourage explicit acknowledgement
of failure while maintaining the readability of script code. You might
think of this as exceptions for scripts.
<P>
Want to learn more?
This document is a short introduction to the fault tolerant shell.
It quickly breezes over the language features in order to get
started. You can learn more about the motivation for the language in
<a href=http://www.cse.nd.edu/~ccl/software/ftsh/ethernet-hpdc12.pdf>"The Ethernet Approach to Grid Computing"</a>,
available from the <a href=http://www.cse.nd.edu/~ccl/software/ftsh>ftsh web page</a>
For a quick introduction, read on!
<a name=basics></a>
<h2>Basics</h2>
<h3>Simple Commands</h3>
An ftsh program is built up from simple commands.
A simple command names a program to be executed, just <tt>sh</tt> or <tt>csh</tt>.
The command is separated from its arguments by whitespaces.
Quotation marks may be used to escape whitespace. For example:
<pre>
ls -l /etc/hosts
</pre>
or:
<pre>
cat /etc/passwd
</pre>
or:
<pre>
cp "This File" "That File"
</pre>
<p>
As you may know, a command (a UNIX process) returns an integer known
as its "exit code." Convention dictates that an exit code of zero
indicates success while any other number indicates failure. Languages
tend to differ in their mapping of integers to success or failure, so
from here on, we will simply use the abstract terms "success" and "failure."
<p>
A command may also fail in a variety of other ways without returning an
exit code. It may be killed by a signal, or it may fail to start
altogether if the program does not exist or its image cannot be loaded.
These cases are also considered failures.
<h3>Groups</h3>
A "group" is simply a list of commands.
Each command only runs if the previous command succeeded.
Let's return our first example:
<pre>
#!/usr/bin/ftsh
cd /work/foo
rm -rf bar
cp -r /fresh/data .
</pre>
This group succeeds only if <b>every command</b> in the group succeeds.
So, if <tt>cd</tt> fails, then the whole group fails and no further commands
are executed.
<p>
This is called the "brittle" property of ftsh. If anything goes wrong,
then processing stops instantly. When something goes wrong, you will
know it, and the program will not "run away" executing more commands blindly.
We will see ways to contain the brittleness of a program below.
<p>
Ftsh itself has an exit code.
Ftsh returns the result of the top-level group that makes up the program.
So, if any command in the top-level group fails, then ftsh itself will
fail. If they all succeed, then ftsh succeeds.
<h3>Try Statements</h3>
A try statement is used to contain and retry group failure.
Here is a simple try statement:
<pre>
#!/usr/bin/ftsh
try 5 times
cd /work/foo
rm -rf bar
cp -r /fresh/data .
end
</pre>
The try statement attempts to execute the enclosed group until
the conditions in its header expire. Here, the group will
be attempted five times. Recall that a group fails as soon as any
one command fails. So, if <tt>rm</tt> fails, then the try statement
will stop and attempt the group again from the beginning.
<p>
If the five times are exceeded, then the try statement itself
fails, and (if it is the top-level try-statement) the whole
shell program itself will fail. If you prefer, you may
catch and react to a try statement in a manner similar to
an exception. The <tt>failure</tt> keyword may be used
to cause a new exception, just like <tt>throw</tt> in other languages.
<pre>
try 5 times
cd /work/foo
rm -rf bar
cp -r /fresh/data .
catch
echo "Oops, it failed. Oh well!"
failure
end
</pre>
<p>
Try statements come in several forms.
They may limit the number of times the group is executed.
For example:
<pre>
try for 10 times
</pre>
A try statement may allow an unlimited number of loops,
terminated by a maximum amount of time, given in seconds,
minutes, hours, or days:
<pre>
try for 45 seconds
</pre>
Both may be combined, yielding a try statement that stops
when either the loop limit or the time limit expires:
<pre>
try for 3 days or 100 times
</pre>
Note that such an statement does not limit the length
of any single attempt to execute the contained group.
