427 lines
12 KiB
Perl
427 lines
12 KiB
Perl
|
#!/usr/bin/perl
|
||
|
|
||
|
=head1 NAME
|
||
|
|
||
|
JMX::Jmx4Perl::Request - A jmx4perl request
|
||
|
|
||
|
=head1 SYNOPSIS
|
||
|
|
||
|
$req = JMX::Jmx4Perl::Request->new(READ,$mbean,$attribute);
|
||
|
|
||
|
=head1 DESCRIPTION
|
||
|
|
||
|
A L<JMX::Jmx4Perl::Request> encapsulates a request for various operational
|
||
|
types.
|
||
|
|
||
|
The following attributes are available:
|
||
|
|
||
|
=over
|
||
|
|
||
|
=item mbean
|
||
|
|
||
|
Name of the targetted mbean in its canonical format.
|
||
|
|
||
|
=item type
|
||
|
|
||
|
Type of request, which should be one of the constants
|
||
|
|
||
|
=over
|
||
|
|
||
|
=item READ
|
||
|
|
||
|
Get the value of a attribute
|
||
|
|
||
|
=item WRITE
|
||
|
|
||
|
Write an attribute
|
||
|
|
||
|
=item EXEC
|
||
|
|
||
|
Execute an JMX operation
|
||
|
|
||
|
=item LIST
|
||
|
|
||
|
List all MBeans available
|
||
|
|
||
|
=item SEARCH
|
||
|
|
||
|
Search for MBeans
|
||
|
|
||
|
=item AGENT_VERSION
|
||
|
|
||
|
Get the agent's version and extra runtime information of the serverside.
|
||
|
|
||
|
=item REGISTER_NOTIFICATION
|
||
|
|
||
|
Register for a JMX notification (not supported yet)
|
||
|
|
||
|
=item REMOVE_NOTIFICATION
|
||
|
|
||
|
Remove a JMX notification (not supported yet)
|
||
|
|
||
|
=back
|
||
|
|
||
|
=item attribute
|
||
|
|
||
|
If type is C<READ> or C<WRITE> this specifies the requested
|
||
|
attribute
|
||
|
|
||
|
=item value
|
||
|
|
||
|
For C<WRITE> this specifies the value to set
|
||
|
|
||
|
=item arguments
|
||
|
|
||
|
List of arguments of C<EXEC> operations
|
||
|
|
||
|
=item path
|
||
|
|
||
|
This optional parameter can be used to specify a nested value in an complex
|
||
|
mbean attribute or nested return value from a JMX operation. For example, the
|
||
|
MBean C<java.lang:type=Memory>'s attribute C<HeapMemoryUsage> is a complex
|
||
|
value, which looks in the JSON representation like
|
||
|
|
||
|
"value":{"init":0,"max":518979584,"committed":41381888,"used":33442568}
|
||
|
|
||
|
So, to fetch the C<"used"> value only, specify C<used> as path within the
|
||
|
request. You can access deeper nested values by building up a path with "/" as
|
||
|
separator. This looks a bit like a simplified form of XPath.
|
||
|
|
||
|
=item maxDepth, maxObjects, maxCollectionSize, ignoreErrors
|
||
|
|
||
|
With these number you can restrict the size of the JSON structure
|
||
|
returned. C<maxDepth> gives the maximum nesting level of the JSON
|
||
|
object,C<maxObjects> returns the maximum number of objects to be returned in
|
||
|
total and C<maxCollectionSize> restrict the number of all arrays and
|
||
|
collections (maps, lists) in the answer. Note, that you should use this
|
||
|
restrictions if you are doing massive bulk operations. C<ignoreErrors> is
|
||
|
useful for read requests with multiple attributes to skip errors while reading
|
||
|
attribute values on the errors side (the error text will be set as value).
|
||
|
|
||
|
=item target
|
||
|
|
||
|
If given, the request is processed by the agent in proxy mode, i.e. it will
|
||
|
proxy to another server exposing via a JSR-160 connector. C<target> is a hash
|
||
|
which contains information how to reach the target service via the proxy. This
|
||
|
hash knows the following keys:
|
||
|
|
||
|
=over
|
||
|
|
||
|
=item url
|
||
|
|
||
|
JMX service URL as specified in JSR-160 pointing to the target server.
|
||
|
|
||
|
=item env
|
||
|
|
||
|
Further context information which is another hash.
