loncom/loncnew - diff

Return to loncnew CVS log

Up to [LON-CAPA] / loncom

Diff for /loncom/loncnew between versions 1.10 and 1.57

version 1.10, 2003/06/24 02:46:04	version 1.57, 2004/09/22 10:22:50
Line 7	Line 7
# Copyright Michigan State University Board of Trustees	# Copyright Michigan State University Board of Trustees
#	#
# This file is part of the LearningOnline Network with CAPA (LON-CAPA).	# This file is part of the LearningOnline Network with CAPA (LON-CAPA).
#	## LON-CAPA is free software; you can redistribute it and/or modify
# LON-CAPA is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by	# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or	# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.	# (at your option) any later version.
Line 27	Line 26
# http://www.lon-capa.org/	# http://www.lon-capa.org/
#	#
#	#
# new lonc handles n requestors spread out bver m connections to londs.	# new lonc handles n request out bver m connections to londs.
# This module is based on the Event class.	# This module is based on the Event class.
# Development iterations:	# Development iterations:
# - Setup basic event loop. (done)	# - Setup basic event loop. (done)
Line 36	Line 35
# - Add ability to create/negotiate lond connections (done).	# - Add ability to create/negotiate lond connections (done).
# - Add general logic for dispatching requests and timeouts. (done).	# - Add general logic for dispatching requests and timeouts. (done).
# - Add support for the lonc/lond requests. (done).	# - Add support for the lonc/lond requests. (done).
# - Add logging/status monitoring.	# - Add logging/status monitoring. (done)
# - Add Signal handling - HUP restarts. USR1 status report.	# - Add Signal handling - HUP restarts. USR1 status report. (done)
# - Add Configuration file I/O (done).	# - Add Configuration file I/O (done).
# - Add management/status request interface.	# - Add management/status request interface. (done)
# - Add deferred request capability. (done)	# - Add deferred request capability. (done)
# - Detect transmission timeouts.	# - Detect transmission timeouts. (done)
#

# Change log:
# $Log$
# Revision 1.10 2003/06/24 02:46:04 foxr
# Put a limit on the number of times we'll retry a connection.
# Start getting the signal stuff put in as well...note that need to get signals
# going or else 6the client will permanently give up on dead servers.
#
# Revision 1.9 2003/06/13 02:38:43 foxr
# Add logging in 'expected format'
#
# Revision 1.8 2003/06/11 02:04:35 foxr
# Support delayed transactions... this is done uniformly by encapsulating
# transactions in an object ... a LondTransaction that is implemented by
# LondTransaction.pm
#
# Revision 1.7 2003/06/03 01:59:39 foxr
# complete coding to support deferred transactions.
#
#	#

	use strict;
use lib "/home/httpd/lib/perl/";	use lib "/home/httpd/lib/perl/";
use lib "/home/foxr/newloncapa/types";
use Event qw(:DEFAULT );	use Event qw(:DEFAULT );
use POSIX qw(:signal_h);	use POSIX qw(:signal_h);
	use POSIX;
use IO::Socket;	use IO::Socket;
use IO::Socket::INET;	use IO::Socket::INET;
use IO::Socket::UNIX;	use IO::Socket::UNIX;
Line 83 use LONCAPA::Configuration;	Line 63 use LONCAPA::Configuration;
use LONCAPA::HashIterator;	use LONCAPA::HashIterator;


#
# Disable all signals we might receive from outside for now.
#
$SIG{QUIT} = IGNORE;
$SIG{HUP} = IGNORE;
$SIG{USR1} = IGNORE;
$SIG{INT} = IGNORE;
$SIG{CHLD} = IGNORE;
$SIG{__DIE__} = IGNORE;


# Read the httpd configuration file to get perl variables	# Read the httpd configuration file to get perl variables
# normally set in apache modules:	# normally set in apache modules:

Line 104 my %perlvar = %{$perlvarref};	Line 73 my %perlvar = %{$perlvarref};
# parent and shared variables.	# parent and shared variables.

my %ChildHash; # by pid -> host.	my %ChildHash; # by pid -> host.
	my %HostToPid; # By host -> pid.
	my %HostHash; # by loncapaname -> IP.


my $MaxConnectionCount = 10; # Will get from config later.	my $MaxConnectionCount = 10; # Will get from config later.
my $ClientConnection = 0; # Uniquifier for client events.	my $ClientConnection = 0; # Uniquifier for client events.

my $DebugLevel = 0;	my $DebugLevel = 0;
my $IdleTimeout= 3600; # Wait an hour before pruning connections.	my $NextDebugLevel= 2; # So Sigint can toggle this.
	my $IdleTimeout= 600; # Wait 10 minutes before pruning connections.

	my $LogTransactions = 0; # When True, all transactions/replies get logged.

