From padb at googlecode.com Fri Dec 3 18:01:29 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Fri, 03 Dec 2010 18:01:29 +0000 Subject: [padb] r421 committed - Add support for using clush as a backend launcher.... Message-ID: <0015175cdab8ec9fd904968553a9@google.com> Revision: 421 Author: apittman at gmail.com Date: Fri Dec 3 10:00:39 2010 Log: Add support for using clush as a backend launcher. This is similar to the pdsh support but uses clush instead. http://sourceforge.net/projects/clustershell/ The mechanism for launch is the same as pdsh so presumably the performance is the same at scale, it might suit some people to use clush rather than pdsh though. Pdsh will still be used in preference but clush will be selected if pdsh cannot be found or if the user selects it. http://code.google.com/p/padb/source/detail?r=421 Modified: /trunk/src/padb ======================================= --- /trunk/src/padb Sun Nov 28 13:43:33 2010 +++ /trunk/src/padb Fri Dec 3 10:00:39 2010 @@ -575,7 +575,7 @@ $conf{tree_width} = '4'; -$conf{launch_mode} = 'local,rmgr,ssh,pdsh'; +$conf{launch_mode} = 'local,rmgr,ssh,pdsh,clush'; # Config options which take boolean values. my @conf_bool = qw(watch_clears_screen inner_callback); @@ -3753,7 +3753,8 @@ @hosts = @{ $pcmd{host_list} }; } - my $have_pdsh = find_exe('pdsh'); + my $have_pdsh = find_exe('pdsh'); + my $have_clush = find_exe('clush'); foreach my $mode (@modes) { if ( $mode eq 'local' ) { @@ -3791,12 +3792,25 @@ return %pcmd; + } elsif ( $mode eq 'clush' ) { + next unless ($have_clush); + next if ( @hosts == 0 ); + + my $fanout = @hosts + 5; + + $pcmd{require_inner_callback} = 1; + my $hlist = join q{,}, @hosts; + $pcmd{command} = "clush -f $fanout --nostdin -w $hlist"; + + return %pcmd; + } else { print "Backend invalid: $mode\n"; } } - print "No suitable backend found (perhaps try installing pdsh?)!\n"; + print + "No suitable backend found (perhaps try installing pdsh or clush ?)!\n"; return; } From padb at googlecode.com Fri Dec 3 18:05:35 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Fri, 03 Dec 2010 18:05:35 +0000 Subject: [padb] r422 committed - Use the -f option to pdsh rather than the FANOUT variable. Message-ID: <0015175ccf7293321204968562bf@google.com> Revision: 422 Author: apittman at gmail.com Date: Fri Dec 3 10:04:09 2010 Log: Use the -f option to pdsh rather than the FANOUT variable. http://code.google.com/p/padb/source/detail?r=422 Modified: /trunk/src/padb ======================================= --- /trunk/src/padb Fri Dec 3 10:00:39 2010 +++ /trunk/src/padb Fri Dec 3 10:04:09 2010 @@ -3777,17 +3777,14 @@ next unless ($have_pdsh); next if ( @hosts == 0 ); + my $fanout = @hosts + 5; + $pcmd{require_inner_callback} = 1; my $hlist = join q{,}, @hosts; - $pcmd{command} = "pdsh -w $hlist"; - - if ( @hosts > 20 ) { - my $fanout = @hosts + 5; - $ENV{FANOUT} = $fanout; - - if ( @hosts > 128 ) { - print "Pdsh backend not recommended for such large jobs\n"; - } + $pcmd{command} = "pdsh -f $fanout -w $hlist"; + + if ( @hosts > 128 ) { + print "Pdsh backend not recommended for such large jobs\n"; } return %pcmd; From ashley at pittman.co.uk Fri Dec 3 18:35:25 2010 From: ashley at pittman.co.uk (Ashley Pittman) Date: Fri, 3 Dec 2010 18:35:25 +0000 Subject: [padb] [padb-users] Upcoming release. In-Reply-To: References: Message-ID: <19A63DCB-946D-417A-B5EE-0C5820676D9A@pittman.co.uk> On 1 Nov 2010, at 19:57, Ashley Pittman wrote: > I'd like to make a formal release in the coming weeks based on the current SVN code, the 3.2 beta has been through an extended testing period and I'm happy that it's ready to move to formal release status. On this basis I propose making a 3.3 release in the next two weeks, probably on Monday the 8th. > > Please test the latest 3.2 beta or trunk and let me know of any problems you have, unless any new issues are reported by the 5th I'll go ahead as planned. A number of issues did come up but I believe these have all been resolved now so the trunk is in a good state for release. I propose a new release date of Wed 8th Dec unless anything further comes up, issues like idb support and