If a single command is delayed for three days, the try statement
will wait that long and then kill the command.
To force individual attempts to be shorter, try statements
may be nested. For example:
<pre>
try for 3 days or 100 times
try for 1 time or 1 minute
/bin/big-simulation
end
end
</pre>
Here, <tt>big-simulation</tt> will be executed for
no more than a minute at a time. Such one-minute attempts
will be tried for up to three days or one hundred attempts
before the outer try statement fails.
<p>
By default, ftsh uses an exponential backoff.
If a group fails once, ftsh will wait one second, and then
try again. If it fails again, it will wait 2 seconds,
then 4 seconds, and so on, doubling the waiting time after
each failure, up to a maximum of one hour.
This prevents failures from consuming
excessive resources in fruitless retries.
<p>
If you prefer to have the retries occur at regular
intervals (though we don't recommend it)
use the <tt>every</tt> keyword to control
how frequently errors are retried. For example:
<pre>
try for 3 days every 1 hour
try for 10 times every 30 seconds
try for 1 minute or 3 times every 15 seconds
</pre>
If a time limit expires in the middle of a try statement, then
the currently running command is forcibly cancelled. If an
<tt>every</tt> clause is used, it merely ensures that each attempt
is at <b>least</b> that long. However, group will not be cancelled
merely to satisfy an <tt>every</tt> clause. To ensure that a single
loop attempt will be time limited, you may combine two try
statements as above:
<pre>
try for 3 days or 100 times <b>every 1 minute</b>
try for 1 time or <b>1 minute</b>
/bin/big-simulation
end
end
</pre>
Try statements themselves return either success or failure
in the same way as a simple command.
If the enclosed group finally succeeds, then the try
expression itself succeeds. If the try expression exhausts
its attempts, then the try statement itself fails.
We will make use of this success or failure value in the next section.
<p>
Cancelling a process is somewhat more complicated than one might think.
For all the details on how this actually works, see the section
on cancelling processes below.
<p>
In (almost) all cases, a try statement absolutely controls
what comes inside of it. There are two ways for a subprogram
to break out of the control of a try. The first is to invoke
<tt>exit</tt>, which causes the entire ftsh process to exit
immediately with the given exit code. The second is to call
<tt>exec</tt>, which causes the given process to be run
in place of the current shell process, thus voiding any
surrounding controls.
<h3>Redirection</h3>
Ftsh uses Bourne shell style I/O redirection. For example:
<pre>
echo "hello" > outfile
</pre>
...sends the output <tt>hello</tt> into the file <tt>outfile</tt>,
likewise, ftsh supports many of the more arcane redirections
of the Bourne shell, such as the redirection of explicit file descriptors:
<pre>
grep needle 0<infile 1>outfile 2>errfile
</pre>
...appending to output files:
<pre>
grep needle >>outfile 2>>errfile
</pre>
...redirection to an open file descriptor:
<pre>
grep needle >outfile 2>&1
</pre>
...and redirection of both input and output at once:
<pre>
grep needle >& out-and-err-file
</pre>
<h2>Variables</h2>
Ftsh provides variables similar to that of the Bourne shell.
For example,
<pre>
name=Douglas
echo "Hello, ${name}!"
echo "Hello, $(name)!"
echo "Hello, $name!"
</pre>
Ftsh also allows variables to be the source and target
of redirections. That is, a variable may act as a file!
The benefit of this approach is that ftsh manages the storage
and name space of variables for you. You don't have to worry
about cleaning up or clashing with other programs.
<p>
Variable redirection looks just like file redirection, except
a dash is put in front of the redirector. For example,
For example, suppose that we want to capture the output of
<tt>grep</tt> and then run it through <tt>sort</tt>:
<pre>
grep needle /tmp/haystack -> needles
sort -< needles
</pre>
This sort of operation takes the place of a pipeline,
which ftsh does not have (yet). However, by using variables
instead of pipelines, different retry conditions may be
placed on each stage of the work:
<pre>
try for 5 times
grep needle /tmp/haystack -> needles
end
try for 1 hour
sort -< needles
end
</pre>
<p>
All of the variations on file redirection are available for variable
redirection, including -> and 2-> and ->>
and 2->> and ->& and ->>&.