|
||
|
|
||
|
=back
|
||
|
|
||
|
=back
|
||
|
|
||
|
=head1 METHODS
|
||
|
|
||
|
=over
|
||
|
|
||
|
=cut
|
||
|
|
||
|
package JMX::Jmx4Perl::Request;
|
||
|
|
||
|
use strict;
|
||
|
use vars qw(@EXPORT);
|
||
|
use Carp;
|
||
|
use Data::Dumper;
|
||
|
|
||
|
use base qw(Exporter);
|
||
|
@EXPORT = (
|
||
|
"READ","WRITE","EXEC","LIST", "SEARCH",
|
||
|
"REGNOTIF","REMNOTIF", "AGENT_VERSION"
|
||
|
);
|
||
|
|
||
|
use constant READ => "read";
|
||
|
use constant WRITE => "write";
|
||
|
use constant EXEC => "exec";
|
||
|
use constant LIST => "list";
|
||
|
use constant SEARCH => "search";
|
||
|
use constant REGNOTIF => "regnotif";
|
||
|
use constant REMNOTIF => "remnotif";
|
||
|
use constant AGENT_VERSION => "version";
|
||
|
|
||
|
my $TYPES =
|
||
|
{ map { $_ => 1 } (READ, WRITE, EXEC, LIST, SEARCH,
|
||
|
REGNOTIF, REMNOTIF, AGENT_VERSION) };
|
||
|
|
||
|
=item $req = new JMX::Jmx4Perl::Request(....);
|
||
|
|
||
|
$req = new JMX::Jmx4Perl::Request(READ,$mbean,$attribute,$path, { ... options ... } );
|
||
|
$req = new JMX::Jmx4Perl::Request(READ,{ mbean => $mbean,... });
|
||
|
$req = new JMX::Jmx4Perl::Request({type => READ, mbean => $mbean, ... });
|
||
|
|
||
|
The constructor can be used in various way. In the simplest form, you provide
|
||
|
the type as first argument and depending on the type one or more additional
|
||
|
attributes which specify the request. The second form uses the type as first
|
||
|
parameter and a hashref containing named parameter for the request parameters
|
||
|
(for the names, see above). Finally you can specify the arguments completely as
|
||
|
a hashref, using 'type' for the entry specifying the request type.
|
||
|
|
||
|
For the options C<maxDepth>, C<maxObjects> and C<maxCollectionSize>, you can mix
|
||
|
them in into the hashref if using the hashed argument format. For the first
|
||
|
format, these options are given as a final hashref.
|
||
|
|
||
|
The option C<method> can be used to suggest a HTTP request method to use. By
|
||
|
default, the agent decides automatically which HTTP method to use depending on
|
||
|
the number of requests and whether an extended format should be used (which is
|
||
|
only possible with an HTTP POST request). The value of this option can be
|
||
|
either C<post> or C<get>, dependening on your preference.
|
||
|
|
||
|
If the request should be proxied through this request, a target configuration
|
||
|
needs to be given as optional parameter. The target configuration consists of a
|
||
|
JMX service C<url> and a optional environment, which is given as a key-value
|
||
|
map. For example
|
||
|
|
||
|
$req = new JMX::Jmx4Perl::Request(..., {
|
||
|
target => {
|
||
|
url => "",
|
||
|
env => { ..... }
|
||
|
}
|
||
|
} );
|
||
|
|
||
|
Note, depending on the type, some parameters are mandatory. The mandatory
|
||
|
parameters and the order of the arguments for the constructor variant without
|
||
|
named parameters are:
|
||
|
|
||
|
=over
|
||
|
|
||
|
=item C<READ>
|
||
|
|
||
|
Order : $mbean, $attribute, $path
|
||
|
Mandatory: $mbean, $attribute
|
||
|
|
||
|
Note that C<$attribute> can be either a single name or a reference to a list
|
||
|
of attribute names.
|
||
|
|
||
|
=item C<WRITE>
|
||
|
|
||
|
Order : $mbean, $attribute, $value, $path
|
||
|
Mandatory: $mbean, $attribute, $value
|
||
|
|
||
|
=item C<EXEC>
|
||
|
|
||
|
Order : $mbean, $operation, $arg1, $arg2, ...