#	#
# The variables below are only used by the child processes.	# The variables below are only used by the child processes.
#	#
my $RemoteHost; # Name of host child is talking to.	my $RemoteHost; # Name of host child is talking to.
my $UnixSocketDir= "/home/httpd/sockets";	my $UnixSocketDir= $perlvar{'lonSockDir'};
my $IdleConnections = Stack->new(); # Set of idle connections	my $IdleConnections = Stack->new(); # Set of idle connections
my %ActiveConnections; # Connections to the remote lond.	my %ActiveConnections; # Connections to the remote lond.
my %ActiveTransactions; # LondTransactions in flight.	my %ActiveTransactions; # LondTransactions in flight.
Line 125 my $WorkQueue = Queue->new(); # Qu	Line 99 my $WorkQueue = Queue->new(); # Qu
my $ConnectionCount = 0;	my $ConnectionCount = 0;
my $IdleSeconds = 0; # Number of seconds idle.	my $IdleSeconds = 0; # Number of seconds idle.
my $Status = ""; # Current status string.	my $Status = ""; # Current status string.
my $ConnectionRetries=5; # Number of connection retries allowed.	my $RecentLogEntry = "";
my $ConnectionRetriesLeft=5; # Number of connection retries remaining.	my $ConnectionRetries=2; # Number of connection retries allowed.
	my $ConnectionRetriesLeft=2; # Number of connection retries remaining.
	my $LondVersion = "unknown"; # Version of lond we talk with.
	my $KeyMode = ""; # e.g. ssl, local, insecure from last connect.
	my $LondConnecting = 0; # True when a connection is being built.

	my $DieWhenIdle = 0; # When true children die when trimmed -> 0.

#	#
# The hash below gives the HTML format for log messages	# The hash below gives the HTML format for log messages
Line 134 my $ConnectionRetriesLeft=5; # Number of	Line 114 my $ConnectionRetriesLeft=5; # Number of
#	#
my %LogFormats;	my %LogFormats;

$LogFormats{"CRITICAL"} = "<font color=red>CRITICAL: %s</font>";	$LogFormats{"CRITICAL"} = "<font color='red'>CRITICAL: %s</font>";
$LogFormats{"SUCCESS"} = "<font color=green>SUCCESS: %s</font>";	$LogFormats{"SUCCESS"} = "<font color='green'>SUCCESS: %s</font>";
$LogFormats{"INFO"} = "<font color=yellow>INFO: %s</font>";	$LogFormats{"INFO"} = "<font color='yellow'>INFO: %s</font>";
$LogFormats{"WARNING"} = "<font color=blue>WARNING: %s</font>";	$LogFormats{"WARNING"} = "<font color='blue'>WARNING: %s</font>";
$LogFormats{"DEFAULT"} = " %s ";	$LogFormats{"DEFAULT"} = " %s ";


	# UpdateStatus;
	# Update the idle status display to show how many connections
	# are left, retries and other stuff.
	#
	sub UpdateStatus {
	if ($ConnectionRetriesLeft > 0) {
	ShowStatus(GetServerHost()." Connection count: ".$ConnectionCount
	." Retries remaining: ".$ConnectionRetriesLeft
	." ($KeyMode)");
	} else {
	ShowStatus(GetServerHost()." >> DEAD <<");
	}
	}


=pod	=pod

Line 180 host and the time will be formatted into	Line 174 host and the time will be formatted into
=cut	=cut

sub Log {	sub Log {
my $severity = shift;
my $message = shift;	my ($severity, $message) = @_;

if(!$LogFormats{$severity}) {	if(!$LogFormats{$severity}) {
$severity = "DEFAULT";	$severity = "DEFAULT";
}	}
Line 201 sub Log {	Line 195 sub Log {
my $execdir = $perlvar{'lonDaemons'};	my $execdir = $perlvar{'lonDaemons'};
my $fh = IO::File->new(">>$execdir/logs/lonc.log");	my $fh = IO::File->new(">>$execdir/logs/lonc.log");
my $msg = sprintf($finalformat, $message);	my $msg = sprintf($finalformat, $message);
	$RecentLogEntry = $msg;
print $fh $msg;	print $fh $msg;


Line 216 Returns the name of the host that a sock	Line 211 Returns the name of the host that a sock
=cut	=cut

sub GetPeername {	sub GetPeername {
my $connection = shift;
my $AdrFamily = shift;
	my ($connection, $AdrFamily) = @_;

my $peer = $connection->peername();	my $peer = $connection->peername();
my $peerport;	my $peerport;
my $peerip;	my $peerip;
if($AdrFamily == AF_INET) {	if($AdrFamily == AF_INET) {
($peerport, $peerip) = sockaddr_in($peer);	($peerport, $peerip) = sockaddr_in($peer);
my $peername = gethostbyaddr($iaddr, $AdrFamily);	my $peername = gethostbyaddr($peerip, $AdrFamily);
return $peername;	return $peername;
} elsif ($AdrFamily == AF_UNIX) {	} elsif ($AdrFamily == AF_UNIX) {
my $peerfile;	my $peerfile;
Line 231 sub GetPeername {	Line 228 sub GetPeername {
return $peerfile;	return $peerfile;
}	}
}	}
#----------------------------- Timer management ------------------------
=pod	=pod

