Fixed bug #34955: Timed-out processes are not identified as failures.

Fixed bug #34954: --timeout should kill process groups.
Released as 20111209 alpha.
This commit is contained in:
Ole Tange 2011-12-09 23:25:20 +01:00
parent ee4b2c7406
commit 8a7f45c6a2
13 changed files with 125 additions and 106 deletions

2
NEWS
View file

@ -241,7 +241,7 @@
20110422
* {#} now works an alias for $PARALLEL_PID.
* {#} now works as an alias for $PARALLEL_PID.
* --eta now estimates way more accurately.

29
README
View file

@ -5,35 +5,44 @@ Please send problems and feedback to bug-parallel@gnu.org.
= Presentation of GNU Parallel =
GNU parallel is a shell tool for executing jobs in parallel using one
GNU Parallel is a shell tool for executing jobs in parallel using one
or more computers. A job is can be a single command or a small script
that has to be run for each of the lines in the input. The typical
input is a list of files, a list of hosts, a list of users, a list of
URLs, or a list of tables. A job can also be a command that reads from
a pipe. GNU parallel can then split the input and pipe it into
a pipe. GNU Parallel can then split the input and pipe it into
commands in parallel.
If you use xargs and tee today you will find GNU parallel very easy to
use as GNU parallel is written to have the same options as xargs. If
you write loops in shell, you will find GNU parallel may be able to
If you use xargs and tee today you will find GNU Parallel very easy to
use as GNU Parallel is written to have the same options as xargs. If
you write loops in shell, you will find GNU Parallel may be able to
replace most of the loops and make them run faster by running several
jobs in parallel.
GNU parallel makes sure output from the commands is the same output as
GNU Parallel makes sure output from the commands is the same output as
you would get had you run the commands sequentially. This makes it
possible to use output from GNU parallel as input for other programs.
possible to use output from GNU Parallel as input for other programs.
= Installation =
Installation of GNU Parallel is a simple as:
Installation of GNU Parallel is as simple as:
./configure && make && make install
If you are not root you can install in ~/bin and ~/share:
If you are not root you can add ~/bin to your path and install in
~/bin and ~/share:
./configure --prefix=$HOME && make && make install
Or if your system lacks 'make' you can simply copy src/parallel
src/sem src/niceload src/sql to a dir in your path.
If you just need parallel in a pinch:
wget http://git.savannah.gnu.org/cgit/parallel.git/plain/src/parallel
chmod 755 parallel
After this you should be able to do:
parallel traceroute ::: foss.org.my gnu.org freenetproject.org
@ -42,7 +51,7 @@ This will run traceroute to 3 different hosts in parallel and print
the output when the traceroutes are done.
Watch the intro video for a quick introduction:
http://www.youtube.com/watch?v=OpaiGYxkSuQ or at
https://www.youtube.com/playlist?list=PL284C9FF2488BC6D1 or at
http://tinyogg.com/watch/TORaR/ and http://tinyogg.com/watch/hfxKj/

20
configure vendored
View file

@ -1,6 +1,6 @@
#! /bin/sh
# Guess values for system-dependent variables and create Makefiles.
# Generated by GNU Autoconf 2.68 for parallel 20111122.
# Generated by GNU Autoconf 2.68 for parallel 20111209.
#
# Report bugs to <bug-parallel@gnu.org>.
#
@ -559,8 +559,8 @@ MAKEFLAGS=
# Identity of this package.
PACKAGE_NAME='parallel'
PACKAGE_TARNAME='parallel'
PACKAGE_VERSION='20111122'
PACKAGE_STRING='parallel 20111122'
PACKAGE_VERSION='20111209'
PACKAGE_STRING='parallel 20111209'
PACKAGE_BUGREPORT='bug-parallel@gnu.org'
PACKAGE_URL=''
@ -1176,7 +1176,7 @@ if test "$ac_init_help" = "long"; then
# Omit some internal or obsolete options to make the list less imposing.
# This message is too long to be a string in the A/UX 3.1 sh.
cat <<_ACEOF
\`configure' configures parallel 20111122 to adapt to many kinds of systems.
\`configure' configures parallel 20111209 to adapt to many kinds of systems.
Usage: $0 [OPTION]... [VAR=VALUE]...
@ -1242,7 +1242,7 @@ fi
if test -n "$ac_init_help"; then
case $ac_init_help in
short | recursive ) echo "Configuration of parallel 20111122:";;
short | recursive ) echo "Configuration of parallel 20111209:";;
esac
cat <<\_ACEOF
@ -1309,7 +1309,7 @@ fi
test -n "$ac_init_help" && exit $ac_status
if $ac_init_version; then
cat <<\_ACEOF
parallel configure 20111122
parallel configure 20111209
generated by GNU Autoconf 2.68
Copyright (C) 2010 Free Software Foundation, Inc.
@ -1326,7 +1326,7 @@ cat >config.log <<_ACEOF
This file contains any messages produced by compilers while
running configure, to aid debugging if configure makes a mistake.
It was created by parallel $as_me 20111122, which was
It was created by parallel $as_me 20111209, which was
generated by GNU Autoconf 2.68. Invocation command line was
$ $0 $@
@ -2141,7 +2141,7 @@ fi
# Define the identity of the package.
PACKAGE='parallel'
VERSION='20111122'
VERSION='20111209'
cat >>confdefs.h <<_ACEOF
@ -2704,7 +2704,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
# report actual input values of CONFIG_FILES etc. instead of their
# values after options handling.
ac_log="
This file was extended by parallel $as_me 20111122, which was
This file was extended by parallel $as_me 20111209, which was
generated by GNU Autoconf 2.68. Invocation command line was
CONFIG_FILES = $CONFIG_FILES
@ -2766,7 +2766,7 @@ _ACEOF
cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
ac_cs_config="`$as_echo "$ac_configure_args" | sed 's/^ //; s/[\\""\`\$]/\\\\&/g'`"
ac_cs_version="\\
parallel config.status 20111122
parallel config.status 20111209
configured by $0, generated by GNU Autoconf 2.68,
with options \\"\$ac_cs_config\\"