|
||
|
Mandatory: $mbean, $operation
|
||
|
|
||
|
|
||
|
=item C<LIST>
|
||
|
|
||
|
Order : $path
|
||
|
|
||
|
=item C<SEARCH>
|
||
|
|
||
|
Order : $pattern
|
||
|
Mandatory: $pattern
|
||
|
|
||
|
=back
|
||
|
|
||
|
=cut
|
||
|
|
||
|
sub new {
|
||
|
my $class = shift;
|
||
|
my $type = shift;
|
||
|
my $self;
|
||
|
# Hash as argument
|
||
|
if (ref($type) eq "HASH") {
|
||
|
$self = $type;
|
||
|
$type = $self->{type};
|
||
|
}
|
||
|
croak "Invalid type '",$type,"' given (should be one of ",join(" ",keys %$TYPES),")" unless $TYPES->{$type};
|
||
|
|
||
|
# Hash comes after type
|
||
|
if (!$self) {
|
||
|
if (ref($_[0]) eq "HASH") {
|
||
|
$self = $_[0];
|
||
|
$self->{type} = $type;
|
||
|
} else {
|
||
|
# Unnamed arguments
|
||
|
$self = {type => $type};
|
||
|
|
||
|
# Options are given as last part
|
||
|
my $opts = $_[scalar(@_)-1];
|
||
|
if (ref($opts) eq "HASH") {
|
||
|
pop @_;
|
||
|
map { $self->{$_} = $opts->{$_} } keys %$opts;
|
||
|
if ($self->{method}) {
|
||
|
# Canonicalize and verify
|
||
|
method($self,$self->{method});
|
||
|
}
|
||
|
}
|
||
|
if ($type eq READ) {
|
||
|
$self->{mbean} = shift;
|
||
|
$self->{attribute} = shift;
|
||
|
$self->{path} = shift;
|
||
|
# Use post for complex read requests
|
||
|
if (ref($self->{attribute}) eq "ARRAY") {
|
||
|
my $method = method($self);
|
||
|
if (defined($method) && $method eq "GET") {
|
||
|
# Was already explicitely set
|
||
|
die "Cannot query for multiple attributes " . join(",",@{$self->{attributes}}) . " with a GET request"
|
||
|
if ref($self->{attribute}) eq "ARRAY";
|
||
|
}
|
||
|
method($self,"POST");
|
||
|
}
|
||
|
} elsif ($type eq WRITE) {
|
||
|
$self->{mbean} = shift;
|
||
|
$self->{attribute} = shift;
|
||
|
$self->{value} = shift;
|
||
|
$self->{path} = shift;
|
||
|
} elsif ($type eq EXEC) {
|
||
|
$self->{mbean} = shift;
|
||
|
$self->{operation} = shift;
|
||
|
$self->{arguments} = [ @_ ];
|
||
|
} elsif ($type eq LIST) {
|
||
|
$self->{path} = shift;
|
||
|
} elsif ($type eq SEARCH) {
|
||
|
$self->{mbean} = shift;
|
||
|
#No check here until now, is done on the server side as well.
|
||
|
#die "MBean name ",$self->{mbean}," is not a pattern" unless is_mbean_pattern($self);
|
||
|
} elsif ($type eq AGENT_VERSION) {
|
||
|
# No extra parameters required
|
||
|
} else {
|
||
|
croak "Type ",$type," not supported yet";
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
bless $self,(ref($class) || $class);
|
||
|
$self->_validate();
|
||
|
return $self;
|
||
|
}
|
||
|
|
||
|
=item $req->method()
|
||
|
|
||
|
=item $req->method("POST")
|
||
|
|
||
|
Set the HTTP request method for this requst excplicitely. If not provided
|
||
|
either during construction time (config key 'method') a prefered request
|
||
|
method is determined dynamically based on the request contents.