changes to the message queue code are liable to be de-stabiling so are best left until after the branch has been made. Ashley. -- Ashley Pittman, Bath, UK. Padb - A parallel job inspection tool for cluster computing http://padb.pittman.org.uk From d.love at liverpool.ac.uk Sun Dec 5 19:08:23 2010 From: d.love at liverpool.ac.uk (Dave Love) Date: Sun, 05 Dec 2010 19:08:23 +0000 Subject: [padb] PBS false positive Message-ID: <87bp50f36g.fsf@liv.ac.uk> A minor point, but the test for the PBS resource manager gives a false positive in an SGE installation as it just looks for qstat on the path. That's also an SGE command (maybe actually defined by the POSIX batch system standard?). If it helps to disambiguate them, the SGE version invoked as `qstat -help' produces output whose first line starts with `GE '. From d.love at liverpool.ac.uk Sun Dec 5 19:16:11 2010 From: d.love at liverpool.ac.uk (Dave Love) Date: Sun, 05 Dec 2010 19:16:11 +0000 Subject: [padb] (version-dependent?) problem with ORTE Message-ID: <878w04f2tg.fsf@liv.ac.uk> I reported a while ago that I couldn't make ORTE work, and I've found out why. With open-mpi 1.4.1 or 1.4.2, the problem is that the format padb expects from ompi-ps is wrong. I assume it's changed at some stage. Here's a (truncated) sample: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ompi-ps.out URL: -------------- next part -------------- The following patch fixes it for me, but presumably it will break whatever version the support was done for originally, so I don't know what to do for a real patch. Maybe you need to match patterns in the records, rather than just checking the number of fields? Let me know if I can provide any more info to help disambiguate things. -------------- next part -------------- A non-text attachment was scrubbed... Name: padb.diff Type: text/x-diff Size: 731 bytes Desc: padb diff for OMPI 1.4.2 URL: -------------- next part -------------- By the way, in case it catches other people, I need to run with OMPI_MCA_orte_allocation_required=0 in the environment to defeat what we have in the open-mpi configuration: # mpirun requires an SGE allocation, i.e. can't just run it on the head. orte_allocation_required = 1 From ashley at pittman.co.uk Sun Dec 5 21:36:31 2010 From: ashley at pittman.co.uk (Ashley Pittman) Date: Sun, 5 Dec 2010 21:36:31 +0000 Subject: [padb] PBS false positive In-Reply-To: <87bp50f36g.fsf@liv.ac.uk> References: <87bp50f36g.fsf@liv.ac.uk> Message-ID: On 5 Dec 2010, at 19:08, Dave Love wrote: > A minor point, but the test for the PBS resource manager gives a false > positive in an SGE installation as it just looks for qstat on the path. > That's also an SGE command (maybe actually defined by the POSIX batch > system standard?). If it helps to disambiguate them, the SGE version > invoked as `qstat -help' produces output whose first line starts with > `GE '. I suspected this would happen at some point, can you give me a list of binaries that SGE installs and I'll try and find a matching list for PBS and hopefully we can find a way of distinguishing them without having to start any processes, in case that isn't possible though could you send me some sample output as well. Ashley. -- Ashley Pittman, Bath, UK. Padb - A parallel job inspection tool for cluster computing http://padb.pittman.org.uk From ashley at pittman.co.uk Sun Dec 5 22:06:39 2010 From: ashley at pittman.co.uk (Ashley Pittman) Date: Sun, 5 Dec 2010 22:06:39 +0000 Subject: [padb] (version-dependent?) problem with ORTE In-Reply-To: <878w04f2tg.fsf@liv.ac.uk> References: <878w04f2tg.fsf@liv.ac.uk> Message-ID: On 5 Dec 2010, at 19:16, Dave Love wrote: > I reported a while ago that I couldn't make ORTE work, and I've found > out why. With open-mpi 1.4.1 or 1.4.2, the problem is that the format > padb expects from ompi-ps is wrong. I assume it's changed at some > stage. Here's a (truncated) sample: This looks to be a Checkpoint dependant change, I regularly update my open-mpi install and have never had this problem, I don't have the Ckpt entries in the output though. > The following patch fixes it for me, but presumably it