diff options
author | Scott Bronson <bronson@rinspin.com> | 2001-04-10 16:58:33 -0700 |
---|---|---|
committer | Jarkko Hietaniemi <jhi@iki.fi> | 2001-04-15 23:40:35 +0000 |
commit | 072620d9f868834f547a0c37d377003e022a1a62 (patch) | |
tree | 7694fe4c68176f06edf450bb138f2b1f6a47c11e /lib | |
parent | 0e172911d4da2e72124db457593beb934daca235 (diff) | |
download | perl-072620d9f868834f547a0c37d377003e022a1a62.tar.gz |
Re: Net::Ping patch, adds stream protocol
Message-ID: <20010410235833.N29719@rinspin.com>
p4raw-id: //depot/perl@9715
Diffstat (limited to 'lib')
-rw-r--r-- | lib/Net/Ping.pm | 397 |
1 files changed, 281 insertions, 116 deletions
diff --git a/lib/Net/Ping.pm b/lib/Net/Ping.pm index b3c6b97cad..5ab3508a64 100644 --- a/lib/Net/Ping.pm +++ b/lib/Net/Ping.pm @@ -1,6 +1,7 @@ package Net::Ping; # Current maintainer: colinm@cpan.org (Colin McMillen) +# stream protocol: bronson@trestle.com (Scott Bronson) # # Original author: mose@ccsn.edu (Russell Mosemann) # @@ -23,7 +24,7 @@ use Carp; @ISA = qw(Exporter); @EXPORT = qw(pingecho); -$VERSION = 2.03; +$VERSION = 2.04; # Constants @@ -70,8 +71,8 @@ sub new bless($self, $class); $proto = $def_proto unless $proto; # Determine the protocol - croak('Protocol for ping must be "icmp", "tcp", "udp", or "external"') - unless $proto =~ m/^(tcp|udp|icmp|external)$/; + croak('Protocol for ping must be "icmp", "udp", "tcp", "stream", or "external"') + unless $proto =~ m/^(icmp|udp|tcp|stream|external)$/; $self->{"proto"} = $proto; $timeout = $def_timeout unless $timeout; # Determine the timeout @@ -114,7 +115,7 @@ sub new socket($self->{"fh"}, &PF_INET(), &SOCK_RAW(), $self->{"proto_num"}) || croak("icmp socket error - $!"); } - elsif ($self->{"proto"} eq "tcp") # Just a file handle for now + elsif ($self->{"proto"} eq "tcp" || $self->{"proto"} eq "stream") { $self->{"proto_num"} = (getprotobyname('tcp'))[2] || croak("Can't get tcp protocol by name"); @@ -154,6 +155,7 @@ sub ping return $self->ping_udp($ip, $timeout) if $self->{"proto"} eq "udp"; return $self->ping_icmp($ip, $timeout) if $self->{"proto"} eq "icmp"; return $self->ping_tcp($ip, $timeout) if $self->{"proto"} eq "tcp"; + return $self->ping_stream($ip, $timeout) if $self->{"proto"} eq "stream"; croak("Unknown protocol \"$self->{proto}\" in ping()"); } @@ -283,96 +285,10 @@ sub checksum return(~(($chk >> 16) + $chk) & 0xffff); # Again and complement } -# Description: Perform a tcp echo ping. Since a tcp connection is -# host specific, we have to open and close each connection here. We -# can't just leave a socket open. Because of the robust nature of -# tcp, it will take a while before it gives up trying to establish a -# connection. Therefore, we use select() on a non-blocking socket to -# check against our timeout. No data bytes are actually -# sent since the successful establishment of a connection is proof -# enough of the reachability of the remote host. Also, tcp is -# expensive and doesn't need our help to add to the overhead. - -sub ping_tcp -{ - my ($self, - $ip, # Packed IP number of the host - $timeout # Seconds after which ping times out - ) = @_; - my ($saddr, # sockaddr_in with port and ip - $rin, # Used in select() - $ret # The return value - ); - - socket($self->{"fh"}, &PF_INET(), &SOCK_STREAM(), $self->{"proto_num"}) || - croak("tcp socket error - $!"); - - $saddr = sockaddr_in($self->{"port_num"}, $ip); - - $ret = 0; # Default to unreachable - - # Buggy Winsock API doesn't allow us to use non-blocking connect() - # calls. Hence, if our OS is Windows, we need to create a new process - # to run a blocking connect attempt, and kill it after the timeout has - # passed. - if ($^O =~ /win32/i) - { - my ($child, $ret, $pid, $time); - my $host = inet_ntoa($ip); - - # The code we will be executing in our new process. - my $code = '"use Net::Ping; $p = Net::Ping->new(\'tcp\'); '; - $code .