|
||
|
|
||
|
=cut
|
||
|
|
||
|
sub method {
|
||
|
my $self = shift;
|
||
|
my $value = shift;
|
||
|
if (defined($value)) {
|
||
|
die "Unknown request method ",$value if length($value) && uc($value) !~ /^(POST|GET)$/i;
|
||
|
$self->{method} = uc($value);
|
||
|
}
|
||
|
return defined($self->{method}) ? $self->{method} : undef;
|
||
|
}
|
||
|
|
||
|
=item $req->is_mbean_pattern
|
||
|
|
||
|
Returns true, if the MBean name used in this request is a MBean pattern (which
|
||
|
can be used in C<SEARCH> or for C<READ>) or not
|
||
|
|
||
|
=cut
|
||
|
|
||
|
sub is_mbean_pattern {
|
||
|
my $self = shift;
|
||
|
my $mbean = shift || $self->{mbean};
|
||
|
return 1 unless $mbean;
|
||
|
my ($domain,$rest) = split(/:/,$mbean,2);
|
||
|
return 1 if $domain =~ /[*?]/;
|
||
|
return 1 if $rest =~ /\*$/;
|
||
|
|
||
|
while ($rest) {
|
||
|
#print "R: $rest\n";
|
||
|
$rest =~ s/([^=]+)\s*=\s*//;
|
||
|
my $key = $1;
|
||
|
my $value;
|
||
|
if ($rest =~ /^"/) {
|
||
|
$rest =~ s/("(\\"|[^"])+")(\s*,\s*|$)//;
|
||
|
$value = $1;
|
||
|
# Pattern in quoted values must not be preceded by a \
|
||
|
return 1 if $value =~ /(?<!\\)[\*\?]/;
|
||
|
} else {
|
||
|
$rest =~ s/([^,]+)(\s*,\s*|$)//;
|
||
|
$value = $1;
|
||
|
return 1 if $value =~ /[\*\?]/;
|
||
|
}
|
||
|
#print "K: $key V: $value\n";
|
||
|
}
|
||
|
return 0;
|
||
|
}
|
||
|
|
||
|
=item $request->get("type")
|
||
|
|
||
|
Get a request parameter
|
||
|
|
||
|
=cut
|
||
|
|
||
|
sub get {
|
||
|
my $self = shift;
|
||
|
my $name = shift;
|
||
|
return $self->{$name};
|
||
|
}
|
||
|
|
||
|
# Internal check for validating that all arguments are given
|
||
|
sub _validate {
|
||
|
my $self = shift;
|
||
|
if ($self->{type} eq READ || $self->{type} eq WRITE) {
|
||
|
die $self->{type} . ": No mbean name given\n",Dumper($self) unless $self->{mbean};
|
||
|
die $self->{type} . ": No attribute name but path is given\n" if (!$self->{attribute} && $self->{path});
|
||
|
}
|
||
|
if ($self->{type} eq WRITE) {
|
||
|
die $self->{type} . ": No value given\n" unless defined($self->{value});
|
||
|
}
|
||
|
if ($self->{type} eq EXEC) {
|
||
|
die $self->{type} . ": No mbean name given\n" unless $self->{mbean};
|
||
|
die $self->{type} . ": No operation name given\n" unless $self->{operation};
|
||
|
}
|
||
|
}
|
||
|
|
||
|
# Called for post requests
|
||
|
sub TO_JSON {
|
||
|
my $self = shift;
|
||
|
my $ret = {
|
||
|
type => $self->{type} ? uc($self->{type}) : undef,
|
||
|
};
|
||
|
for my $k (qw(mbean attribute path value operation arguments target)) {
|
||
|
$ret->{$k} = $self->{$k} if defined($self->{$k});
|
||
|
}
|
||
|
my %config;
|
||
|
for my $k (qw(maxDepth maxObjects maxCollectionSize ignoreErrors)) {
|
||
|
$config{$k} = $self->{$k} if defined($self->{$k});
|
||
|
}
|
||
|
$ret->{config} = \%config if keys(%config);
|
||
|
return $ret;
|
||
|
}
|
||
|
|
||
|
=back
|
||
|
|
||
|
=head1 LICENSE
|
||
|
|
||
|
This file is part of jmx4perl.
|
||
|
|
||
|
Jmx4perl is free software: you can redistribute it and/or modify
|
||
|
it under the terms of the GNU General Public License as published by
|
||
|
the Free Software Foundation, either version 2 of the License, or
|
||
|
(at your option) any later version.
|
||
|
|
||
|
jmx4perl is distributed in the hope that it will be useful,
|
||
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||
|
GNU General Public License for more details.
|
||
|
|
||
|
You should have received a copy of the GNU General Public License
|
||
|
along with jmx4perl. If not, see <http://www.gnu.org/licenses/>.
|
||
|
|
||
|
A commercial license is available as well. Please contact roland@cpan.org for
|
||
|
further details.
|
||
|
|
||
|
=head1 AUTHOR
|
||
|
|
||
|
roland@cpan.org
|
||
|
|
||
|
=cut
|
||
|
|
||
|
1;
|