View file

@ -1,4 +1,4 @@
AC_INIT([parallel], [20111122], [bug-parallel@gnu.org])
AC_INIT([parallel], [20111209], [bug-parallel@gnu.org])
AM_INIT_AUTOMAKE([-Wall -Werror foreign])
AC_CONFIG_HEADERS([config.h])
AC_CONFIG_FILES([

View file

@ -1,3 +1,12 @@
Luk filen ved EOF - lad være med bare at læse videre.
> /tmp/ged; tail -f /tmp/ged| xargs -n1 -E eof & sleep 1; echo echo a >>/tmp/ged; echo eof >>/tmp/ged; seq 10 >>/tmp/ged; wait
> /tmp/ged; tail -f /tmp/ged| parallel -n1 -E eof & sleep 1; echo echo a >>/tmp/ged; echo eof >>/tmp/ged; seq 10 >>/tmp/ged; wait
Fhqwhgads som 20120122 pga 10 års jubliæum.
--timeout auto: If a jobs takes > 3*average runtime then kill it. Only after job 3.
niceload seeks last column:
@ -182,6 +191,16 @@ sem
single file installation
./configure --prefix=$HOME && make && make install
Or if your system lacks 'make' you can simply copy src/parallel
src/sem src/niceload src/sql to a dir in your path.
wget http://git.savannah.gnu.org/cgit/parallel.git/plain/src/parallel
== Video: GNU Parallel 20110522 - The Pakistan Release ==

View file

@ -25,10 +25,10 @@ http://nd.gd/0s http://www.youtube.com/watch?v=OpaiGYxkSuQ
http://nd.gd/0t http://en.wikipedia.org/wiki/Xargs#The_separator_problem
http://nd.gd/3k http://www.gnu.org/software/parallel/man.html#differences_between_xargs_and_gnu_parallel
http://nd.gd/po http://www.gnu.org/software/parallel/man.html#example__distributing_work_to_local_and_remote_computers
http://nd.gd/039 http://www.youtube.com/playlist?list=PL284C9FF2488BC6D1
If you like xargs you may love GNU Parallel: http://nd.gd/0s
With GNU Parallel (http://nd.gd/0s) you can do:
ls | grep jpeg | parallel mv {} {.}.jpg
@ -55,27 +55,19 @@ problem http://en.wikipedia.org/wiki/Xargs#The_separator_problem
GNU Parallel http://www.gnu.org/software/parallel/ does not have that
problem.
Watch the intro video for GNU Parallel:
http://www.youtube.com/watch?v=OpaiGYxkSuQ
If you have GNU Parallel http://www.gnu.org/software/parallel/ installed you can do this:
Watch the intro video for GNU Parallel to learn more:
http://www.youtube.com/watch?v=OpaiGYxkSuQ
You can install GNU Parallel simply by:
wget http://git.savannah.gnu.org/cgit/parallel.git/plain/src/parallel
chmod 755 parallel
cp parallel sem
Watch the intro videos for GNU Parallel to learn more: http://nd.gd/039
GNU Parallel also makes it possible to run small scripts. Try this:
ls *.zip | parallel mkdir {.}; cd {.}; unzip ../{}
ls *.zip | parallel 'mkdir {.}; cd {.}; unzip ../{}'
= Other people says =
fawxtin: Using GNU Parallel to run things simultaneously, neat! ;)
cowsandmilk: if you haven't quit xargs for gnu parallel yet, you're really missing out
afader: GNU Parallel is cool
ecyrd: Okay, why hasn't anyone told me of GNU parallel before?
cashion: very cool! Did not know about GNU parallel.
toorghezi: Writing shell scripts? You should know about GNU Parallel!
Racecar564 : GNU Parallel rocks!
arclight : Very nice! I especially like the {} -> {.}.bz2 syntax.
nwp8861 : I surprised that it's so comfortable!