=head2 Debug	=head2 Debug
Line 241 Invoked to issue a debug message.	Line 237 Invoked to issue a debug message.
=cut	=cut

sub Debug {	sub Debug {
my $level = shift;
my $message = shift;	my ($level, $message) = @_;

if ($level <= $DebugLevel) {	if ($level <= $DebugLevel) {
print $message." host = ".$RemoteHost."\n";	Log("INFO", "-Debug- $message host = $RemoteHost");
}	}
}	}

sub SocketDump {	sub SocketDump {
my $level = shift;
my $socket= shift;	my ($level, $socket) = @_;

if($level <= $DebugLevel) {	if($level <= $DebugLevel) {
$socket->Dump();	$socket->Dump(-1); # Ensure it will get dumped.
}	}
}	}

Line 274 sub ShowStatus {	Line 272 sub ShowStatus {

=pod	=pod

	=head 2 SocketTimeout

	Called when an action on the socket times out. The socket is
	destroyed and any active transaction is failed.


	=cut
	sub SocketTimeout {
	my $Socket = shift;
	Log("WARNING", "A socket timeout was detected");
	Debug(5, " SocketTimeout called: ");
	$Socket->Dump(0);
	if(exists($ActiveTransactions{$Socket})) {
	FailTransaction($ActiveTransactions{$Socket});
	}
	KillSocket($Socket); # A transaction timeout also counts as
	# a connection failure:
	$ConnectionRetriesLeft--;
	if($ConnectionRetriesLeft <= 0) {
	Log("CRITICAL", "Host marked DEAD: ".GetServerHost());
	$LondConnecting = 0;
	}

	}
	#----------------------------- Timer management ------------------------

	=pod

=head2 Tick	=head2 Tick

Invoked each timer tick.	Invoked each timer tick.
Line 282 Invoked each timer tick.	Line 308 Invoked each timer tick.


sub Tick {	sub Tick {
	my ($Event) = @_;
	my $clock_watcher = $Event->w;

my $client;	my $client;
ShowStatus(GetServerHost()." Connection count: ".$ConnectionCount);	UpdateStatus();
Debug(6, "Tick");
Debug(6, " Current connection count: ".$ConnectionCount);
foreach $client (keys %ActiveClients) {
Debug(7, " Have client: with id: ".$ActiveClients{$client});
}
# Is it time to prune connection count:	# Is it time to prune connection count:


Line 296 sub Tick {	Line 321 sub Tick {
($WorkQueue->Count() == 0)) { # Idle connections and nothing to do?	($WorkQueue->Count() == 0)) { # Idle connections and nothing to do?
$IdleSeconds++;	$IdleSeconds++;
if($IdleSeconds > $IdleTimeout) { # Prune a connection...	if($IdleSeconds > $IdleTimeout) { # Prune a connection...
$Socket = $IdleConnections->pop();	my $Socket = $IdleConnections->pop();
KillSocket($Socket);	KillSocket($Socket);
	$IdleSeconds = 0; # Otherwise all connections get trimmed to fast.
	UpdateStatus();
	if(($ConnectionCount == 0) && $DieWhenIdle) {
	#
	# Create a lock file since there will be a time window
	# between our exit and the parent's picking up the listen
	# during which no listens will be done on the
	# lonnet client socket.
	#
	my $lock_file = GetLoncSocketPath().".lock";
	open(LOCK,">$lock_file");
	print LOCK "Contents not important";
	close(LOCK);

	exit(0);
	}
}	}
} else {	} else {
$IdleSeconds = 0; # Reset idle count if not idle.	$IdleSeconds = 0; # Reset idle count if not idle.
}	}
	#
	# For each inflight transaction, tick down its timeout counter.
	#

