From jbrooks4 at austin.rr.com  Wed Sep  1 08:40:16 2004
From: jbrooks4 at austin.rr.com (Jeremy Brooks)
Date: Wed Sep  1 08:26:44 2004
Subject: APM: Net::FTP recieves file but still reaches timeout without
	saving
In-Reply-To: <1093978940.12780.23.camel@willow.mobile-o.com>
References: <1093978940.12780.23.camel@willow.mobile-o.com>
Message-ID: <1094046016.22706.3.camel@willow.mobile-o.com>

Well,  the problem was zone-alarm running on the server.  
That begs the question, why did ncftp work just fine?

On Tue, 2004-08-31 at 14:02, Jeremy Brooks wrote:
> good afternoon,
> 
> I've written a simple test script (see inline) that connects,
> authenticates and gets a file from a Microsoft IIS 5.0 ftp server.
> In about 19 out of twenty executions the script timesout before actually
> getting the file.  However, without fail, I see the <file size> number
> of bytes received over the passive mode data connection.
> The file's size is 5169 bytes.
> 
> ----------------Netstat -an --inet--------------------
> tcp  0    25 65.16.122.206:33723  207.200.10.26:21        ESTABLISHED
> tcp  5170  0 65.16.122.206:33724  207.200.10.26:1577      CLOSE_WAIT
> ------------------------------------------------------
> 
> 
> I've traced the time out to Net::Cmd.pm line 269 where we use select to
> block on the handle used to read data from the server.  
> 
> I've enabled passive connections in Net::FTP using FTP_PASSIVE, pasv(),
> and Passive=>1.  All  these methods seem to work in that the passive
> data connection is established back the server but it's as if Net::FTP
> just doesn't do anything with it after the server sends the file.
> I've also tried rasing the timeout to 6 minutes which changed nothing
> but how long it took me to trouble shoot.  When things work, the
> transfer takes on a couple of seconds so I set Timeout=>10.
> 
> ncftp works 100% of the time from the same client machine to the same
> server.  
> 
> Have you any ideas on what's happening here and why it works
> occasionally?
> 
> 
> the script:
> -----------------------------------------------
> #!/usr/bin/perl
> 
> use Net::FTP;
> 
> $ftp =  Net::FTP->new("data.teledynamics.com",Hash=>\*STDOUT,
>                       Debug=>1,Passive=>1,Timeout=>10)
>         or die "could not make handle\n";
> 
> $ftp->login("user","pass") or die "could not login\n";
> $ftp->binary();
> 
> ### Here's where we hang
> $ftp->get("/images/ZES120.jpg") or die "could not get file\n";
> 
> 
> $ftp->quit;
> 
> print "OK ", $ftp->message,"\n";
> ---------------------------------------------------------------
> 
> 
> thanks in advance!
> 
> Jeremy Brooks
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin

From wwalker at bybent.com  Wed Sep  1 10:04:30 2004
From: wwalker at bybent.com (Wayne Walker)
Date: Thu Sep  2 19:37:59 2004
Subject: APM: Net::FTP recieves file but still reaches timeout without
	saving
In-Reply-To: <1094046016.22706.3.camel@willow.mobile-o.com>
References: <1093978940.12780.23.camel@willow.mobile-o.com>
	<1094046016.22706.3.camel@willow.mobile-o.com>
Message-ID: <20040901150430.GB14859@bybent.com>

On Wed, Sep 01, 2004 at 08:40:16AM -0500, Jeremy Brooks wrote:
> Well,  the problem was zone-alarm running on the server.  
> That begs the question, why did ncftp work just fine?

Probably ncftp defaults to passive mode ftp and Net::FTP defaults to
active (or vice versa, I can never get the modes straight).  One is
considered a serious security problem, the other a possible denial of
service attack.

-- 

Wayne Walker
wwalker@bybent.com                 Do you use Linux?!
http://www.bybent.com              Get Counted!  http://counter.li.org/
Perl - http://www.perl.org/        Perl User Groups - http://www.pm.org/
Jabber:  wwalker@jabber.ntrc.net   AIM:     lwwalkerbybent
IRC:     wwalker on freenode.net
From jbrooks4 at austin.rr.com  Fri Sep  3 15:43:13 2004
From: jbrooks4 at austin.rr.com (Jeremy Brooks)
Date: Fri Sep  3 15:29:37 2004
Subject: APM: Net::FTP recieves file but still reaches timeout without
	saving
In-Reply-To: <20040901150430.GB14859@bybent.com>
References: <1093978940.12780.23.camel@willow.mobile-o.com>
	<1094046016.22706.3.camel@willow.mobile-o.com>
	<20040901150430.GB14859@bybent.com>
Message-ID: <1094244193.32660.6.camel@willow.mobile-o.com>

Thanks Wayne,
  I'm using Passive with Net::FTP.  I have, in fact, turned on passive
using all three methods provided in the Net::FTP docs.  (see the
original post).  It turns out that the firewall on the server was
broken.  The problem has been resolved.  However, we still don't know
why ncftp worked and Net::FTP didn't.  

thanks for the help,
  Jeremy

On Wed, 2004-09-01 at 10:04, Wayne Walker wrote:
> On Wed, Sep 01, 2004 at 08:40:16AM -0500, Jeremy Brooks wrote:
> > Well,  the problem was zone-alarm running on the server.  
> > That begs the question, why did ncftp work just fine?
> 
> Probably ncftp defaults to passive mode ftp and Net::FTP defaults to
> active (or vice versa, I can never get the modes straight).  One is
> considered a serious security problem, the other a possible denial of
> service attack.

From mlehmann at marklehmann.com  Fri Sep  3 15:56:37 2004
From: mlehmann at marklehmann.com (Mark Lehmann)
Date: Fri Sep  3 15:53:06 2004
Subject: APM: Net::FTP recieves file but still reaches timeout without
	saving
In-Reply-To: <1094244193.32660.6.camel@willow.mobile-o.com>
References: <1093978940.12780.23.camel@willow.mobile-o.com>
	<1094046016.22706.3.camel@willow.mobile-o.com>
	<20040901150430.GB14859@bybent.com>
	<1094244193.32660.6.camel@willow.mobile-o.com>
Message-ID: <48775.66.45.70.241.1094244997.squirrel@www.marklehmann.com>

I suspect that the zone-alarm can detect certain types of FTP clients and
it lets the ones it thinks are safe work and blocks the others.
-- 
Mark Lehmannn
mlehmann@marklehmann.com - mobile 512 689-7705

<quote who="Jeremy Brooks">
> Thanks Wayne,
>   I'm using Passive with Net::FTP.  I have, in fact, turned on passive
> using all three methods provided in the Net::FTP docs.  (see the
> original post).  It turns out that the firewall on the server was
> broken.  The problem has been resolved.  However, we still don't know
> why ncftp worked and Net::FTP didn't.
>
> thanks for the help,
>   Jeremy
>
> On Wed, 2004-09-01 at 10:04, Wayne Walker wrote:
>> On Wed, Sep 01, 2004 at 08:40:16AM -0500, Jeremy Brooks wrote:
>> > Well,  the problem was zone-alarm running on the server.
>> > That begs the question, why did ncftp work just fine?
>>
>> Probably ncftp defaults to passive mode ftp and Net::FTP defaults to
>> active (or vice versa, I can never get the modes straight).  One is
>> considered a serious security problem, the other a possible denial of
>> service attack.
>
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin



From dbii at interaction.net  Fri Sep 10 08:41:04 2004
From: dbii at interaction.net (David Bluestein II)
Date: Fri Sep 10 08:41:07 2004
Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?
Message-ID: <20040910134104.GN1171@interaction.net>

Next week we're scheduled for our monthly meeting, and Ian has been great in reserving the room, projector, and internet access for us up at ARL (see website for directions).