View file

@ -140,7 +140,7 @@ http://freshmeat.net/projects/parallel/releases/new
== Update Twitter ==
New version of #GNU Parallel released. See what is new in this release
http://nd.gd/2j Watch the intro video http://nd.gd/0s
http://nd.gd/2j Watch the intro video http://nd.gd/039
https://savannah.gnu.org/news/?group=parallel
@ -182,9 +182,9 @@ cc:Sandro Cazzaniga <kharec@mandriva.org>,
Ryoichiro Suzuki <ryoichiro.suzuki@gmail.com>, Jesse Alama
<jesse.alama@gmail.com>
Subject: GNU Parallel 20111122 ('Silvio') released
Subject: GNU Parallel 20111222 ('?') released
GNU Parallel 20111122 ('Silvio') has been released. It is
GNU Parallel 20111122 ('?') has been released. It is
available for download at: http://ftp.gnu.org/gnu/parallel/
This is a bugfix release with no new features. Probably a good release
@ -192,44 +192,8 @@ for stable long-term use.
New in this release:
* GNU Parallel package for Fedora.
https://admin.fedoraproject.org/pkgdb/acls/name/parallel
* Using GNU Parallel with Mozilla cross referencing.
https://blog.mozilla.com/it/2011/11/15/mxr-improvements/
* Neat flash presentation of GNU Parallel.
http://prezi.com/0nq74b2elgtx/about-gnu-parallel/
* How to create tar files of dirs in parallel.
https://joneslee85.wordpress.com/2011/11/22/shell-tips-how-to-recursively-archive-many-folders/
* Article in Admin Network&Security by Ben Martin.
http://www.admin-magazine.com/HPC/Articles/GNU-Parallel
* Blog post in Polish.
http://blog.tiger.com.pl/2011/01/05/gnu-parallel/
* Blog post on using GNU Parallel of Mac OS X.
http://www.0xcb0.com/2011/10/19/running-parallel-bash-tasks-on-os-x/
* Blog post in English about using zip with GNU Parallel.
http://nuclear-imaging.info/site_content/2011/05/11/gnu-parallel/
* Blog post in English by researcher.
http://zandyware.wordpress.com/2011/10/15/gnu-parallell-make-best-use-of-your-multicore-computer/
* Blog entry in Chinese.
http://blog.csdn.net/tossense/article/details/6889026
* Blog entry in Japanese on managing many servers.
http://hiroakis.com/blog/2011/11/13/gnu-parallel/
* GNU Parallel is now part of ppbs.
http://physiology.med.cornell.edu/faculty/mason/lab/ppbs/introduction/introduction.html
* How to use GNU Parallel with MCX.
http://mcx.sourceforge.net/cgi-bin/index.cgi?MMC/Doc/MMCCluster
* Interview in Hacker Public Radio.
http://hackerpublicradio.org/eps.php?id=0860
* Bug fixes and man page updates.

View file

@ -24,7 +24,7 @@
use strict;
use Getopt::Long;
$Global::progname="niceload";
$Global::version = 20111122;
$Global::version = 20111209;
Getopt::Long::Configure("bundling","require_order");
get_options_from_array(\@ARGV) || die_usage();
if($::opt_version) {