	foreach my $item (keys %ActiveConnections) {
	my $State = $ActiveConnections{$item}->data->GetState();
	if ($State ne 'Idle') {
	Debug(5,"Ticking Socket $State $item");
	$ActiveConnections{$item}->data->Tick();
	}
	}
# Do we have work in the queue, but no connections to service them?	# Do we have work in the queue, but no connections to service them?
# If so, try to make some new connections to get things going again.	# If so, try to make some new connections to get things going again.
#	#
	# Note this code is dead now...
	#
my $Requests = $WorkQueue->Count();	my $Requests = $WorkQueue->Count();
if (($ConnectionCount == 0) && ($Requests > 0)) {	if (($ConnectionCount == 0) && ($Requests > 0) && (!$LondConnecting)) {
if ($ConnectionRetriesLeft > 0) {	if ($ConnectionRetriesLeft > 0) {
my $Connections = ($Requests <= $MaxConnectionCount) ?	Debug(5,"Work but no connections, Make a new one");
$Requests : $MaxConnectionCount;	my $success;
Debug(1,"Work but no connections, start ".$Connections." of them");	$success = &MakeLondConnection;
for ($i =0; $i < $Connections; $i++) {	if($success == 0) { # All connections failed:
MakeLondConnection();	Debug(5,"Work in queue failed to make any connectiouns\n");
	EmptyQueue(); # Fail pending transactions with con_lost.
	CloseAllLondConnections(); # Should all be closed but....
}	}
} else {	} else {
Debug(1,"Work in queue, but gave up on connections..flushing\n");	$LondConnecting = 0;
	ShowStatus(GetServerHost()." >>> DEAD!!! <<<");
	Debug(5,"Work in queue, but gave up on connections..flushing\n");
EmptyQueue(); # Connections can't be established.	EmptyQueue(); # Connections can't be established.
	CloseAllLondConnections(); # Should all already be closed but...
}	}

}	}
	if ($ConnectionCount == 0) {
	$KeyMode = "";
	$clock_watcher->cancel();
	}
}	}

=pod	=pod
Line 342 Trigger disconnections of idle sockets.	Line 403 Trigger disconnections of idle sockets.

sub SetupTimer {	sub SetupTimer {
Debug(6, "SetupTimer");	Debug(6, "SetupTimer");
Event->timer(interval => 1, debug => 1, cb => \&Tick );	Event->timer(interval => 1, cb => \&Tick );
}	}

=pod	=pod
Line 362 long enough, it will be shut down and re	Line 423 long enough, it will be shut down and re

sub ServerToIdle {	sub ServerToIdle {
my $Socket = shift; # Get the socket.	my $Socket = shift; # Get the socket.
	$KeyMode = $Socket->{AuthenticationMode};
delete($ActiveTransactions{$Socket}); # Server has no transaction	delete($ActiveTransactions{$Socket}); # Server has no transaction

&Debug(6, "Server to idle");	&Debug(5, "Server to idle");

# If there's work to do, start the transaction:	# If there's work to do, start the transaction:

$reqdata = $WorkQueue->dequeue(); # This is a LondTransaction	my $reqdata = $WorkQueue->dequeue(); # This is a LondTransaction
unless($reqdata eq undef) {	if ($reqdata ne undef) {
Debug(9, "Queue gave request data: ".$reqdata->getRequest());	Debug(5, "Queue gave request data: ".$reqdata->getRequest());
&StartRequest($Socket, $reqdata);	&StartRequest($Socket, $reqdata);

} else {	} else {

# There's no work waiting, so push the server to idle list.	# There's no work waiting, so push the server to idle list.
&Debug(8, "No new work requests, server connection going idle");	&Debug(5, "No new work requests, server connection going idle");
$IdleConnections->push($Socket);	$IdleConnections->push($Socket);
}	}
}	}
Line 422 sub ClientWritable {	Line 484 sub ClientWritable {
# request.	# request.

&Debug(9,"Send result is ".$result." Defined: ".defined($result));	&Debug(9,"Send result is ".$result." Defined: ".defined($result));
if(defined($result)) {	if($result ne undef) {
&Debug(9, "send result was defined");	&Debug(9, "send result was defined");
if($result == length($Data)) { # Entire string sent.	if($result == length($Data)) { # Entire string sent.
&Debug(9, "ClientWritable data all written");	&Debug(9, "ClientWritable data all written");
Line 440 sub ClientWritable {	Line 502 sub ClientWritable {

} else { # Partial string sent.	} else { # Partial string sent.
$Watcher->data(substr($Data, $result));	$Watcher->data(substr($Data, $result));
	if($result == 0) { # client hung up on us!!
	# Log("INFO", "lonc pipe client hung up on us!");
	$Watcher->cancel;
	$Socket->shutdown(2);
	$Socket->close();
	}
}	}