At the last meeting, we tossed around some ideas, but didn't have any definite discussion. Does anyone have something they want to present on? Or heard presented too?

If not, I can present on FormBuilder, Data::Grouper and SQL::Abstract, which allows you to create a webform, tie to database, and display information using 3-4 modules without having to do a lot of code. Basically, create a form definition (including self-validation) and then you can easily insert into a database or retrieve data from it in tabular form.

David

-- 
David H. Bluestein II
President & Lead Developer        dbii@interaction.net
ii, inc.                          http://www.interaction.net
 -- Specializing in Designing Interactive Database Driven Websites -- 
From jbrooks4 at austin.rr.com  Fri Sep 10 08:57:41 2004
From: jbrooks4 at austin.rr.com (Jeremy Brooks)
Date: Fri Sep 10 08:56:01 2004
Subject: APM: RE: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <20040910134104.GN1171@interaction.net>
Message-ID: <004201c4973e$24096ae0$641ea8c0@THEDOGHOUSE.HMFIC>

I'd like to learn about SQL::Abstract.  By the way, did I miss the YAML
presentation?

-Jeremy Brooks

-----Original Message-----
From: austin-bounces@mail.pm.org [mailto:austin-bounces@mail.pm.org]On
Behalf Of David Bluestein II
Sent: Friday, September 10, 2004 8:41 AM
To: austin@mail.pm.org
Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?


Next week we're scheduled for our monthly meeting, and Ian has been great in
reserving the room, projector, and internet access for us up at ARL (see
website for directions).

At the last meeting, we tossed around some ideas, but didn't have any
definite discussion. Does anyone have something they want to present on? Or
heard presented too?

If not, I can present on FormBuilder, Data::Grouper and SQL::Abstract, which
allows you to create a webform, tie to database, and display information
using 3-4 modules without having to do a lot of code. Basically, create a
form definition (including self-validation) and then you can easily insert
into a database or retrieve data from it in tabular form.

David

--
David H. Bluestein II
President & Lead Developer        dbii@interaction.net
ii, inc.                          http://www.interaction.net
 -- Specializing in Designing Interactive Database Driven Websites --
_______________________________________________
Austin mailing list
Austin@mail.pm.org
http://mail.pm.org/mailman/listinfo/austin




Protected by Polesoft Lockspam
http://www.polesoft.com/refer.html

From dbii at interaction.net  Fri Sep 10 08:59:44 2004
From: dbii at interaction.net (David Bluestein II)
Date: Fri Sep 10 08:59:47 2004
Subject: APM: Re: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <004201c4973e$24096ae0$641ea8c0@THEDOGHOUSE.HMFIC>
References: <20040910134104.GN1171@interaction.net>
	<004201c4973e$24096ae0$641ea8c0@THEDOGHOUSE.HMFIC>
Message-ID: <20040910135944.GP1171@interaction.net>

No, I forget who was going to do YAML. I think they wanted to do it in October or November to have time to work with it some.

David

On Fri, Sep 10, 2004 at 08:57:41AM -0500, Jeremy Brooks wrote:
> I'd like to learn about SQL::Abstract.  By the way, did I miss the YAML
> presentation?
> 
> -Jeremy Brooks
> 
> -----Original Message-----
> From: austin-bounces@mail.pm.org [mailto:austin-bounces@mail.pm.org]On
> Behalf Of David Bluestein II
> Sent: Friday, September 10, 2004 8:41 AM
> To: austin@mail.pm.org
> Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?
> 
> 
> Next week we're scheduled for our monthly meeting, and Ian has been great in
> reserving the room, projector, and internet access for us up at ARL (see
> website for directions).
> 
> At the last meeting, we tossed around some ideas, but didn't have any
> definite discussion. Does anyone have something they want to present on? Or
> heard presented too?
> 
> If not, I can present on FormBuilder, Data::Grouper and SQL::Abstract, which
> allows you to create a webform, tie to database, and display information
> using 3-4 modules without having to do a lot of code. Basically, create a
> form definition (including self-validation) and then you can easily insert
> into a database or retrieve data from it in tabular form.
> 
> David
> 
> --
> David H. Bluestein II
> President & Lead Developer        dbii@interaction.net
> ii, inc.                          http://www.interaction.net
>  -- Specializing in Designing Interactive Database Driven Websites --
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin
> 
> 
> 
> 
> Protected by Polesoft Lockspam
> http://www.polesoft.com/refer.html

-- 
David H. Bluestein II
President & Lead Developer        dbii@interaction.net
ii, inc.                          http://www.interaction.net
 -- Specializing in Designing Interactive Database Driven Websites -- 
From mlehmann at marklehmann.com  Fri Sep 10 10:04:30 2004
From: mlehmann at marklehmann.com (Mark Lehmann)
Date: Fri Sep 10 10:00:31 2004
Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <20040910134104.GN1171@interaction.net>
References: <20040910134104.GN1171@interaction.net>
Message-ID: <44842.66.45.70.241.1094828670.squirrel@www.marklehmann.com>

Spam Assassin.

Really I just want to present on something were I can legitimately say
"ass" twice without being impolite.
--
Mark Lehmannn
mlehmann@marklehmann.com - mobile 512 689-7705

<quote who="David Bluestein II">
> Next week we're scheduled for our monthly meeting, and Ian has been
> great in reserving the room, projector, and internet access for us up at
> ARL (see website for directions).
>
> At the last meeting, we tossed around some ideas, but didn't have any
> definite discussion. Does anyone have something they want to present on?
> Or heard presented too?
>
> If not, I can present on FormBuilder, Data::Grouper and SQL::Abstract,
> which allows you to create a webform, tie to database, and display
> information using 3-4 modules without having to do a lot of code.
> Basically, create a form definition (including self-validation) and then
> you can easily insert into a database or retrieve data from it in
> tabular form.
>
> David
>
> --
> David H. Bluestein II
> President & Lead Developer        dbii@interaction.net
> ii, inc.                          http://www.interaction.net
>  -- Specializing in Designing Interactive Database Driven Websites --
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin



From parkerm at pobox.com  Fri Sep 10 10:28:19 2004
From: parkerm at pobox.com (Michael Parker)
Date: Fri Sep 10 10:28:24 2004
Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <44842.66.45.70.241.1094828670.squirrel@www.marklehmann.com>
References: <20040910134104.GN1171@interaction.net>
	<44842.66.45.70.241.1094828670.squirrel@www.marklehmann.com>
Message-ID: <20040910152818.GA16128@mail.herk.net>

On Fri, Sep 10, 2004 at 10:04:30AM -0500, Mark Lehmann wrote:
> Spam Assassin.
> 

What kind of a talk on SpamAssassin would people like to have?
Something general?  Something about the upcoming 3.0.0 release?