will break > whatever version the support was done for originally, so I don't know > what to do for a real patch. Maybe you need to match patterns in the > records, rather than just checking the number of fields? Let me know if > I can provide any more info to help disambiguate things. Yes, this would break it for most other people although I'm glad it works for you. Interestingly perl seems to be ignoring the empty fields after splitting on | so it's likely that if you were using checkpoint-restart it would also break for you. The attached should fix it and still works for me, I'm incredibly reluctant to apply this so close to release though because I've no idea what the valid values the opmi_is_valid_state() function should check for are. I'll have another go tomorrow and try adding some code to keep track of where in the output the parsing has progressed to, alternatively it's possible to make ompi_is_valid_state() always return 1 with this patch and have it still work but then it's based on matching of the jobid only which doesn't strike me as a good idea. Ashley, -------------- next part -------------- A non-text attachment was scrubbed... Name: ompi-ps-format.patch Type: application/octet-stream Size: 2043 bytes Desc: not available URL: -------------- next part -------------- -- Ashley Pittman, Bath, UK. Padb - A parallel job inspection tool for cluster computing http://padb.pittman.org.uk From d.love at liverpool.ac.uk Mon Dec 6 10:36:38 2010 From: d.love at liverpool.ac.uk (Dave Love) Date: Mon, 06 Dec 2010 10:36:38 +0000 Subject: [padb] PBS false positive In-Reply-To: (Ashley Pittman's message of "Sun, 5 Dec 2010 21:36:31 +0000") References: <87bp50f36g.fsf@liv.ac.uk> Message-ID: <87k4jnfart.fsf@liv.ac.uk> You wrote: > On 5 Dec 2010, at 19:08, Dave Love wrote: > >> A minor point, but the test for the PBS resource manager gives a false >> positive in an SGE installation as it just looks for qstat on the path. >> That's also an SGE command (maybe actually defined by the POSIX batch >> system standard?). If it helps to disambiguate them, the SGE version >> invoked as `qstat -help' produces output whose first line starts with >> `GE '. > > I suspected this would happen at some point, can you give me a list of > binaries that SGE installs and I'll try and find a matching list for > PBS and hopefully we can find a way of distinguishing them without > having to start any processes, in case that isn't possible though > could you send me some sample output as well. > > Ashley. The binaries you can probably rely on being installed for relevant versions are: qacct qhost qmod qalter qsh qconf qping qstat qdel qquota qsub. I guess the best way to distinguish SGE and PBS is to find a qstat option that SGE doesn't support, like the -n that padb actually uses, and try invoking it. Here's the qstat help output, which you'll also get for an unrecognized option. I think it's SGE version-dependent whether or not it appears on stdout or stderr. $ qstat -help GE 6.2u5 usage: qstat [options] [-cb] view additional binding specific parameters [-ext] view additional attributes [-explain a|c|A|E] show reason for c(onfiguration ambiguous), a(larm), suspend A(larm), E(rror) state [-f] full output [-F [resource_attributes]] full output and show (selected) resources of queue(s) [-g {c}] display cluster queue summary [-g {d}] display all job-array tasks (do not group) [-g {t}] display all parallel job tasks (do not group) [-help] print this help [-j job_identifier_list ] show scheduler job information [-l resource_list] request the given resources [-ne] hide empty queues [-pe pe_list] select only queues with one of these parallel environments [-q wc_queue_list] print information on given queue [-qs {a|c|d|o|s|u|A|C|D|E|S}] selects queues, which are in the given state(s) [-r] show requested resources of job(s) [-s {p|r|s|z|hu|ho|hs|hd|hj|ha|h|a}] show pending, running, suspended, zombie jobs, jobs with a user/operator/system/array-dependency hold, jobs with a start time in future or any combination only. h is an abbreviation for huhohshdhjha a is an abbreviation for prsh [-t] show