} else { # Error of some sort...	} else { # Error of some sort...
Line 487 The transaction that is being completed.	Line 555 The transaction that is being completed.
=cut	=cut

sub CompleteTransaction {	sub CompleteTransaction {
&Debug(6,"Complete transaction");	&Debug(5,"Complete transaction");
my $Socket = shift;
my $Transaction = shift;	my ($Socket, $Transaction) = @_;

if (!$Transaction->isDeferred()) { # Normal transaction	if (!$Transaction->isDeferred()) { # Normal transaction
my $data = $Socket->GetReply(); # Data to send.	my $data = $Socket->GetReply(); # Data to send.
	if($LogTransactions) {
	Log("SUCCESS", "Reply from lond: '$data'");
	}
StartClientReply($Transaction, $data);	StartClientReply($Transaction, $data);
} else { # Delete deferred transaction file.	} else { # Delete deferred transaction file.
Log("SUCCESS", "A delayed transaction was completed");	Log("SUCCESS", "A delayed transaction was completed");
LogPerm("S:$Client:".$Transaction->getRequest());	LogPerm("S:$Transaction->getClient() :".$Transaction->getRequest());
unlink $Transaction->getFile();	unlink $Transaction->getFile();
}	}
}	}

=pod	=pod

=head1 StartClientReply	=head1 StartClientReply

Initiates a reply to a client where the reply data is a parameter.	Initiates a reply to a client where the reply data is a parameter.
Line 516 sub CompleteTransaction {	Line 589 sub CompleteTransaction {
The data to send to apached client.	The data to send to apached client.

=cut	=cut

sub StartClientReply {	sub StartClientReply {
my $Transaction = shift;
my $data = shift;	my ($Transaction, $data) = @_;

my $Client = $Transaction->getClient();	my $Client = $Transaction->getClient();

&Debug(8," Reply was: ".$data);	&Debug(8," Reply was: ".$data);
my $Serial = $ActiveClients{$Client};	my $Serial = $ActiveClients{$Client};
my $desc = sprintf("Connection to lonc client %d",	my $desc = sprintf("Connection to lonc client %d",

$Serial);	$Serial);
Event->io(fd => $Client,	Event->io(fd => $Client,
poll => "w",	poll => "w",
Line 533 sub StartClientReply {	Line 606 sub StartClientReply {
cb => \&ClientWritable,	cb => \&ClientWritable,
data => $data);	data => $data);
}	}

=pod	=pod

=head2 FailTransaction	=head2 FailTransaction

Finishes a transaction with failure because the associated lond socket	Finishes a transaction with failure because the associated lond socket
Line 543 sub StartClientReply {	Line 618 sub StartClientReply {
- The transaction is 'live' in which case we initiate the sending	- The transaction is 'live' in which case we initiate the sending
of "con_lost" to the client.	of "con_lost" to the client.

Deleting the transaction means killing it from the	Deleting the transaction means killing it from the %ActiveTransactions hash.
%ActiveTransactions hash.

Parameters:	Parameters:

Line 552 Parameters:	Line 626 Parameters:

The LondTransaction we are failing.	The LondTransaction we are failing.


=cut	=cut

sub FailTransaction {	sub FailTransaction {
my $transaction = shift;	my $transaction = shift;

	# If the socket is dead, that's already logged.

	if ($ConnectionRetriesLeft > 0) {
	Log("WARNING", "Failing transaction "
	.$transaction->getRequest());
	}
Debug(1, "Failing transaction: ".$transaction->getRequest());	Debug(1, "Failing transaction: ".$transaction->getRequest());
if (!$transaction->isDeferred()) { # If the transaction is deferred we'll get to it.	if (!$transaction->isDeferred()) { # If the transaction is deferred we'll get to it.
my $client = $transcation->getClient();	my $client = $transaction->getClient();
Debug(1," Replying con_lost to ".$transaction->getRequest());	Debug(1," Replying con_lost to ".$transaction->getRequest());
StartClientReply($client, "con_lost\n");	StartClientReply($transaction, "con_lost\n");
}	}

}	}
Line 573 sub FailTransaction {	Line 655 sub FailTransaction {

=cut	=cut
sub EmptyQueue {	sub EmptyQueue {
	$ConnectionRetriesLeft--; # Counts as connection failure too.
while($WorkQueue->Count()) {	while($WorkQueue->Count()) {
my $request = $WorkQueue->dequeue(); # This is a transaction	my $request = $WorkQueue->dequeue(); # This is a transaction
FailTransaction($request);	FailTransaction($request);
Line 587 Close all connections open on lond prior	Line 670 Close all connections open on lond prior

=cut	=cut
sub CloseAllLondConnections {	sub CloseAllLondConnections {
foreach $Socket (keys %ActiveConnections) {	foreach my $Socket (keys %ActiveConnections) {
KillSocket($Socket);	if(exists($ActiveTransactions{$Socket})) {
	FailTransaction($ActiveTransactions{$Socket});
	}
	KillSocket($Socket);
}	}
}	}
=cut	=cut
Line 618 nonzero if we are allowed to create a ne	Line 704 nonzero if we are allowed to create a ne
sub KillSocket {	sub KillSocket {
my $Socket = shift;	my $Socket = shift;

	Log("WARNING", "Shutting down a socket");
$Socket->Shutdown();	$Socket->Shutdown();