I'm presenting a talk at ApacheCon in November on "Storing
SpamAssassin User Data in SQL Databases," it might be a good time for
me to practice that talk.  It will mostly cover the new Bayes/AWL data
in SQL functionality that is being added for the 3.0.0 release but
also putting user prefs in SQL.

It would have to be October, I'm not done with the talk yet.

Michael
SpamAssassin Developer
From ian at remmler.org  Fri Sep 10 10:41:03 2004
From: ian at remmler.org (Ian Remmler)
Date: Fri Sep 10 10:41:08 2004
Subject: APM: Re: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <20040910135944.GP1171@interaction.net>
References: <20040910134104.GN1171@interaction.net>
	<004201c4973e$24096ae0$641ea8c0@THEDOGHOUSE.HMFIC>
	<20040910135944.GP1171@interaction.net>
Message-ID: <20040910154103.GA4740@remmler.org>

On Fri, Sep 10, 2004 at 08:59:44AM -0500, David Bluestein II wrote:
> No, I forget who was going to do YAML. I think they wanted to
> do it in October or November to have time to work with it
> some.
> 
> On Fri, Sep 10, 2004 at 08:57:41AM -0500, Jeremy Brooks wrote:
> > I'd like to learn about SQL::Abstract.  By the way, did I miss the YAML
> > presentation?

I was the YAML volunteer.  If we don't have a full lineup for
Wednesday I could talk a bit about YAML.  There's really not
much to it (which is a good thing), so it wouldn't take too
long.  Or I could do it next month.

I also started looking at Maypole, the web app framework.  I
could probably talk about it in the not-so-distant furure (i.e.
not this meeting :)

-- 
Go banana!
From dbii at interaction.net  Fri Sep 10 10:56:49 2004
From: dbii at interaction.net (David Bluestein II)
Date: Fri Sep 10 10:56:52 2004
Subject: APM: Meeting Topics Schedule--DRAFT
Message-ID: <20040910155649.GQ1171@interaction.net>

Wow, lots of topics. How about the following (if good with Michael and Ian--thanks for the volunteering):

Sept: Ian on YAML, David on SQL::Abstract et al.
Oct: Michael on Spamassassin (we'll need to see if we let Mark come to this one)

Nov: Ian on Maypole (that gives you two months)
Dec: Christmas Dinner
Jan: TBD

Any other topics? Wayne's boss (I forget his name) wanted a presentation on the Windows .exe maker, which is probably a 15-30 minute talk. Maybe we can add that on somewhere?

David

-- 
David H. Bluestein II
President & Lead Developer        dbii@interaction.net
ii, inc.                          http://www.interaction.net
 -- Specializing in Designing Interactive Database Driven Websites -- 
From chris at chrisbaker.net  Fri Sep 10 22:30:11 2004
From: chris at chrisbaker.net (Christopher Baker)
Date: Fri Sep 10 22:30:05 2004
Subject: APM: Next Weeks Meeting Topic-Suggestion and Volunteers?
In-Reply-To: <20040910152818.GA16128@mail.herk.net>
References: <44842.66.45.70.241.1094828670.squirrel@www.marklehmann.com>
	<20040910152818.GA16128@mail.herk.net>
Message-ID: <20040910222736.4A65.CHRIS@chrisbaker.net>

When and where is the meeting?

Chris

============================================================
"I'm not gonna go up onstage and dress different.
Britney Spears goes up onstage and dresses like a
showgirl. She's not being herself up there because
she's dancing like a ho." --Avril Lavigne
============================================================
Chris Baker -- www.chrisbaker.net
chris@chrisbaker.net, chrisbaker@iname.com, cbaker2@columbus.rr.com
"When you stop growing, you start dying."

From dbii at interaction.net  Mon Sep 13 08:38:13 2004
From: dbii at interaction.net (David Bluestein II)
Date: Mon Sep 13 08:38:15 2004
Subject: APM: Meeting Annoucement: YAML and SQL::Abstract
Message-ID: <20040913133813.GE8929@interaction.net>

The APM Meeting this month is coming up this Wednesday. Ian is going to cover YAML, and I'm going to cover SQL::Abstract (and a few related modules like FormBuilder that help create quick database calls and displays).

Meeting will be at 7:00 pm at ARL, those who wish can join us at Pok-e Jo's in the Arboretum at 5:45 for dinner.

David

-- 
David H. Bluestein II
President & Lead Developer        dbii@interaction.net
ii, inc.                          http://www.interaction.net
 -- Specializing in Designing Interactive Database Driven Websites -- 
From tom.bakken at tx.usda.gov  Fri Sep 17 11:06:51 2004
From: tom.bakken at tx.usda.gov (Bakken, Tom - Temple, TX)
Date: Fri Sep 17 11:06:54 2004
Subject: APM: Process/System questions
Message-ID: <DD06C4236551DD4994ADA36ECAEC5EF83E3A27@COFORTCOL2S301.agwest.one.usda.gov>

Some time ago I posted a problem I was having with a script that was
supposed to restart workstations but stalled when it hit certain PCs.  I
used Bill Raty's suggestion of the spawn subroutine.   I wasn't able to
get it to work and put it aside until I recently had more time and now
seem to have gotten it running.

Shutdown.exe has a known bug where if the workstation that's to be shut
down is at the Windows login screen, it will lock up.  Bill's solution
works around that problem by waiting 20 seconds and then continuing if
the workstation stalls.  Unfortunately, the system command returns a 0
exit status for success even if shutdown.exe stalls.  Is there any way
to capture the output from shutdown.exe which is "the device is not
ready"?  Either that or if the script does wait 20 seconds and fails,
can I test for that somehow?

I appreciate the help.