= 'exit($p->_ping_tcp_win(' . $host . '))"'; - - # Call the process. - $pid = system(1, "perl", "-e", $code); - - # Import the POSIX version of <sys/wait.h> - require POSIX; - import POSIX qw(:sys_wait_h); - - # Get the current time; will be used to tell if we've timed out. - $time = time; - - # Wait for the child to return or for the timeout to expire. - do { - $child = waitpid($pid, &WNOHANG); - $ret = $?; - } until time > ($time + $timeout) or $child; - - # Return an appropriate value; 0 if the child didn't return, - # the return value of the child otherwise. - return $ret >> 8 if $child; - - kill $pid; - return 0; - } - - # If our OS isn't Windows, do this stuff instead... - else - { - # Try a non-blocking TCP connect to the remote echo port. - # Our call to select() below will stop after the timeout has - # passed or set the return value to true if the connection - # succeeds in time. - $self->{"fh"}->blocking(0); - connect($self->{"fh"}, $saddr); - - $rin = ""; - vec($rin, fileno($self->{"fh"}), 1) = 1; - $ret = 1 if select($rin, undef, undef, $timeout); - - # Close our filehandle, restore it to its default state (i.e. blocking), - # and return our result. - $self->{"fh"}->blocking(1); - $self->{"fh"}->close(); - } - return($ret); -} - # Warning: this method may generate false positives. # It is meant to be a private method and should only # be invoked by ping_tcp() if $^O =~ /win32/i. + sub _ping_tcp_win { my ($self, @@ -401,6 +317,218 @@ sub _ping_tcp_win return $ret; } +# Buggy Winsock API doesn't allow us to use non-blocking connect() +# calls. Hence, if our OS is Windows, we need to create a new process +# to run a blocking connect attempt, and kill it after the timeout has +# passed. Unfortunately, this won't work with the stream protocol. + +sub ping_tcp_win32 +{ + my ($self, + $ip, # Packed IP number of the host + $timeout # Seconds after which open times out + ) = @_; + + socket($self->{"fh"}, &PF_INET(), &SOCK_STREAM(), $self->{"proto_num"}) || + croak("tcp socket error - $!"); + + my $saddr = sockaddr_in($self->{"port_num"}, $ip); + + my ($child, $ret, $pid, $time); + my $host = inet_ntoa($ip); + + # The code we will be executing in our new process. + my $code = '"use Net::Ping; $p = Net::Ping->new(\'tcp\'); '; + $code .= 'exit($p->_ping_tcp_win(' . $host . '))"'; + + # Call the process. + $pid = system(1, "perl", "-e", $code); + + # Import the POSIX version of <sys/wait.h> + require POSIX; + import POSIX qw(:sys_wait_h); + + # Get the current time; will be used to tell if we've timed out. + $time = time; + + # Wait for the child to return or for the timeout to expire. + do { + $child = waitpid($pid, &WNOHANG); + $ret = $?; + } until time > ($time + $timeout) or $child; + + # Return an appropriate value; 0 if the child didn't return, + # the return value of the child otherwise. + return $ret >> 8 if $child; + + kill $pid; + return 0; +} + +# This writes the given string to the socket and then reads it +# back. It returns 1 on success, 0 on failure. +sub tcp_echo +{ + my $self = shift; + my $timeout = shift; + my $pingstring = shift; + + my $ret = undef; + my $time = time; + my $wrstr = $pingstring; + my $rdstr = ""; + + eval <<'EOM'; + do { + my $rin = ""; + vec($rin, $self->{"fh"}->fileno(), 1) = 1; + + my $rout = undef; + if($wrstr) { + $rout = ""; + vec($rout, $self->{"fh"}->fileno(), 1) = 1; + } + + if(select($rin, $rout, undef, ($time + $timeout) - time())) { + + if($rout && vec($rout,$self->{"fh"}->fileno(),1)) { + my $num = syswrite($self->{"fh"}, $wrstr); + if($num) { + # If it was a partial write, update and try again. + $wrstr = substr($wrstr,$num); + } else { + # There was an error. + $ret = 0; + } + } + + if(vec($rin,$self->{"fh"}->fileno(),1)) { + my $reply; + if(sysread($self->{"fh"},$reply,length($pingstring)-length($rdstr))) { + $rdstr .= $reply; + $ret = 1 if $rdstr eq $pingstring; + } else { + # There was an error. + $ret = 0; + } + } + + } + } until time() > ($time + $timeout) || defined($ret); +EOM + + return $ret; +} + +sub tcp_connect +{ + my ($self, + $ip, # Packed IP number of the host + $timeout # Seconds after which open times out + ) = @_; + + # Should we go back to using blocking IO and alarms to implement + # the stream protocol on win32? + croak "no nonblocking io -- can't stream ping on win32" + if ($^O =~ /win32/i); + + $self->{"ip"} = $ip; + + socket($self->{"fh"}, &PF_INET(), &SOCK_STREAM(), $self->{"proto_num"}) || + croak("tcp socket error - $!"); + + my $saddr = sockaddr_in($self->{"port_num"}, $ip); + my $ret = 0; + + # Try a non-blocking TCP connect to the remote echo port. + # Our call to select() below will stop after the timeout has + # passed or set the return value to true if the connection + # succeeds in time. + $self->{"fh"}->blocking(0); + connect($self->{"fh"}, $saddr); + + # This replaces the breakage where we were listening on a + # socket that would never produce any data. This works, but + # it's now quite a bit heavier than the old Net::Ping. I'd + # like to see it reverted. + return $self->tcp_echo($timeout, "ping!