<p>
Like the Bourne shell, several variable names are reserved.
Simple integers are used to refer
to the command line arguments given to ftsh itself.
<tt>$$</tt> names the current process.
<tt>$#</tt> gives the number of arguments passed to the program.
<tt>$*</tt> gives all of the unquoted current arguments, while
<tt>"$@"</tt> gives all of the arguments individually quoted.
The <tt>shift</tt> command can be used to pop off the
first positional argument.
<p>
Variables are implemented by creating temporary files and immediately unlinking
them after creation. Thus, no matter how ftsh exits
-- even if it crashes -- the kernel deletes buffer space after you.
This prevents both the namespace and garbage
collection problem left by scripts that manually read and write to files.
<a name=structures></a>
<h2>Structures</h2>
Complex programs are built up by combining basic elements
with programming structures. Ftsh has most of the decision
elements of other programming languages, such as conditionals
and loops. Each of these elements behaves in a very precise way with
respect to successes and failures.
<h3>For-Statements</h3>
A for-statement executes a command group once for each
word in a list. For example:
<pre>
for food in bread wine meatballs
echo "I like ${food}"
end
</pre>
Of course, the list of items may also come from a variable:
<pre>
packages="bread.tar.gz wine.tar.gz meatballs.tar.gz"
for p in ${packages
echo "Unpacking package ${p}..."
tar xvzf ${p}
end
</pre>
The more interesting variations are <tt>forany</tt> and <tt>forall</tt>.
A <tt>forany</tt> attempts to make a group succeed once for any of the options given
in the header, chosen randomly. After the for-statement has run, the branch that
succeeds in made available through the control variable:
<pre>
hosts="mirror1.wisc.edu mirror2.wisc.edu mirror3.wisc.edu"
forany h in ${hosts}
echo "Attempting host ${host}"
wget http://${h}/some-file
end
echo "Got file from ${h}"
</pre>
A <tt>forall</tt> attempts to make a group succeed for all of the options
given in the header, simultaneously:
<pre>
forall h in ${hosts}
ssh ${h} reboot
end
</pre>
Both <tt>for</tt> and <tt>forall</tt> are brittle with respect to failures.
If any instance fails, then the entire for-statement
fails. A try-statement may be added in one of two ways.
If you wish to make each iteration resilient, place
the try-statement inside the for-statement:
<pre>
for p in ${packages}
try for 1 hour every 5 minutes
echo "Unpacking package ${p}..."
tar xvzf ${p}
end
end
</pre>
Or, if you wish to make the entire for-statement
restart after a failure, place it outside:
<pre>
try for 1 hour every 5 minutes
for p in ${packages}
echo "Unpacking package ${p}..."
tar xvzf ${p}
end
end
</pre>
<h3>Loops, Conditionals, and Expressions</h3>
Ftsh has loops and conditionals similar to other languages.
For example:
<pre>
n=0
while $n .lt. 10
echo "n is now ${n}"
n=$n .add. 1
end
</pre>
And:
<pre>
if $n .lt. 1000
echo "n is less than 1000"
else if $n .eq. 1000
echo "n is equal to 1000"
else
echo "n is greater than 1000"
end
</pre>
You'll notice right away that arithmetic expressions look a little
different than other languages.
Here's how it works:
<p>
The arithmetic operators .add. .sub. .mul. .div. .mod. .pow.
represent addition, subtraction, multiplication, division, modulus,
and exponentiation, including parenthesis and the usual order
of operations. For example:
<pre>
a=$x .mul. ( $y .add. $z )
</pre>
The comparison operators .eq. .ne. .le. .lt. .ge. .gt
represent equal, not-equal, less-than-or-equal, less-than,
greater-than-or-equal, and greater-than. These return
the literal strings "true" and "false".