#############
#!perl -w
# Purpose:	Create a list of workstations, restart them and store
results in log file
# Notes:	Portablilty: shutdown switch to abort programs is /C on
Windows 2000 and /F on XP	

use strict;
use Text::ParseWords;		# Help for comma separated data
use POSIX 'strftime';		# Load function
use Time::Local;

# Create list of servers to analyze:

my $LogFile = 'WorkstationRestartLog.txt';
unless (open LOGFILE, "> $LogFile") {
  die "Cannot open $LogFile: $!";
}

my $LocalServer = 'MyServer';
my $Domain = 'MyDomain';
my $ServerFile = 'c:/bin/reboot/csv.log';
my $StateAbbrev = 'TX';
my @ListOfWorkstations = ();
my @MachineNameAndDomain = ();

GetWorkstations(\@ListOfWorkstations, $Domain, $LocalServer,
$ServerFile, $StateAbbrev);
my $Status;  

WorkstationRestart(\@ListOfWorkstations);		# Restart the
list of workstations

######################   Subroutines
########################################
sub GetHumanTime {
  my $TimeStamp = strftime("%m/%d/%Y %I:%M %p", localtime(time));
  return $TimeStamp;
}
sub spawn (&) {
  my ($CodeRef) = @_;
  my $pid;
  unless ($pid = fork) {
    $CodeRef->();		# Start the child process.  Make sure it
exits too!
    exit;
  }
  $SIG{CHLD} = "IGNORE";	# Make sure we don't have any ghost
processes.  Have perl do the waitpid
  return $pid;
}
sub GetWorkstations {
  my $ListOfWorkstationsArrayRef = shift;
  my $Domain = shift;
  my $LocalServer = shift;
  my $ServerFile = shift;
  my $StateAbbrev = shift;
  my $Status = system "csvde -f $ServerFile \
               -s $LocalServer.$Domain.one.usda.gov \
               -d \"ou=$StateAbbrev, dc=$Domain, dc=one, dc=usda,
dc=gov\" \
               -p subtree \
	       -r \"(&(Objectcategory=Computer)(objectClass=computer))\"
	       -l \"name, dNSHostName, description, operatingsystem,
whencreated, whenchanged\"";
  warn "Active Directory query failed: $!" unless $Status == 0;

  unless (open WORKSTATIONS, "$ServerFile") {
    die "Cannot open $ServerFile: $!";
  }
  while(<WORKSTATIONS>) {
    chomp;			# Remove ctrl+M
    s/\000//g;			# Remove null padding
    my @Fields = parse_csv($_);
    my ($DN, $Description, $DnsHostName, $OS, $Name, $WhenChanged,
$WhenCreated) = @Fields;
##print "$Name = \$Name\n";
    push @{$ListOfWorkstationsArrayRef}, join '|', $Name, $Description,
$DnsHostName,
           $OS, $WhenChanged, $WhenCreated;
  }
}			# End GetWorkstationUptime subroutine
sub WorkstationRestart {	# Copy files to server using a list of
machine names 
  my $MachineNameAndDomainArrayRef = shift;
  my $Count = 0;
  my $Error = 0;
  my $StartTime = GetHumanTime();
  print LOGFILE "                  Start: Computer Restart Log:
$StartTime\n\n";
  my $i = ();
  foreach $i (sort @{$MachineNameAndDomainArrayRef}) {
    unless (open LOGFILE, ">> $LogFile") {		# Close write
filehandle and open append
      die "Cannot open $LogFile: $!";
    }
    $Count++;
    my ($MachineName, $Domain) = split /\|/, $i;
#print "\$MachineName = $MachineName\n";
    next unless $MachineName;				# Skip blank
records
    print "$Count:	Attempting Reboot on: $MachineName\n";
    $Status = system "uptime \\\\$MachineName > /nul";
    unless ($Status == 0) {
      print LOGFILE "$Count	ALERT!!! Could not contact $MachineName.
Error: $?\n";
      print "$Count   ALERT!!! Could not contact $MachineName. Error:
$?\n";
      $Error++;
      next;
    }
print "Parent PID: $$\n";
#########    
    my $ChildPid = spawn {
     $Status = system "shutdown.exe -m \\\\$MachineName -F    -R -T 90";
# Windows XP
###     $Status = system "shutdown.exe \\\\$MachineName /C /Y /R /T:90";
# Windows 2K
###      $Status = system "uptime \\\\$MachineName > /nul";
      unless ($Status == 0) {
        print LOGFILE "$Count   ALERT!!! Could not restart $MachineName.
Error: $?\n";
        print  "$Count   ALERT!!! Could not restart $MachineName. Error:
$?\n";
        $Error++;
        next; 
      }
      my $TimeStamp = GetHumanTime();
      print LOGFILE "$Count	$MachineName successfully rebooted at
$TimeStamp\n";
      print "$Count	$MachineName successfully rebooted at
$TimeStamp\n";
    };				# Don't forget the semicolon here!!!!
###########
    sleep 20;			# Give uptime 20 seconds to run
    print "Child pid: $ChildPid\n";
    kill TERM => $ChildPid;	# Kill after that amount of time
  }		# End server loop
  my $EndTime = GetHumanTime();
  print LOGFILE "\n                   $Count Computer restarts
attempted.  $Error Failed.\n\n" . 
                "                   End: Computer Restart Log:
$EndTime\n" .
                '=' x 50 . "\n\n";
}				# End WorkstationRestart subroutine
sub parse_csv {
  return quotewords(",", 0, $_[0]);
}
###################

Tom Bakken

From ezra at jdba.org  Fri Sep 17 11:42:39 2004
From: ezra at jdba.org (ezra pagel)
Date: Fri Sep 17 11:42:53 2004
Subject: APM: Process/System questions
In-Reply-To: <DD06C4236551DD4994ADA36ECAEC5EF83E3A27@COFORTCOL2S301.agwest.one.usda.gov>
References: <DD06C4236551DD4994ADA36ECAEC5EF83E3A27@COFORTCOL2S301.agwest.one.usda.gov>
Message-ID: <414B13FF.4020000@jdba.org>

Using backticks/qx instead of system() will let you capture standard 
output; you can redirect stderr if you need it. You can still check $? 
for the system exit code.

Bakken, Tom - Temple, TX wrote:

>Some time ago I posted a problem I was having with a script that was
>supposed to restart workstations but stalled when it hit certain PCs.  I
>used Bill Raty's suggestion of the spawn subroutine.   I wasn't able to
>get it to work and put it aside until I recently had more time and now
>seem to have gotten it running.
>
>Shutdown.exe has a known bug where if the workstation that's to be shut
>down is at the Windows login screen, it will lock up.  Bill's solution
>works around that problem by waiting 20 seconds and then continuing if
>the workstation stalls.  Unfortunately, the system command returns a 0
>exit status for success even if shutdown.exe stalls.  Is there any way
>to capture the output from shutdown.exe which is "the device is not
>ready"?  Either that or if the script does wait 20 seconds and fails,
>can I test for that somehow?
>  
>
<snip>

-- 

-------------------------------------------------------
Ezra Pagel, Software Engineer
Austin Logistics, Inc.
Office: 512-651-5644
Mobile: 512-694-1107
epagel@austinlogistics.com
ezra@cpan.org
-------------------------------------------------------
The information in this Internet email is confidential and may be 
legally privileged. It is intended solely for the addressee. Access to 
this Internet email by anyone else is unauthorized. If you are not the 
intended recipient, any disclosure, copying, distribution or any action 
taken or any action omitted in reliance on it, is prohibited and may be 
unlawful. When addressed to our clients any opinions or advice contained 
in this Internet email are subject to the terms and conditions expressed 
in any applicable governing Austin Logistics' contract terms.