task information (implicitly -g t) [-u user_list] view only jobs of this user [-U user_list] select only queues where these users have access [-urg] display job urgency information [-pri] display job priority information [-xml] display the information in XML-Format pe_list pe[,pe,...] job_identifier_list [job_id|job_name|pattern]{, [job_id|job_name|pattern]} resource_list resource[=value][,resource[=value],...] user_list user|@group[,user|@group],...] resource_attributes resource,resource,... wc_cqueue wildcard expression matching a cluster queue wc_host wildcard expression matching a host wc_hostgroup wildcard expression matching a hostgroup wc_qinstance wc_cqueue at wc_host wc_qdomain wc_cqueue at wc_hostgroup wc_queue wc_cqueue|wc_qdomain|wc_qinstance wc_queue_list wc_queue[,wc_queue,...] From d.love at liverpool.ac.uk Mon Dec 6 10:47:17 2010 From: d.love at liverpool.ac.uk (Dave Love) Date: Mon, 06 Dec 2010 10:47:17 +0000 Subject: [padb] (version-dependent?) problem with ORTE In-Reply-To: (Ashley Pittman's message of "Sun, 5 Dec 2010 22:06:39 +0000") References: <878w04f2tg.fsf@liv.ac.uk> Message-ID: <87d3pffaa2.fsf@liv.ac.uk> Ashley Pittman writes: > On 5 Dec 2010, at 19:16, Dave Love wrote: > >> I reported a while ago that I couldn't make ORTE work, and I've found >> out why. With open-mpi 1.4.1 or 1.4.2, the problem is that the format >> padb expects from ompi-ps is wrong. I assume it's changed at some >> stage. Here's a (truncated) sample: > > This looks to be a Checkpoint dependant change, I regularly update my > open-mpi install and have never had this problem, I don't have the > Ckpt entries in the output though. Oops, yes. I should have realized it must be the difference between having CR and not. It's not obvious without looking at the code whether there's a way to turn that off with an MCA configuration variable. > Yes, this would break it for most other people although I'm glad it > works for you. Interestingly perl seems to be ignoring the empty > fields after splitting on | so it's likely that if you were using > checkpoint-restart it would also break for you. Oh, yes. I really wasn't awake yesterday. I was confused initially by it ignoring empty fields, but I'm not a perl expert. It's no big problem for me, anyhow. I can have a look sometime and see if I can find a clean way round it. From padb at googlecode.com Mon Dec 6 21:23:20 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Mon, 06 Dec 2010 21:23:20 +0000 Subject: [padb] r423 committed - Handle ompi-ps output when it's compiled with support for checkpoint-r... Message-ID: <0015175cdab850a9e30496c47f33@google.com> Revision: 423 Author: apittman at gmail.com Date: Mon Dec 6 13:22:27 2010 Log: Handle ompi-ps output when it's compiled with support for checkpoint-restart. This adds extra output to the ompi-ps command which was causing the pattern matching to fail so be more careful in what we check for to ensure we match both the case when it's enabled and also the case where it isn't. http://code.google.com/p/padb/source/detail?r=423 Modified: /trunk/src/padb ======================================= --- /trunk/src/padb Fri Dec 3 10:04:09 2010 +++ /trunk/src/padb Mon Dec 6 13:22:27 2010 @@ -635,16 +635,19 @@ printf {*STDERR} "Time value \"%s\" not recognised, aborting.\n", $str; exit 1; } + +sub is_digit { + my ($str) = @_; + + return $str =~ m{\A # Start of line + \d+ # A number + \z}x +} sub check_int { my ($str) = @_; - return - if ( - $str =~ m{\A # Start of line - \d+ # A number - \z}x - ); + return if ( is_digit($str) ); printf {*STDERR} "Integer value \"%s\" not recognised, aborting.