# If the socket came from the active connection set,	# If the socket came from the active connection set,
Line 631 sub KillSocket {	Line 718 sub KillSocket {
}	}
if(exists($ActiveConnections{$Socket})) {	if(exists($ActiveConnections{$Socket})) {
delete($ActiveConnections{$Socket});	delete($ActiveConnections{$Socket});
	$ConnectionCount--;
	if ($ConnectionCount < 0) { $ConnectionCount = 0; }
}	}
$ConnectionCount--;

# If the connection count has gone to zero and there is work in the	# If the connection count has gone to zero and there is work in the
# work queue, the work all gets failed with con_lost.	# work queue, the work all gets failed with con_lost.
#	#
if($ConnectionCount == 0) {	if($ConnectionCount == 0) {
EmptyQueue;	EmptyQueue();
	CloseAllLondConnections; # Should all already be closed but...
}	}
}	}

Line 664 The connection must echo the challenge b	Line 752 The connection must echo the challenge b
The challenge has been replied to. The we are receiveing the	The challenge has been replied to. The we are receiveing the
'ok' from the partner.	'ok' from the partner.

	=head3 State=ReadingVersionString

	We have requested the lond version and are reading the
	version back. Upon completion, we'll store the version away
	for future use(?).

	=head3 State=HostSet

	We have selected the domain name of our peer (multhomed hosts)
	and are getting the reply (presumably ok) back.

=head3 State=RequestingKey	=head3 State=RequestingKey

The ok has been received and we need to send the request for	The ok has been received and we need to send the request for
Line 707 sub LondReadable {	Line 806 sub LondReadable {
my $Socket = $Watcher->data;	my $Socket = $Watcher->data;
my $client = undef;	my $client = undef;

&Debug(6,"LondReadable called state = ".$State);	&Debug(6,"LondReadable called state = ".$Socket->GetState());


my $State = $Socket->GetState(); # All action depends on the state.	my $State = $Socket->GetState(); # All action depends on the state.

SocketDump(6, $Socket);	SocketDump(6, $Socket);
	my $status = $Socket->Readable();

if($Socket->Readable() != 0) {	&Debug(2, "Socket->Readable returned: $status");
# bad return from socket read. Currently this means that
	if($status != 0) {
	# bad return from socket read. Currently this means that
# The socket has become disconnected. We fail the transaction.	# The socket has become disconnected. We fail the transaction.

	Log("WARNING",
	"Lond connection lost.");
if(exists($ActiveTransactions{$Socket})) {	if(exists($ActiveTransactions{$Socket})) {
Debug(3,"Lond connection lost failing transaction");
FailTransaction($ActiveTransactions{$Socket});	FailTransaction($ActiveTransactions{$Socket});
	} else {
	# Socket is connecting and failed... need to mark
	# no longer connecting.

	$LondConnecting = 0;
}	}
$Watcher->cancel();	$Watcher->cancel();
KillSocket($Socket);	KillSocket($Socket);
	$ConnectionRetriesLeft--; # Counts as connection failure
return;	return;
}	}
SocketDump(6,$Socket);	SocketDump(6,$Socket);
Line 731 sub LondReadable {	Line 840 sub LondReadable {
$State = $Socket->GetState(); # Update in case of transition.	$State = $Socket->GetState(); # Update in case of transition.
&Debug(6, "After read, state is ".$State);	&Debug(6, "After read, state is ".$State);

if($State eq "Initialized") {	if($State eq "Initialized") {


} elsif ($State eq "ChallengeReceived") {	} elsif ($State eq "ChallengeReceived") {
# The challenge must be echoed back; The state machine	# The challenge must be echoed back; The state machine
# in the connection takes care of setting that up. Just	# in the connection takes care of setting that up. Just
# need to transition to writable:	# need to transition to writable:

$Watcher->cb(\&LondWritable);	$Watcher->cb(\&LondWritable);
$Watcher->poll("w");	$Watcher->poll("w");

} elsif ($State eq "ChallengeReplied") {	} elsif ($State eq "ChallengeReplied") {

	} elsif ($State eq "RequestingVersion") {
	# Need to ask for the version... that is writiability:

	$Watcher->cb(\&LondWritable);
	$Watcher->poll("w");

	} elsif ($State eq "ReadingVersionString") {
	# Read the rest of the version string...
	} elsif ($State eq "SetHost") {
	# Need to request the actual domain get set...

	$Watcher->cb(\&LondWritable);
	$Watcher->poll("w");
	} elsif ($State eq "HostSet") {
	# Reading the 'ok' from the peer.