From jbrooks4 at austin.rr.com  Fri Sep 17 12:50:33 2004
From: jbrooks4 at austin.rr.com (Jeremy Brooks)
Date: Fri Sep 17 12:38:23 2004
Subject: APM: RPC on windows
Message-ID: <1095443433.12859.26.camel@willow.mobile-o.com>

Tom Bakken's last post got me thinking about the possibilities of remote
administration of Windows machines.  To what extent can 
one administer Windows machines from a remote client, be it Linux,Unix
or Windows?  Is it possible, assuming the correct
permissions/credentials are in place, to do remote installation of
software or file uploads?  Perlifying something akin to Symantec
Corporate Edition's ultra-nifty remote client installation functionality
in my own software would be very handy! 

I'm not very familiar with Perl/CPAN code for Windows but I'm an
experienced Perlite so simply pointing me in the right direction would
be helpful enough!

thanks,
   Jeremy Brooks

From jbrooks4 at austin.rr.com  Tue Sep 21 10:40:52 2004
From: jbrooks4 at austin.rr.com (Jeremy Brooks)
Date: Tue Sep 21 10:28:33 2004
Subject: APM: RE: RPC on windows
In-Reply-To: <297203B1AD107A438D2616D3A8E9B382E0FCAB@dsae2k.dsad.ad.tamu.edu>
References: <297203B1AD107A438D2616D3A8E9B382E0FCAB@dsae2k.dsad.ad.tamu.edu>
Message-ID: <1095781252.9707.12.camel@willow.mobile-o.com>

thanks,  I'll check his site out.
Yes, I have some modules you'd probably be interested in.
My company (www.mobile-o.com) developed a set of MS Exchange interaction
modules that use WebDAV to bring Outlook like functionality to Perl. 
They've only been tested on Linux/Perl 5.8 but with Exchange 2000 and
2003.  I've written a website for hand held devices that uses these
modules web email access that brings pop and Exchange accounts
together.  I plan on getting the package onto CPAN toward the end of the
year, or sooner.  The package includes modifications to HTTP::DAV and
several MS Exchange specific modules that currently reside in
Mobileo::Exchange.  The location of the latter will need to be changed
and may end up in HTTP::Exchange,DAV::Exchange, or somewhere like that. 
If you'd like to work on testing and/or adding to them I'd be willing to
upload them this month. 

-Jeremy Brooks

On Mon, 2004-09-20 at 10:49, Fluhmann, Jeremy wrote:
> Jeremy,
> 
> All of my Perl experience is on Windows.  Dave Roth has a good
> Perl/Win32 site with several scripts available for download.  He has
> also writen a couple of books on the subject.  I've used his modules in
> some of my Windows Adminstration scripts.
> 
> http://www.roth.net/
> 
> I used some of his modules on my 'new web account' web page for work.  I
> input a new username and description, and the user account, user's
> folder (with specific permissions), virtual directory, and the FTP
> account are all created if they don't already exist.  I'm still looking
> into remote installation and update possibilities.
> 
> Since all of my Perl experience is in a Windows environment, I'm very
> interested in the subject as well as extending my knowledge.  So, please
> let me know if you come across some good places or have some good ideas
> about Perl and Windows.
> 
> Thanks,
> Jeremy Fluhmann
> Microcomputer Specialist
> Department of Student Activities
> Texas A&M University
> 979-458-4649
> jeremy@stuact.tamu.edu
> 
> 
> -----Original Message-----
> From: austin-bounces@mail.pm.org [mailto:austin-bounces@mail.pm.org] On
> Behalf Of Jeremy Brooks
> Sent: Friday, September 17, 2004 12:51 PM
> To: apm
> Subject: APM: RPC on windows
> 
> Tom Bakken's last post got me thinking about the possibilities of remote
> administration of Windows machines.  To what extent can one administer
> Windows machines from a remote client, be it Linux,Unix or Windows?  Is
> it possible, assuming the correct permissions/credentials are in place,
> to do remote installation of software or file uploads?  Perlifying
> something akin to Symantec Corporate Edition's ultra-nifty remote client
> installation functionality in my own software would be very handy! 
> 
> I'm not very familiar with Perl/CPAN code for Windows but I'm an
> experienced Perlite so simply pointing me in the right direction would
> be helpful enough!
> 
> thanks,
>    Jeremy Brooks
> 
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin
> 
> 

From michalk at awpi.com  Tue Sep 21 12:52:12 2004
From: michalk at awpi.com (Brian Michalk)
Date: Tue Sep 21 12:52:38 2004
Subject: APM: arrays and hashes
In-Reply-To: <1095443433.12859.26.camel@willow.mobile-o.com>
Message-ID: <NGBBLKIBLMDNILPAICAJEENKGDAA.michalk@awpi.com>

I'm having problems making a hash reference to a 2D array.

I would like to get the following output:

  DB<40> x @res
0  'hello'
1  'there'
2  'world'

  DB<45> x $self->{'ary'}
0  ARRAY(0x8cfd1f0)
   0  'hello'
   1  'there'
   2  'world'
1  ARRAY(0x8cf2ab0)
   0  'hello'
   1  'there'
   2  'world'
2  ARRAY(0x8cd385c)
   0  'hello'
   1  'there'
   2  'world'

One would think one could get it like this:
  DB<43> push @{$self->{'ary'}}, [@res]
  DB<43> push @{$self->{'ary'}}, [@res]
  DB<43> push @{$self->{'ary'}}, [@res]


However, the above generates the following:
  DB<52> x $self->{'ary'}              
0  ARRAY(0x8cfc4e8)
   0  ARRAY(0x8cfc494)
      0  'hello'
      1  'there'
      2  'world'
   1  ARRAY(0x8cfce30)
      0  'hello'
      1  'there'
      2  'world'
   2  ARRAY(0x8cfd0d0)
      0  'hello'
      1  'there'
      2  'world'

Which is a 3D array, and I can't figure out how to get it to work correctly.
From ezra at jdba.org  Tue Sep 21 13:18:13 2004
From: ezra at jdba.org (ezra pagel)
Date: Tue Sep 21 13:18:44 2004
Subject: APM: arrays and hashes
In-Reply-To: <NGBBLKIBLMDNILPAICAJEENKGDAA.michalk@awpi.com>
References: <NGBBLKIBLMDNILPAICAJEENKGDAA.michalk@awpi.com>
Message-ID: <41507065.1030505@jdba.org>

Brian Michalk wrote:

>I'm having problems making a hash reference to a 2D array.
>  
>
Sorry, no such thing ;)