\n", $str; exit 1; @@ -3248,26 +3251,36 @@ next if ( $l eq $EMPTY_STRING ); my @elems = split qr{\s*\|\s*}, $l; - - if ( @elems == 4 ) { + next unless @elems >= 4; + +# We used to check for the number of elements here and then match the jobid seperatly but that doesn't +# work for all cases as sometimes Open-Mpi has extra checkpoint-restart related data on the end +# of each line changing the element count and hence the test to fail. Now we check the element count +# is at least as long as we expect, we check that all fields that should be integers are actually integers +# and I've moved the regexp for matching the job ID into the line test. +# For performance reasons I've put the per-rank test before the per-job test as it triggers more often. + if ( @elems >= 6 + and ( $elems[1] =~ m{\A\[\[(\d+)\,(\d+)\]\,(\d+)\]}x ) + and is_digit( $elems[2] ) + and is_digit( $elems[3] ) ) + { + my $job = $1; + my $step = $2; + my $rank = $3; + my $pid = $elems[3]; + my $host = $elems[4]; + $open_jobs{$job}{$step}{hosts}{$host}++; + $open_jobs{$job}{$step}{ranks}{$host}{$rank} = $pid; + + } elsif ( @elems >= 4 + and ( $elems[0] =~ m{\A\[(\d+)\,(\d+)]\z}x ) + and is_digit( $elems[2] ) + and is_digit( $elems[3] ) ) + { my $nprocs = $elems[3]; - my $name = $elems[0]; - if ( $name =~ m{\A\[(\d+)\,(\d+)]\z}x ) { - my $job = $1; - my $step = $2; - $open_jobs{$job}{$step}{nprocs} = $nprocs; - } - } elsif ( @elems == 6 ) { - my $name = $elems[1]; - if ( $name =~ m{\A\[\[(\d+)\,(\d+)\]\,(\d+)\]}x ) { - my $job = $1; - my $step = $2; - my $rank = $3; - my $pid = $elems[3]; - my $host = $elems[4]; - $open_jobs{$job}{$step}{hosts}{$host}++; - $open_jobs{$job}{$step}{ranks}{$host}{$rank} = $pid; - } + my $job = $1; + my $step = $2; + $open_jobs{$job}{$step}{nprocs} = $nprocs; } } return; From ashley at pittman.co.uk Mon Dec 6 21:26:29 2010 From: ashley at pittman.co.uk (Ashley Pittman) Date: Mon, 6 Dec 2010 21:26:29 +0000 Subject: [padb] (version-dependent?) problem with ORTE In-Reply-To: <87d3pffaa2.fsf@liv.ac.uk> References: <878w04f2tg.fsf@liv.ac.uk> <87d3pffaa2.fsf@liv.ac.uk> Message-ID: On 6 Dec 2010, at 10:47, Dave Love wrote: > Oh, yes. I really wasn't awake yesterday. I was confused initially by > it ignoring empty fields, but I'm not a perl expert. It's no big > problem for me, anyhow. I can have a look sometime and see if I can > find a clean way round it. I've committed a slightly modified version of the patch I sent last night, as there was a regexp test inside of each conditional block anyway it's not actually any different in having the regexp inside the test statement itself. I was slightly worried about the performance of this at scale but the number of regexps done is exactly the same and my testing shows that most of the time is spent in the split function anyway. Can you try r423 and let me know if it works for you and I'll push forward with the release. Ashley. -- Ashley Pittman, Bath, UK. Padb - A parallel job inspection tool for cluster computing http://padb.pittman.org.uk From padb at googlecode.com Wed Dec 8 15:55:52 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Wed, 08 Dec 2010 15:55:52 +0000 Subject: [padb] r424 committed - Update the release notes and version numbers in preperation... Message-ID: <0015175ccfb2e5dcdc0496e827bf@google.com> Revision: 424 Author: apittman at gmail.com Date: Wed Dec 8 07:55:22 2010 Log: Update the release notes and version numbers in preperation for a 3.3 release. http://code.google.com/p/padb/source/detail?r=424 Modified: /trunk/configure.in /trunk/src/padb ======================================= --- /trunk/configure.in Sat Oct 23 18:06:44 2010 +++ /trunk/configure.in Wed Dec 8 07:55:22 2010 @@ -1,5 +1,5 @@ AC_INIT(src/padb) -AM_INIT_AUTOMAKE(padb,3.2-beta1) +AM_INIT_AUTOMAKE(padb,3.3-beta0) AC_PROG_CC AC_PROG_INSTALL AM_PROG_CC_C_O ======================================= --- /trunk/src/padb Mon Dec 6 13:22:27 2010 +++ /trunk/src/padb Wed Dec 8 07:55:22 2010 @@ -10,7 +10,7 @@ # or email padb-users at pittman.org.uk # Copyright (C) 2005-2007 Quadrics. -# Copyright (C) 2009 Ashley Pittman. +# Copyright (C) 2009-2010 Ashley Pittman. # # This program is free software; you can redistribute it and/or # modify it under the terms of the GNU Lesser General Public @@ -28,6 +28,42 @@ # Revision history +# Version 3.3 +# * Add a 'launch-mode' option to controll the back-end used to launch +# the inner processes, previously padb would use the underlying resource +# manager if possible and pdsh if not, now this is configurable and +# padb can be told to use either, the resource manager, pdsh or clush. +# The default setting for