} elsif ($State eq "RequestingKey") {	} elsif ($State eq "RequestingKey") {
# The ok was received. Now we need to request the key	# The ok was received. Now we need to request the key
Line 755 sub LondReadable {	Line 879 sub LondReadable {
} elsif ($State eq "ReceivingKey") {	} elsif ($State eq "ReceivingKey") {

} elsif ($State eq "Idle") {	} elsif ($State eq "Idle") {

	# This is as good a spot as any to get the peer version
	# string:

	if($LondVersion eq "unknown") {
	$LondVersion = $Socket->PeerVersion();
	Log("INFO", "Connected to lond version: $LondVersion");
	}
# If necessary, complete a transaction and then go into the	# If necessary, complete a transaction and then go into the
# idle queue.	# idle queue.
	# Note that a trasition to idle indicates a live lond
	# on the other end so reset the connection retries.
	#
	$ConnectionRetriesLeft = $ConnectionRetries; # success resets the count
$Watcher->cancel();	$Watcher->cancel();
if(exists($ActiveTransactions{$Socket})) {	if(exists($ActiveTransactions{$Socket})) {
Debug(8,"Completing transaction!!");	Debug(5,"Completing transaction!!");
CompleteTransaction($Socket,	CompleteTransaction($Socket,
$ActiveTransactions{$Socket});	$ActiveTransactions{$Socket});
} else {	} else {
Line 767 sub LondReadable {	Line 903 sub LondReadable {
.$RemoteHost." now ready for action");	.$RemoteHost." now ready for action");
}	}
ServerToIdle($Socket); # Next work unit or idle.	ServerToIdle($Socket); # Next work unit or idle.

	#
	$LondConnecting = 0; # Best spot I can think of for this.
	#

} elsif ($State eq "SendingRequest") {	} elsif ($State eq "SendingRequest") {
# We need to be writable for this and probably don't belong	# We need to be writable for this and probably don't belong
Line 780 sub LondReadable {	Line 920 sub LondReadable {


} else {	} else {
# Invalid state.	# Invalid state.
Debug(4, "Invalid state in LondReadable");	Debug(4, "Invalid state in LondReadable");
}	}
}	}
Line 862 sub LondWritable {	Line 1002 sub LondWritable {

SocketDump(6,$Socket);	SocketDump(6,$Socket);

	# If the socket is writable, we must always write.
	# Only by writing will we undergo state transitions.
	# Old logic wrote in state specific code below, however
	# That forces us at least through another invocation of
	# this function after writability is possible again.
	# This logic also factors out common code for handling
	# write failures... in all cases, write failures
	# Kill the socket.
	# This logic makes the branches of the >big< if below
	# so that the writing states are actually NO-OPs.

	if ($Socket->Writable() != 0) {
	# The write resulted in an error.
	# We'll treat this as if the socket got disconnected:
	Log("WARNING", "Connection to ".$RemoteHost.
	" has been disconnected");
	if(exists($ActiveTransactions{$Socket})) {
	FailTransaction($ActiveTransactions{$Socket});
	} else {
	# In the process of conneting, so need to turn that off.

	$LondConnecting = 0;
	}
	$Watcher->cancel();
	KillSocket($Socket);
	return;
	}



if ($State eq "Connected") {	if ($State eq "Connected") {

if ($Socket->Writable() != 0) {
# The write resulted in an error.
# We'll treat this as if the socket got disconnected:
Log("WARNING", "Connection to ".$RemoteHost.
" has been disconnected");
$Watcher->cancel();
KillSocket($Socket);
return;
}
# "init" is being sent...	# "init" is being sent...


} elsif ($State eq "Initialized") {	} elsif ($State eq "Initialized") {

# Now that init was sent, we switch	# Now that init was sent, we switch
Line 883 sub LondWritable {	Line 1043 sub LondWritable {

$Watcher->cb(\&LondReadable);	$Watcher->cb(\&LondReadable);
$Watcher->poll("r");	$Watcher->poll("r");

} elsif ($State eq "ChallengeReceived") {	} elsif ($State eq "ChallengeReceived") {
# We received the challenge, now we	# We received the challenge, now we
# are echoing it back. This is a no-op,	# are echoing it back. This is a no-op,
# we're waiting for the state to change	# we're waiting for the state to change

if($Socket->Writable() != 0) {

$Watcher->cancel();
KillSocket($Socket);
return;
}

} elsif ($State eq "ChallengeReplied") {	} elsif ($State eq "ChallengeReplied") {
# The echo was sent back, so we switch	# The echo was sent back, so we switch
# to watching readability.	# to watching readability.

$Watcher->cb(\&LondReadable);	$Watcher->cb(\&LondReadable);
$Watcher->poll("r");	$Watcher->poll("r");
	} elsif ($State eq "RequestingVersion") {
	# Sending the peer a version request...