>I would like to get the following output:
>
>  DB<40> x @res
>0  'hello'
>1  'there'
>2  'world'
>
>  DB<45> x $self->{'ary'}
>0  ARRAY(0x8cfd1f0)
>   0  'hello'
>   1  'there'
>   2  'world'
>1  ARRAY(0x8cf2ab0)
>   0  'hello'
>   1  'there'
>   2  'world'
>2  ARRAY(0x8cd385c)
>   0  'hello'
>   1  'there'
>   2  'world'
>
>One would think one could get it like this:
>  DB<43> push @{$self->{'ary'}}, [@res]
>  DB<43> push @{$self->{'ary'}}, [@res]
>  DB<43> push @{$self->{'ary'}}, [@res]
>
>
>However, the above generates the following:
>  DB<52> x $self->{'ary'}              
>0  ARRAY(0x8cfc4e8)
>   0  ARRAY(0x8cfc494)
>      0  'hello'
>      1  'there'
>      2  'world'
>   1  ARRAY(0x8cfce30)
>      0  'hello'
>      1  'there'
>      2  'world'
>   2  ARRAY(0x8cfd0d0)
>      0  'hello'
>      1  'there'
>      2  'world'
>
>Which is a 3D array, and I can't figure out how to get it to work correctly.
>_______________________________________________
>Austin mailing list
>Austin@mail.pm.org
>http://mail.pm.org/mailman/listinfo/austin
>  
>

Actually, that's still just a 2D array; I think that the debugger lines 
are throwing you off. Try x $self->{'ary'}->[0][0]
From austin.pm at sam-i-am.com  Wed Sep 22 13:01:20 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Wed Sep 22 13:01:30 2004
Subject: APM: Cookie-less cgi sessions
Message-ID: <4151BDF0.3090501@sam-i-am.com>

Scanning CGI::Session's perldocs, it looks to rely on http cookies. Is 
there another module anyone can recommend that gives you options (e.g. 
adding the session id to the querystring, or a hidden field) or better 
yet will pick the appropriate or available mechanism invisibly?

Sorry my searching has not been exhaustive - I just figured this might 
be a easy one for someone out there to get me on the right track.

Sam
From austin.pm at sam-i-am.com  Wed Sep 22 13:16:29 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Wed Sep 22 13:16:32 2004
Subject: APM: YAML url mappings
Message-ID: <4151C17D.6020609@sam-i-am.com>

So, after this month's very intersting YAML presentation, I thought I'd 
give it a whirl with a little 404 handler I'm making. It needs to 
attempt to lookup a (bad) url, or fragment of the url in a simple old => 
new hash structure, eg
{
   '/some/old/url.html' 	=> '/new/location/for/url.html',
   '/old/dir'		=> '/new/dir'
}
Having a nice (safe) human editable datafile to store this would be 
ideal.. a nice spot for YAML I thought.
Except the obvious doesn't work:
---
/some/old/url.html: /new/location/for/url.html
/old/dir: /new/dir

it doesn't like '/' as a starting character (or '~,' or '.'. Is it 
matching with something like /\s*\w+/ ?.. must take a look.)

I looked through the YAML perldocs, and the spec and found nothing about 
illegal or allowed characters that might explain this. Does that make it 
a bug? Anyone have any thoughts, workarounds or suggestions?

Right now, I've prepended 'ROOT' so I get
---
ROOT/some/old/url.html: ROOT/new/location/for/url.html

which works just fine - and I strip it back off before I return the url 
to the user... But kind of dirties my data a little. The goal is to keep 
the data format clean and intuitive, and leave any complexity in the 
script. I can obviously write my own "parser" or use Config::Auto or 
something else.. but I've just been telling everyone at work all about 
YAML and need to not look silly :)

thanks,
Sam





From taylor at codecafe.com  Wed Sep 22 13:54:13 2004
From: taylor at codecafe.com (Taylor Carpenter)
Date: Wed Sep 22 13:54:23 2004
Subject: APM: Cookie-less cgi sessions
In-Reply-To: <4151BDF0.3090501@sam-i-am.com>
References: <4151BDF0.3090501@sam-i-am.com>
Message-ID: <1095879253.20375.204936116@webmail.messagingengine.com>


On Wed, 22 Sep 2004 13:01:20 -0500, "Sam Foster"
<austin.pm@sam-i-am.com> said:
> Scanning CGI::Session's perldocs, it looks to rely on http cookies. Is 
> there another module anyone can recommend that gives you options (e.g. 
> adding the session id to the querystring, or a hidden field) or better 
> yet will pick the appropriate or available mechanism invisibly?
> 
> Sorry my searching has not been exhaustive - I just figured this might 
> be a easy one for someone out there to get me on the right track.


Hello Sam,

CGI::Session also supports using query string to pass the session ID. 
It will do this automatically.

>From the page

http://search.cpan.org/~sherzodr/CGI-Session-3.95/Session/Tutorial.pm#WHAT_YOU_NEED_TO_KNOW_FIRST

It discusses the second argument to CGI::Session.

"In this case, the library will try to retrieve the session id from
either CGISESSID cookie or CGISESSID CGI parameter (query string)"

Hope this helps.

On a side note... It is very easy to extend CGI::Session for storage
methods and session ID generation.

Taylor
From ian at remmler.org  Wed Sep 22 15:09:45 2004
From: ian at remmler.org (Ian Remmler)
Date: Wed Sep 22 15:09:52 2004
Subject: APM: YAML url mappings
In-Reply-To: <4151C17D.6020609@sam-i-am.com>
References: <4151C17D.6020609@sam-i-am.com>
Message-ID: <20040922200945.GA7139@remmler.org>

On Wed, Sep 22, 2004 at 01:16:29PM -0500, Sam Foster wrote:
> it doesn't like '/' as a starting character (or '~,' or '.'. Is it 
> matching with something like /\s*\w+/ ?.. must take a look.)
> 
> I looked through the YAML perldocs, and the spec and found nothing about 
> illegal or allowed characters that might explain this. Does that make it 
> a bug? Anyone have any thoughts, workarounds or suggestions?

I just installed YAML::Parser::Syck, and it worked just fine
with that.  I'm guessing it's a bug in YAML.pm, but I haven't
looked at it to find it.  I think the only special characters at
the start of an entry are !, &, and *.  YAML.pm does work if you
quote, e.g.

---
'/old/dir': '/new/dir'
...

	- Ian.

-- 
Go banana!
From tom.bakken at tx.usda.gov  Thu Sep 23 09:21:17 2004
From: tom.bakken at tx.usda.gov (Bakken, Tom - Temple, TX)
Date: Thu Sep 23 09:21:23 2004
Subject: APM: Process/System questions
Message-ID: <DD06C4236551DD4994ADA36ECAEC5EF83E3A41@COFORTCOL2S301.agwest.one.usda.gov>

That did the trick.  The system command would wait until shutdown.exe
gracefully exited.  The bug in windows prevented it from doing that, so
the entire program would stall.  Using back ticks doesn't wait for a
graceful exit worked great.  

Thanks for the help.

Tom Bakken
Information Resource Manager
Texas USDA, Rural Development 

-----Original Message-----
From: ezra pagel [mailto:ezra@jdba.org] 
Sent: Friday, September 17, 2004 11:43 AM
To: Bakken, Tom - Temple, TX
Cc: austin@mail.pm.org
Subject: Re: APM: Process/System questions


Using backticks/qx instead of system() will let you capture standard 
output; you can redirect stderr if you need it. You can still check $? 
for the system exit code.