this value is "local,rmgr,ssh,pdsh,clush" which +# mimics the old behaviour but offers the user more choice about the +# specifics +# * Change the default to be not to show paramaters and locals in +# stack traces, it's very useful to have but adds a lot of noise +# which makes it harder to spot simple problems. +# * Add proper support for viewing stack traces in threaded +# programmes, show a full tree-based stack view for each thread-id +# * Add limited for LSF jobs. +# * Better handling of gdb errors, add a catch-all signal handler +# for sigpipe such that if any undetected error occours we can +# report it rather than crash. +# * Detect errors introduced by the "yama" security feature in new +# kernels, this prevents padb from attaching to running jobs if +# it's enabled. +# * Convert between hostnames and fully-qualified-domain-names as +# required to work on systems which use either or a combination +# of both. +# * Sort values properly in proc-summary mode, if a value is numeric +# for all ranks then do a numberical sort, otherwise perform a +# dictionary sort. +# * Modify the PBS resource manager to also work on Torque. These +# two resource managers are fundamentally the same but differ +# slightly in the command line options they accept. +# * Build fixes for edb (For Quadrics systems) +# * Fix a problem with reading MPI message queues from Fortran programmes, +# switch gdb into the 'c' language mode when required rather than just +# using the default all the time. +# * Fix orte support for systems where OpenMPI was built with check-point +# restart enabled. +# # Version 3.2 # * Support of PBS Pro # * Support for OpenMPI jobs run by mpirun under a slurm allocation. @@ -85,7 +121,7 @@ # callback handle it. This means that persistent attachments can # be used in full-report mode. # * Speed up attaching gdb to the target job greatly by attaching to -# all target processes on a not simultanously rather than one at +# all target processes on a node simultanously rather than one at # a time. # * Better handling of jobs that dissapear whilst we are monitoring them, # there should be no perl errors shown if this happens. From padb at googlecode.com Wed Dec 8 18:57:31 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Wed, 08 Dec 2010 18:57:31 +0000 Subject: [padb] r425 committed - Change the version to 3.3 Message-ID: <0015175cb23e8880410496eab179@google.com> Revision: 425 Author: apittman at gmail.com Date: Wed Dec 8 10:56:59 2010 Log: Change the version to 3.3 http://code.google.com/p/padb/source/detail?r=425 Modified: /trunk/src/padb ======================================= --- /trunk/src/padb Wed Dec 8 07:55:22 2010 +++ /trunk/src/padb Wed Dec 8 10:56:59 2010 @@ -386,7 +386,7 @@ } my $prog = basename $0; -my $version = "3.2 (Revision $svn_revision)"; +my $version = "3.3 (Revision $svn_revision)"; my %conf; From padb at googlecode.com Wed Dec 8 23:14:34 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Wed, 08 Dec 2010 23:14:34 +0000 Subject: [padb] r426 committed - Make a 3.3 release branch Message-ID: <0015175cba5ecb50ed0496ee484a@google.com> Revision: 426 Author: apittman at gmail.com Date: Wed Dec 8 15:13:41 2010 Log: Make a 3.3 release branch http://code.google.com/p/padb/source/detail?r=426 Added: /branches/3.3 From padb at googlecode.com Wed Dec 8 23:42:35 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Wed, 08 Dec 2010 23:42:35 +0000 Subject: [padb] r427 committed - Make the release official. Message-ID: <000e0cd76eb8f7e4560496eeac4b@google.com> Revision: 427 Author: apittman at gmail.com Date: Wed Dec 8 15:42:04 2010 Log: Make the release official. http://code.google.com/p/padb/source/detail?r=427 Modified: /branches/3.3/configure.in ======================================= --- /branches/3.3/configure.in Wed Dec 8 07:55:22 2010 +++ /branches/3.3/configure.in Wed Dec 8 15:42:04 2010 @@ -1,5 +1,5 @@ AC_INIT(src/padb) -AM_INIT_AUTOMAKE(padb,3.3-beta0) +AM_INIT_AUTOMAKE(padb,3.3) AC_PROG_CC AC_PROG_INSTALL AM_PROG_CC_C_O From padb at googlecode.com Wed Dec 8 23:57:23 2010 From: padb at googlecode.com (padb at googlecode.com) Date: Wed, 08 Dec 2010 23:57:23 +0000 Subject: [padb] r428 committed - Update the website to point to the 3.3 release. Message-ID: <0015175cb124f1cac10496eee1df@google.com> Revision: 428 Author: apittman at gmail.com Date: Wed Dec 8 15:56:56 2010 Log: Update the website to point to the 3.3 release. http://code.google.com/p/padb/source/detail?r=428 Modified: /trunk/doc/download.html /trunk/doc/header.html /trunk/doc/index.html ======================================= --- /trunk/doc/download.html Sat Oct 23 18:39:59 2010 +++ /trunk/doc/download.html Wed Dec 8 15:56:56 2010 @@ -1,14 +1,8 @@