View file

@ -502,7 +502,7 @@ sub get_options_from_array {
sub parse_options {
# Returns: N/A
# Defaults:
$Global::version = 20111122;
$Global::version = 20111209;
$Global::progname = 'parallel';
$Global::infinity = 2**31;
$Global::debug = 0;
@ -1691,6 +1691,7 @@ sub usage {
"{3} {3.} {3/} {3/.} Positional replacement strings",
"",
"-S sshlogin Example: foo\@server.example.com",
"--slf .. Use ~/.parallel/sshloginfile as the list of sshlogins",
"--trc {}.bar Shorthand for --transfer --return {}.bar --cleanup",
"--onall Run the given command with argument on all sshlogins",
"--nonall Run the given command with no arguments on all sshlogins",
@ -3033,19 +3034,44 @@ sub timedout {
sub kill {
# kill the jobs
my $self = shift;
my $pid = $self->pid();
for my $signal ("TERM", "TERM", "KILL") {
my @family_pids = $self->family_pids();
# Record this jobs as failed
$self->set_exitstatus(1);
# Send two TERMs to give time to clean up
for my $signal ("TERM", "TERM", "KILL") {
my $alive = 0;
for my $pid (@family_pids) {
if(kill 0, $pid) {
# The job still running
kill $signal, $pid;
if($signal eq "TERM") { ::usleep(200); }
} else {
# The job is already finished
last;
$alive = 1;
}
}
# Wait 200 ms between TERMs - but only if any pids are alive
if($signal eq "TERM" and $alive) { ::usleep(200); }
}
}
sub family_pids {
# Find the pids with this->pid as (grand)*parent
# TODO test this on different OS as 'ps' is known to be different
my $self = shift;
my $pid = $self->pid();
my $script = q{
family_pids() {
for CHILDPID in `ps --ppid "$@" -o pid --no-headers`; do
family_pids $CHILDPID &
done
echo "$@"
}
} .
"family_pids $pid; wait";
my @pids = qx{$script};
chomp(@pids);
return ($pid,@pids);
}
sub failed {
# return number of times failed for this $sshlogin
my $self = shift;
@ -3584,7 +3610,14 @@ sub exitstatus {
sub set_exitstatus {
my $self = shift;
my $exitstatus = shift;
if($exitstatus) {
# Overwrite status if non-zero
$self->{'exitstatus'} = $exitstatus;
} else {
# Set status but do not overwrite
# Status may have been set by --timeout
$self->{'exitstatus'} ||= $exitstatus;
}
}
sub exitsignal {

View file

@ -460,18 +460,13 @@ B<--gnu> takes precedence.
=item B<-g> (-g will be retired 20120122)
Group output. Output from each jobs is grouped together and is only
printed when the command is finished. stderr (standard error) first
followed by stdout (standard output). B<--group> is the default. Can be
reversed with B<-u>.
Group output. Output from each jobs is grouped together and is only
printed when the command is finished. stderr (standard error) first
followed by stdout (standard output). This takes some CPU time. In
rare situations GNU B<parallel> takes up lots of CPU time and if it is
acceptable that the output from different commands are mixed together,
then disabling grouping with B<-u> can speedup GNU B<parallel> by a
factor of 10.
acceptable that the outputs from different commands are mixed
together, then disabling grouping with B<-u> can speedup GNU
B<parallel> by a factor of 10.
B<--group> is the default. Can be reversed with B<-u>.
@ -902,8 +897,8 @@ If B<--recstart> is given I<startstring> will be used to split at record start.
If B<--recend> is given I<endstring> will be used to split at record end.
If both B<--recstart> and B<--recend> are given the string
I<startstring>I<endstring> will have to match to find a split
If both B<--recstart> and B<--recend> are given the combined string
I<endstring>I<startstring> will have to match to find a split
position. This is useful if either I<startstring> or I<endstring>
match in the middle of a record.

View file

@ -556,7 +556,7 @@ $Global::Initfile && unlink $Global::Initfile;
exit ($err);
sub parse_options {
$Global::version = 20111122;
$Global::version = 20111209;
$Global::progname = 'sql';
# This must be done first as this may exec myself

View file

@ -87,6 +87,9 @@ echo '### Test -x'
echo '### Test -a and --arg-file: Read input from file instead of stdin'
seq 1 10 >/tmp/$$-1; parallel -k -a /tmp/$$-1 echo
seq 1 10 >/tmp/$$-2; parallel -k --arg-file /tmp/$$-2 echo
echo '### Test killing children with --timeout and exit value (failed if timed out)'
pstree |grep sleep | wc; parallel --timeout 3 'true {} ; for i in `seq 100 120`; do bash -c "(sleep $i)" & sleep $i & done; wait; echo No good' ::: 1000000000 1000000001 ; echo $?; pstree |grep sleep | wc
EOF
#echo '### Test bugfix if no command given'

View file

@ -172,6 +172,10 @@ xargs: argument line too long
8
9
10
### Test killing children with --timeout and exit value (failed if timed out)
0 0 0
2
0 0 0
xargs Expect: 3 1 2
3
1
@ -288,8 +292,8 @@ line 2
### Test --no-run-if-empty and -r: This should give no output
### Test --help and -h: Help output (just check we get the same amount of lines)
Output from -h and --help
27
27
28
28
### Test --version: Version output (just check we get the same amount of lines)
12
### Test --verbose and -t