Bakken, Tom - Temple, TX wrote:

>Some time ago I posted a problem I was having with a script that was
>supposed to restart workstations but stalled when it hit certain PCs.
I
>used Bill Raty's suggestion of the spawn subroutine.   I wasn't able to
>get it to work and put it aside until I recently had more time and now
>seem to have gotten it running.
>
>Shutdown.exe has a known bug where if the workstation that's to be shut
>down is at the Windows login screen, it will lock up.  Bill's solution
>works around that problem by waiting 20 seconds and then continuing if
>the workstation stalls.  Unfortunately, the system command returns a 0
>exit status for success even if shutdown.exe stalls.  Is there any way
>to capture the output from shutdown.exe which is "the device is not
>ready"?  Either that or if the script does wait 20 seconds and fails,
>can I test for that somehow?
>  
>
<snip>

-- 

-------------------------------------------------------
Ezra Pagel, Software Engineer
Austin Logistics, Inc.
Office: 512-651-5644
Mobile: 512-694-1107
epagel@austinlogistics.com
ezra@cpan.org
-------------------------------------------------------
The information in this Internet email is confidential and may be 
legally privileged. It is intended solely for the addressee. Access to 
this Internet email by anyone else is unauthorized. If you are not the 
intended recipient, any disclosure, copying, distribution or any action 
taken or any action omitted in reliance on it, is prohibited and may be 
unlawful. When addressed to our clients any opinions or advice contained

in this Internet email are subject to the terms and conditions expressed

in any applicable governing Austin Logistics' contract terms.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Tom.Bakken@tx.usda.gov.vcf
Type: text/x-vcard
Size: 460 bytes
Desc: Tom.Bakken@tx.usda.gov.vcf
Url : http://mail.pm.org/pipermail/austin/attachments/20040923/67ed5516/Tom.Bakkentx.usda.gov.vcf
From austin.pm at sam-i-am.com  Tue Sep 28 11:00:23 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Tue Sep 28 11:00:33 2004
Subject: APM: installing modules on a virtual server
Message-ID: <41598A97.1070106@sam-i-am.com>

Got a problem getting some modules onto a virtual server - this is a 
client's basic shared hosting package.

I have ssh access.
I can run perl -MCPAN -e shell;
   During the config setup, I'm adding PREFIX=~/perl5lib
I can install stuff there just fine, but a lot of the modules obviously 
have dependancies.. and as ~/perl5lib isn't in @INC, I cant complete the 
tests or installation of the modules I need.

Just for kicks I tried adding BEGIN { unshift @INC, '~/perl5lib' } to 
the top of the .cpan/CPAN/MyConfig.pm .. but I guess that falls out of 
scope by the time it comes to running the make tests

Apparently I should be able to do:
% setenv PERL5LIB ~/perl5lib
but setenv is also apparently not available.

I'm putting in a tech support call on this... any thoughts or alternatives?
It seems like when I run perl -MCPAN -e shell, I should be able to add 
the directory to @INC at that point.. But i'm not clear - looking at the 
perldocs - how I would do that.

TIA
Sam
From austin.pm at sam-i-am.com  Tue Sep 28 13:37:01 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Tue Sep 28 13:37:35 2004
Subject: APM: installing modules on a virtual server
In-Reply-To: <1096389006.8500.205331481@webmail.messagingengine.com>
References: <41598A97.1070106@sam-i-am.com>
	<1096389006.8500.205331481@webmail.messagingengine.com>
Message-ID: <4159AF4D.8050202@sam-i-am.com>

(excuse the top posting.. I did trim a little honest)
I did some searching yesterday and came up with this:
http://www.bluereef.net/support/extensions/development/perlmodown.html

I'm using bash.
	setenv PERL5LIB ~/perl
gives me the error that setenv is not available.

	export PERL5LIB=~/perl
gave me nothing.
I also tried
	perl -MCPAN -Mlib -e "unshift(@INC, '~/perl5lib');print join(\"\n\", 
@INC); shell;"

and this worked in that it printed out @INC and showed it to include 
~/perl5lib.. But when I went to cpan> test , it was still not picking up 
the dependancies even after I let it install them again.
E.g. to install File::Slurp, it runs tests which want Test::More. 
Test::More wants Test::Harness. Each time it spots the dependancy, 
installs the module, but the test still fails as it cant find it (in 
~/perl5lib)
This is a trivial example, and I could probably make it skip those 
tests, but with others (e.g XML::LibXML) it becomes more of a hassle.

Which brings me back to setenv PERL5LIB.. still no word back from tech 
support on that one (they said they were getting a technician to look at 
it.. and there's me thinking I'd called technical support :)

Sam


Taylor Carpenter wrote:
> On Tue, 28 Sep 2004 11:00:23 -0500, "Sam Foster"
> <austin.pm@sam-i-am.com> said:
> 
>>Apparently I should be able to do:
>>% setenv PERL5LIB ~/perl5lib
>>but setenv is also apparently not available.
> 
> Did you try export PERL5LIB (after doing PERL5LIB=~/perl5lib)?  For bash
> you can do 
>    export PERL5LIB=~/perl5lib
> but for some shells you have to first set then export the variable. 
> Be sure to add
>    use lib '/path/to/my/perl5lib'; 
> to your scripts.
> 
> For CPAN you want the following set when configuring CPAN
> 
>    makepl_arg: PREFIX=~/perl5lib/
>    cpan_home: /path/to/my/perl5lib/ANY_NAME_YOU_WANT
> 
From cbl_mbi1 at yahoo.com  Tue Sep 28 14:21:31 2004
From: cbl_mbi1 at yahoo.com (Brian Little)
Date: Tue Sep 28 14:21:35 2004
Subject: APM: installing modules on a virtual server
In-Reply-To: <4159AF4D.8050202@sam-i-am.com>
Message-ID: <20040928192131.81597.qmail@web20024.mail.yahoo.com>

   Setting PERL5LIB to ~/perl isn't going to point to the right place (i.e.
~/perl5lib).  If you're using bash and you're installing your modules into perl5lib
then you might want to try

      export PERL5LIB=~/perl5lib



--- Sam Foster <austin.pm@sam-i-am.com> wrote:

> (excuse the top posting.. I did trim a little honest)
> I did some searching yesterday and came up with this:
> http://www.bluereef.net/support/extensions/development/perlmodown.html
> 
> I'm using bash.
> 	setenv PERL5LIB ~/perl
> gives me the error that setenv is not available.
> 
> 	export PERL5LIB=~/perl
> gave me nothing.
> I also tried
> 	perl -MCPAN -Mlib -e "unshift(@INC, '~/perl5lib');print join(\"\n\", 
> @INC); shell;"
> 
> and this worked in that it printed out @INC and showed it to include 
> ~/perl5lib.. But when I went to cpan> test , it was still not picking up 
> the dependancies even after I let it install them again.
> E.g. to install File::Slurp, it runs tests which want Test::More. 
> Test::More wants Test::Harness. Each time it spots the dependancy, 
> installs the module, but the test still fails as it cant find it (in 
> ~/perl5lib)
> This is a trivial example, and I could probably make it skip those 
> tests, but with others (e.g XML::LibXML) it becomes more of a hassle.
> 
> Which brings me back to setenv PERL5LIB.. still no word back from tech 
> support on that one (they said they were getting a technician to look at 
> it.. and there's me thinking I'd called technical support :)
> 
> Sam
> 
> 
> Taylor Carpenter wrote:
> > On Tue, 28 Sep 2004 11:00:23 -0500, "Sam Foster"
> > <austin.pm@sam-i-am.com> said:
> > 
> >>Apparently I should be able to do:
> >>% setenv PERL5LIB ~/perl5lib
> >>but setenv is also apparently not available.
> > 
> > Did you try export PERL5LIB (after doing PERL5LIB=~/perl5lib)?  For bash
> > you can do 
> >    export PERL5LIB=~/perl5lib
> > but for some shells you have to first set then export the variable. 
> > Be sure to add
> >    use lib '/path/to/my/perl5lib'; 
> > to your scripts.
> > 
> > For CPAN you want the following set when configuring CPAN
> > 
> >    makepl_arg: PREFIX=~/perl5lib/
> >    cpan_home: /path/to/my/perl5lib/ANY_NAME_YOU_WANT
> > 
> _______________________________________________
> Austin mailing list
> Austin@mail.pm.org
> http://mail.pm.org/mailman/listinfo/austin
> 



		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com
From ian at remmler.org  Tue Sep 28 14:25:08 2004
From: ian at remmler.org (Ian Remmler)
Date: Tue Sep 28 14:25:15 2004
Subject: APM: installing modules on a virtual server
In-Reply-To: <4159AF4D.8050202@sam-i-am.com>
References: <41598A97.1070106@sam-i-am.com>
	<1096389006.8500.205331481@webmail.messagingengine.com>
	<4159AF4D.8050202@sam-i-am.com>
Message-ID: <20040928192508.GA852@remmler.org>

On Tue, Sep 28, 2004 at 01:37:01PM -0500, Sam Foster wrote:
> I'm using bash.
> 	setenv PERL5LIB ~/perl
> gives me the error that setenv is not available.
> 
> 	export PERL5LIB=~/perl
> gave me nothing.

The latter in bash does exactly what the former does in cshell.
That should be all you need.  setenv does not exist in
bourne-based shells.  If you want to use setenv you can always
run csh, then try it from there.

If by "gave me nothing" above, you mean there was no output from
the command, that's normal.  Just do a "set |grep PERL5LIB" or
"echo $PERL5LIB" to see the contents.  If you already know this
stuff, excuse the wasted bytes.

Unless the perl binary is wonkey, I can't see a reason why that
shouldn't work.  I'm using a similar setup at work with no
problems.  I haven't needed the 'use lib' trick.

-- 
Go banana!
From austin.pm at sam-i-am.com  Tue Sep 28 14:25:28 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Tue Sep 28 14:25:31 2004
Subject: APM: installing modules on a virtual server
In-Reply-To: <20040928192131.81597.qmail@web20024.mail.yahoo.com>
References: <20040928192131.81597.qmail@web20024.mail.yahoo.com>
Message-ID: <4159BAA8.5030000@sam-i-am.com>

Brian Little wrote:

>    Setting PERL5LIB to ~/perl isn't going to point to the right place (i.e.
> ~/perl5lib).  If you're using bash and you're installing your modules into perl5lib
> then you might want to try
> 
>       export PERL5LIB=~/perl5lib

sorry, that's what I meant.. just a typo in my message.

	export PERL5LIB=~/perl5lib

doesn't do anything .. no message, no result.

Sam
From austin.pm at sam-i-am.com  Tue Sep 28 14:42:34 2004
From: austin.pm at sam-i-am.com (Sam Foster)
Date: Tue Sep 28 14:42:37 2004
Subject: APM: installing modules on a virtual server
In-Reply-To: <20040928192508.GA852@remmler.org>
References: <41598A97.1070106@sam-i-am.com>
	<1096389006.8500.205331481@webmail.messagingengine.com>
	<4159AF4D.8050202@sam-i-am.com> <20040928192508.GA852@remmler.org>
Message-ID: <4159BEAA.4060203@sam-i-am.com>

Ian Remmler wrote:
> If by "gave me nothing" above, you mean there was no output from
> the command, that's normal.  Just do a "set |grep PERL5LIB" or
> "echo $PERL5LIB" to see the contents.  If you already know this
> stuff, excuse the wasted bytes.
> 
> Unless the perl binary is wonkey, I can't see a reason why that
> shouldn't work.  I'm using a similar setup at work with no
> problems.  I haven't needed the 'use lib' trick.
> 
Ahh.. Thank you Ian for pointing out the should-have-been-obvious.
PERL5LIB was indeed being set, but running the cpan shell having set it, 
gave me the same result - modules installed to ~/perl5lib are not being 
found.
In the Test::* example, I've got

~/perl5lib/lib/Test/Builder.pm
~/perl5lib/lib/Test/Harness
~/perl5lib/lib/Test/Harness.pm
~/perl5lib/lib/Test/More.pm
~/perl5lib/lib/Test/Simple.pm
~/perl5lib/lib/Test/Tutorial.pod

(~/perl5lib resolves to /home/usgm/perl5lib)

yet, the File::Slurp tests complain:

t/stdin.............Can't locate Test/More.pm in @INC (@INC contains: 
blib/arch blib/lib /usr/lib/perl5/5.6.1/i686-linux 
/usr/lib/perl5/5.6.1/i686-linux /usr/lib/perl5/5.6.1/i686-linux 
/usr/lib/perl5/5.6.1 /home/usgm/perl5lib /usr/lib/perl5/5.6.1/i686-linux 
/usr/lib/perl5/5.6.1/i686-linux /usr/lib/perl5/5.6.1 
/usr/lib/perl5/site_perl/5.6.1/i686-linux 
/usr/lib/perl5/site_perl/5.6.1/i686-linux /usr/lib/perl5/site_perl/5.6.1 
/usr/lib/perl5/site_perl/5.6.1/i686-linux /usr/lib/perl5/site_perl/5.6.1 
/usr/lib/perl5/site_perl . /usr/lib/perl5/5.6.1/i686-linux 
/usr/lib/perl5/5.6.1 /usr/lib/perl5/site_perl/5.6.1/i686-linux 
/usr/lib/perl5/site_perl/5.6.1 /usr/lib/perl5/site_perl .) at t/stdin.t 
line 9.

/home/usgm/perl5lib is in fact in @INC... However,
	export PERL5LIB = ~/perl5lib/lib
does make it all work. I guess I was expecting it to search down from 
~/perl5lib, it didn't occur to me I needed the extra '/lib'.

thanks all, I think I'm back on track.

Sam