-

Beta release

 - -Version 3.2 is currently in Beta - and can be downloaded from the downloads section. -

Current stable release

 -The latest stable release, 3.0 can be downloaded direct from the google downloads section. This release was made on 06-10-09. +The latest stable release, 3.3 can be downloaded direct from the google downloads section. This release was made on 08-12-10.

Source code download

======================================= --- /trunk/doc/header.html Mon Dec 21 15:11:50 2009 +++ /trunk/doc/header.html Wed Dec 8 15:56:56 2010 @@ -35,7 +35,7 @@
  • Download -
    3.0 Release +
    3.3 Release
    Source code
  • Patches ======================================= --- /trunk/doc/index.html Sat Oct 23 18:13:32 2010 +++ /trunk/doc/index.html Wed Dec 8 15:56:56 2010 @@ -18,8 +18,12 @@

    Recent News

      -
    • 23-10-10: 3.2-beta1 avaliable for - download. +
    • 08-12-10: 3.3 release avaliable for + download. It is recommended all + users upgrade to version 3.3. + +
    • 23-10-10: 3.2-beta1 avaliable for + download.
    • 06-10-09: 3.0 release avaliable for download. From ashley at pittman.co.uk Thu Dec 9 00:22:55 2010 From: ashley at pittman.co.uk (Ashley Pittman) Date: Thu, 9 Dec 2010 00:22:55 +0000 Subject: [padb] Announcing padb version 3.3. Message-ID: <97F1007F-B1FF-44E7-9725-B5DAA4C3B937@pittman.co.uk> I am pleased to announce that version 3.3 of padb, the first official release of padb in over a year, is now ready for use and has been uploaded to the website this evening. Release 3.3 represents a major step forward in terms of functionality, usability and stability since 3.0 and is a recommended upgrade for all users. Major changes of note are: - The ability to display variables in tree-based stack traces. - Proper support for threaded applications, in particular the tree-based stack trace mode now reports each thread in a rank individually and makes a number of trees, one for each target thread-id. - Significantly better command line parsing, resulting in better error messages and easier configuration. - Miscellaneous performance improvements, both for absolute job size and for larger process counts within individual nodes. - Selection of back-end launch mode: it is now possible to target jobs without having to rely on the resource manager to launch in many cases. - "MPIR" interface support to enable padb to work on many more resource managers which support this standard. - Solaris port. - PBS/PBS Pro/Torque support. - Limited LFS support. For a full list of changes see the "Revision history" in the source. Many of these changes were already present in the 3.2 beta releases. However a number of improvements have been made since the last beta on this branch so existing 3.2 users should also consider upgrading. The source tarball can be downloaded from the usual downloads page on Google code or directly via: http://padb.googlecode.com/files/padb-3.3.tar.gz SHA 1 Checksum: e2ec75f0d78cfff7df1a97f29dab00ddfa24f501 Work has already started on future developments. As well as supporting an ever increasing number of resource managers, the focus is moving to new modes of operation and better ways of reporting collected information to the user. My thanks to everyone who has helped make this release what it is; I appreciate all user reports, both good and bad, and hope to be able to continue bringing you improvements to padb in the future. Ashley Pittman. -- Ashley Pittman, Bath, UK. Padb - A parallel job inspection tool for cluster computing http://padb.pittman.org.uk