\n"); +} + +# Description: Perform a tcp echo ping. Since a tcp connection is +# host specific, we have to open and close each connection here. We +# can't just leave a socket open. Because of the robust nature of +# tcp, it will take a while before it gives up trying to establish a +# connection. Therefore, we use select() on a non-blocking socket to +# check against our timeout. No data bytes are actually +# sent since the successful establishment of a connection is proof +# enough of the reachability of the remote host. Also, tcp is +# expensive and doesn't need our help to add to the overhead. + +sub ping_tcp +{ + my ($self, + $ip, # Packed IP number of the host + $timeout # Seconds after which ping times out + ) = @_; + + my $ret; + + # tcp_connect won't work on win32, so special-case it if need be. + if ($^O =~ /win32/i) { + $ret = $self->ping_tcp_win32($ip, $timeout); + } else { + $ret = $self->tcp_connect($ip, $timeout); + $self->{"fh"}->close(); + } + + return $ret; +} + +# Description: Perform a stream ping. If the tcp connection isn't +# already open, it opens it. It then sends some data and waits for +# a reply. It leaves the stream open on exit. + +sub ping_stream +{ + my ($self, + $ip, # Packed IP number of the host + $timeout # Seconds after which ping times out + ) = @_; + + my $pingstring = "ping!\n"; # The data we exchange with the server + + # Open the stream if it's not already open + if(!defined $self->{"fh"}->fileno()) { + $self->tcp_connect($ip, $timeout) or return 0; + } + + croak "tried to switch servers while stream pinging" + if $self->{"ip"} ne $ip; + + return $self->tcp_echo($timeout, "pingschwingping!\n"); +} + +# Description: opens the stream. You would do this if you want to +# separate the overhead of opening the stream from the first ping. + +sub open +{ + my ($self, + $ip, # Packed IP number of the host + $timeout # Seconds after which open times out + ) = @_; + + $timeout = $self->{"timeout"} unless $timeout; + + if($self->{"proto"} eq "stream") { + if(defined($self->{"fh"}->fileno())) { + croak("socket is already open"); + } else { + $self->tcp_connect($ip, $timeout); + } + } +} + # Description: Perform a udp echo ping. Construct a message of # at least the one-byte sequence number and any additional data bytes. # Send the message out and wait for a message to come back. If we @@ -527,39 +655,64 @@ hosts on a network. A ping object is first created with optional parameters, a variable number of hosts may be pinged multiple times and then the connection is closed. -You may choose one of four different protocols to use for the -ping. The "udp" protocol is the default. Note that a live remote host -may still fail to be pingable by one or more of these protocols. For -example, www.microsoft.com is generally alive but not pingable. +Ping supports five ping protocols, each with its own strengths +and weaknesses. The "udp" protocol is the default. A host +may be configured to respond to only a few of these protocols, +or even none at all. For example, www.microsoft.com is generally +alive but not pingable. -With the "tcp" protocol the ping() method attempts to establish a -connection to the remote host's echo port. If the connection is -successfully established, the remote host is considered reachable. No -data is actually echoed. This protocol does not require any special -privileges but has higher overhead than the other two protocols. +=over 4 + +=item icmp + +The C<ping()> method sends an icmp echo message to the remote host +(this is what the UNIX ping program does). +If the echoed message is received from the remote host and +the echoed information is correct, the remote host is considered +reachable. Specifying this protocol requires that the program +be run as root or that the program be setuid to root. -Specifying the "udp" protocol causes the ping() method to send a udp +=item udp + +The C<ping()> method sends a udp packet to the remote host's echo port. If the echoed packet is received from the remote host and the received packet contains the same data as the packet that was sent, the remote host is considered reachable. This protocol does not