<pre>
uname -s -> n
if $n .ne. Linux
...
end
</pre>
For integer comparison, use the operators .eql. and .neql..
<pre>
if $x .eql. 5
...
end
</pre>
The Boolean operators .and. .or .not. have the usual meaning.
An exception is thrown if they are given arguments that are
not the boolean strings "true" or "false".
<pre>
while ( $x .lt. 10 ) .and. ( $y .gt. 20 )
...
end
</pre>
The unary file operators .exists. .isr. .isw. .isx. test whether
a filename exists, is readable, writeable, or executable, respectively.
The similar operators .isfile. .isdir. .issock. .isfile. .isblock. .ischar.
test for the type of a named file.
All these operators throw exceptions if the named file is unavailable
for examination.
<pre>
f=/etc/passwd
if ( .isfile. $f ) .and. ( .isr. $f )
...
end
</pre>
Finally, the .to. and .step. operators are conveniences for generating
numeric lists to be used with for-loops:
<pre>
forall x in 1 .to. 100
ssh c$x reboot
end
for x in 1 .to. 100 .step. 5
y=$x .mul. $x
echo "$x times $x is $y"
end
</pre>
Notice that, unlike other shells, there is a distinction
between expressions, which compute a value or throw an
exception, and external commands, which return no value.
Therefore, you cannot do this:
<pre>
# !!! This is wrong !!!
if rm $f
echo "Removed $f"
else
echo "Couldn't remove $f"
end
</pre>
Instead, you want this:
<pre>
try
rm $f
echo "Removed $f"
catch
echo "Couldn't remove $f"
end
</pre>
<h3>Functions</h3>
Simple functions are named groups of commands that may
be called in the same manner as an external program.
The arguments passed to the function are available
in the same way as arguments to the shell:
<pre>
function compress_and_move
echo "Working on ${1}..."
gzip ${1}
mv ${1}.gz ${2}
end
compress_and_move /etc/hosts /tmp/hosts.gz
compress_and_move /etc/passwd /tmp/passwd.gz
compress_and_move /usr/dict/words /tmp/dict.gz
</pre>
A function may also be used to compute and return
a value:
<pre>
function fib
if $1 .le. 1
return 1
else
return fib($1 .sub. 1) .add. fib($1 .sub. 2)
end
end
value=fib(100)
echo $value
</pre>
Functions, like groups, are brittle with respect
to failures. A failure inside a function causes
the entire function to stop and fail immediately.
As in most languages, functions may be both nested
and recursive. However, ftsh aborts recursive
function calls deeper than 1000 steps.
If a function is used in an expression but does not
return a value, then the expression evaluation fails.
<a name=miscellaneous></a>
<h2>Miscellaneous Features</h2>
<h3>Environment</h3>
Variables may be exported into the environment, just like the Bourne shell:
<pre>
PATH="/usr/bin:/usr/local/bin"
export PATH
</pre>
<h3>Nested Shells</h3>
Ftsh is perfectly safe to nest.
That is, an ftsh script may safely call other scripts written in ftsh.
One ftsh passes all of its options to sub-shells using environment
variables, so logs, error settings, and timeouts are uniform from
top to bottom. If a sub-shell provides its own arguments, these
override the environment settings of the parent.
<h3>Error Logging</h3>
Ftsh may optionally keep a log that describes all the details
of a script's execution. The -f option specifies a log file.
Logs are open for appending, so parallel and sub-shells may
share the same log. The time, process number, script, and
line number are all recorded with every event.
<p>
<b>Note: Logs shared between processes must not be recorded in NFS or AFS filesystems.
NFS is not designed to support shared appending: your logs
will be corrupted sooner or later. AFS is not designed to support
simultaneous write sharing of a file: you will end up with the
log of one process or another, but not both. These are deliberate
design limitations of these filesystems and are not bugs in
UNIX or ftsh.
