#!/usr/bin/perl

use strict;
use autodie;
use warnings;
use CPAN;

CPAN::Shell->install(
qw(
IO::Scalar
ExtUtils::Constant
Encode
ExtUtils::ParseXS
Pod::Usage
Getopt::Long
Text::ParseWords
Module::CoreList
Module::Load
Module::Load::Conditional
Params::Check
Locale::Maketext::Simple
IPC::Cmd
ExtUtils::CBuilder
ExtUtils::Manifest
Perl::OSType
CPAN::Meta::Requirements
CPAN::Meta::YAML
JSON::PP
Parse::CPAN::Meta
CPAN::Meta
ExtUtils::Install
File::Temp
version
IO::File
Module::Metadata
Module::Build
constant
Text::Wrap
Pod::Escapes
Pod::Simple
Pod::Man
Test
Exporter
Carp
File::Path
Test::Harness
Test::More
Scalar::Util
File::Spec
ExtUtils::MakeMaker
Net::SSH2));
[download]

Net::SSH2::Channel goes into more detail about what the channel is and does. Read that thoroughly. It'll help.

For more details about "exec", you can do perl -f exec for more info.

See: libssh2.org for more info.

Hi, I looked into Net::SSH2::Channel documentation and it's also just one liner description of all the methods used there.However, I've read about exec.

See A little demo for Net::SSH2.

As far as blocking and channels go, it's not simple, read perldoc Net::SSH2::Channel

If you want to exec something on the server, use this:

#!/usr/bin/perl
use warnings;
use strict;
use Net::SSH2;

my $pass = 'hahahaha';
my $ssh2 = Net::SSH2->new();

$ssh2->connect('localhost') or die "Unable to connect Host $@ \n";

#plain password login
#$ssh2->auth_password('z','ztester') or die "Unable to login $@ \n";

# works when run from z's homedir because you need
# permission to read the keys
$ssh2->auth_publickey('zentara',
                 '/home/zentara/.ssh/id_dsa.pub',
         '/home/zentara/.ssh/id_dsa',
          $pass );


my $chan = $ssh2->channel();
$chan->blocking(1);

$chan->exec("nohup /home/zentara/perlplay/net/zzsleep > foo.out 2> foo
+.err < /dev/null &");

$chan->send_eof;

exit;
[download]

Hi, I want to fire multiple system commands. Also, second system command would depend on the first system command output and so on. therefore, I guess, we cannot use exec here as exec overlays the parent process. We can use shell method then. If it is then can you please help me out to find out details about this shell method?

See Salva's advice in channel SSH2. You probably are better off running a background process on the server itself. You exec with a nohup and detach the pid back into the background. You can log all output. Why do you need the logic of command processing done on the remote machine thru an unreliable network? Launch your control script right on the server, and come back later and get the log of it running.

---

I'm not really a human, but I play one on earth.
Old Perl Programmer Haiku ................... flash japh