require any special privileges. -It should be borne in mind that, for both udp ping, a host +It should be borne in mind that, for both udp and tcp ping, a host will be reported as unreachable if it is not running the appropriate echo service. For Unix-like systems see L<inetd(8)> for more information. -If the "icmp" protocol is specified, the ping() method sends an icmp -echo message to the remote host, which is what the UNIX ping program -does. If the echoed message is received from the remote host and -the echoed information is correct, the remote host is considered -reachable. Specifying the "icmp" protocol requires that the program -be run as root or that the program be setuid to root. +=item tcp -If the "external" protocol is specified, the ping() method attempts to -use the C<Net::Ping::External> module to ping the remote host. -C<Net::Ping::External> interfaces with your system's default C<ping> -utility to perform the ping, and generally produces relatively +The C<ping()> method attempts to establish a +connection to the remote host's echo port. If the connection is +successfully established, the remote host is considered reachable. +Once the connection is made, it is torn down immediately -- no data +is actually echoed. This protocol does not require any special +privileges but has highest overhead of the protocols. + +=item stream + +This is just like the tcp protocol, except that once it establishes +the tcp connection, it keeps it up. Each subsequent ping +request re-uses the existing connection. stream +provides better performance than tcp since the connection +doesn't need to be created and torn down with every ping. It is +also the only protocol that will recognize that the original host is +gone, even if it is immediately replaced by an +identical host responding in exactly the same way. The drawback +is that you can only ping one host per Ping instance. You will get +an error if you neglect to call C<close()> before trying to ping +a different network device. + +=item external + +The ping() method attempts to use the C<Net::Ping::External> module to ping +the remote host. C<Net::Ping::External> interfaces with your system's default +L<ping(8)> utility to perform the ping, and generally produces relatively accurate results. If C<Net::Ping::External> if not installed on your system, specifying the "external" protocol will result in an error. @@ -594,6 +747,17 @@ there is a problem with the IP number, undef is returned. Otherwise, 1 is returned if the host is reachable and 0 if it is not. For all practical purposes, undef and 0 and can be treated as the same case. +=item $p->open($host); + +When you are using the stream protocol, this call pre-opens the +tcp socket. It's only necessary to do this if you want to +provide a different timeout when creating the connection, or +remove the overhead of establishing the connection from the +first ping. If you don't call C<open()>, the connection is +automatically openeed the first time C<ping()> is called. +This call simply does nothing if you are using any protocol other +than stream. + =item $p->close(); Close the network connection for this ping object. The network @@ -622,9 +786,8 @@ to implement a small wait (e.g. 25ms or more) between each ping to avoid flooding your network with packets. The icmp protocol requires that the program be run as root or that it -be setuid to root. The tcp and udp protocols do not require special -privileges, but not all network devices implement the echo protocol -for tcp or udp. +be setuid to root. The other protocols do not require special +privileges, but not all network devices implement tcp or udp echo. Local hosts should normally respond to pings within milliseconds. However, on a very congested network it may take up to 3 seconds or @@ -633,7 +796,9 @@ is set too low under these conditions, it will appear that the remote host is not reachable (which is almost the truth). Reachability doesn't necessarily mean that the remote host is actually -functioning beyond its ability to echo packets. +functioning beyond its ability to echo packets. tcp is slightly better +at indicating the health of a system than icmp because it uses more +of the networking stack to respond. Because of a lack of anything better, this module uses its own routines to pack and unpack ICMP packets. It would be better for a |