</b>
<p>
The amount of detail kept in a log
is controled with the -l option. These logging
levels are currently defined:
<ul>
<li><b>0</b> - Nothing is logged.
<li><b>10</b> - Display failed commands and structures.
<li><b>20</b> - Display executed commands and their exit codes.
<li><b>30</b> - Display structural elements such as TRY and IF-THEN.
<li><b>40</b> - Display process activities such as signals and completions.
</ul>
<h3>Command-Line Arguments</h3>
Ftsh accepts the following command-line arguments:
<ul>
<li> <b>-f <file></b> The name of a log file for tracing the
execution of a script. This log file is opened in append mode.
Equivalent to the environment variable FTSH_LOG_FILE.
<li> <b>-l <level></b> The logging level, on a scale of 0 to 100.
Higher numbers log more data about a script.
Equivalent to the environment variable FTSH_LOG_LEVEL.
<li> <b>-k <mode></b> - Controls whether ftsh trusts the operating
system to actually kill a process. If set to 'weak', ftsh will assume
that processes die without checking. If set to 'strong', ftsh will issue
SIGKILL repeatedly until a process actually dies.
Equivalent to the environment variable FTSH_KILL_MODE.
<li> <b>-t <secs></b> - The amount of time ftsh will wait
between requesting a process to exit (SIGTERM) and killing it forcibly (SIGKILL).
Equivalent to the environment variable FTSH_KILL_TIMEOUT.
<li> <b>-p</b> Parse, but do not execute the script. This option may be used to test the validity of an Ftsh script.
<li> <b>-v</b> Show the version of ftsh.
<li> <b>-h</b> Show the help screen.
</ul>
<h3>Cancelling Processes</h3>
Cancelling a running process in UNIX is rather quite complex.
Although starting and stopping one single process is fairly
easy, there are several complications to manging a tree of
processes, as well as dealing with the various failures that
can occur in the transmission of a signal.
<p>
Ftsh can clean up any set of processes that it starts,
given the following restrictions:
<dir>
<li> Your programs must not create a new UNIX "process session"
with the <tt>setsid()</tt> system call. If you don't know what
this is, then don't worry about it.
<li> The operating system must actually kill a process when
ftsh asks it to. Some variants of Linux won't kill processes
using distributed file systems. Consider using the "weak" mode
of ftsh.
<li> Rogue system administrators must not forcibly kill
an ftsh with a SIGKILL. However, you may safely send a SIGTERM, SIGHUP,
SIGINT, or SIGQUIT to an ftsh, and it will clean up its children and exit.
</dir>
<p>
Ftsh starts every command as a separate UNIX process in its
own process session (i.e. <tt>setsid</tt>).
This simplifies the administration of
large process tress. To cancel a command, ftsh sends a SIGTERM
to every process in the group. Ftsh then waits up to
thirty seconds for the child to exit willingly. At the end of
that time, it forcibly terminates the entire process group
with a SIGKILL.
<p>
Surprisingly, SIGKILL is not always effective.
Some operating systems have bugs in which signals
are occasionally lost or the process may be in such
a state that it cannot be killed at all.
By default, ftsh tries very hard to kill processes
by issuing SIGKILL repeatedly until the process actually
dies. This is called the "strong" kill mode.
If you do not wish to have this behavior -- perhaps
you have a bug resulting in unkillable processes --
then you may run ftsh in the "weak" kill mode, using
the "-k weak" option.
<p>
Ftsh may be safely nested. That is, an ftsh may invoke
another program written using ftsh. However, this child
needs to clean up faster than its parents. If the parent
shell issues forcible kills after waiting for 30 seconds,
then the child must issue forcible kills before that.
This problem is handled transparently for you.
Each ftsh informs its children of the current kill timeout
by setting the FTSH_KILL_TIMEOUT variable to five seconds
less than the current timeout. Thus, subshells are progressively
less tolerant of programs that refuse to exit cleanly.
<hr>
</body>
</html>
|