	} elsif ($State eq "ReadingVersionString") {
	# Transition to read since we have sent the
	# version command and now just need to read the
	# version string from the peer:

	$Watcher->cb(\&LondReadable);
	$Watcher->poll("r");

	} elsif ($State eq "SetHost") {
	# Setting the remote domain...

	} elsif ($State eq "HostSet") {
	# Back to readable to get the ok.

	$Watcher->cb(\&LondReadable);
	$Watcher->poll("r");


} elsif ($State eq "RequestingKey") {	} elsif ($State eq "RequestingKey") {
# At this time we're requesting the key.	# At this time we're requesting the key.
# again, this is essentially a no-op.	# again, this is essentially a no-op.
# we'll write the next chunk until the
# state changes.

if($Socket->Writable() != 0) {
# Write resulted in an error.

$Watcher->cancel();
KillSocket($Socket);
return;

}
} elsif ($State eq "ReceivingKey") {	} elsif ($State eq "ReceivingKey") {
# Now we need to wait for the key	# Now we need to wait for the key
# to come back from the peer:	# to come back from the peer:
Line 925 sub LondWritable {	Line 1088 sub LondWritable {
$Watcher->poll("r");	$Watcher->poll("r");

} elsif ($State eq "SendingRequest") {	} elsif ($State eq "SendingRequest") {

# At this time we are sending a request to the	# At this time we are sending a request to the
# peer... write the next chunk:	# peer... write the next chunk:

if($Socket->Writable() != 0) {

if(exists($ActiveTransactions{$Socket})) {
Debug(3, "Lond connection lost, failing transactions");
FailTransaction($ActiveTransactions{$Socket});
}
$Watcher->cancel();
KillSocket($Socket);
return;

}

} elsif ($State eq "ReceivingReply") {	} elsif ($State eq "ReceivingReply") {
# The send has completed. Wait for the	# The send has completed. Wait for the
Line 968 sub QueueDelayed {	Line 1121 sub QueueDelayed {
Debug(4, "Delayed path: ".$path);	Debug(4, "Delayed path: ".$path);
opendir(DIRHANDLE, $path);	opendir(DIRHANDLE, $path);

@alldelayed = grep /\.$RemoteHost$/, readdir DIRHANDLE;	my @alldelayed = grep /\.$RemoteHost$/, readdir DIRHANDLE;
Debug(4, "Got ".$alldelayed." delayed files");
closedir(DIRHANDLE);	closedir(DIRHANDLE);
my $dfname;	my $dfname;
my $reqfile;	my $reqfile;
Line 979 sub QueueDelayed {	Line 1131 sub QueueDelayed {
my $Handle = IO::File->new($reqfile);	my $Handle = IO::File->new($reqfile);
my $cmd = <$Handle>;	my $cmd = <$Handle>;
chomp $cmd; # There may or may not be a newline...	chomp $cmd; # There may or may not be a newline...
$cmd = $cmd."\ny"; # now for sure there's exactly one newline.	$cmd = $cmd."\n"; # now for sure there's exactly one newline.
my $Transaction = LondTransaction->new($cmd);	my $Transaction = LondTransaction->new($cmd);
$Transaction->SetDeferred($reqfile);	$Transaction->SetDeferred($reqfile);
QueueTransaction($Transaction);	QueueTransaction($Transaction);
Line 1006 sub MakeLondConnection {	Line 1158 sub MakeLondConnection {
my $Connection = LondConnection->new(&GetServerHost(),	my $Connection = LondConnection->new(&GetServerHost(),
&GetServerPort());	&GetServerPort());

if($Connection == undef) { # Needs to be more robust later.	if($Connection eq undef) { # Needs to be more robust later.
Log("CRITICAL","Failed to make a connection with lond.");	Log("CRITICAL","Failed to make a connection with lond.");
$ConnectionRetriesLeft--;	$ConnectionRetriesLeft--;
return 0; # Failure.	return 0; # Failure.
} else {	} else {
$ConnectionRetriesLeft = $ConnectionRetries; # success resets the count
# The connection needs to have writability	# The connection needs to have writability
# monitored in order to send the init sequence	# monitored in order to send the init sequence
# that starts the whole authentication/key	# that starts the whole authentication/key
# exchange underway.	# exchange underway.
#	#
my $Socket = $Connection->GetSocket();	my $Socket = $Connection->GetSocket();
if($Socket == undef) {	if($Socket eq undef) {
die "did not get a socket from the connection";	die "did not get a socket from the connection";
} else {	} else {
&Debug(9,"MakeLondConnection got socket: ".$Socket);	&Debug(9,"MakeLondConnection got socket: ".$Socket);
}	}

	$Connection->SetTimeoutCallback(\&SocketTimeout);
$event = Event->io(fd => $Socket,
	my $event = Event->io(fd => $Socket,
poll => 'w',	poll => 'w',
cb => \&LondWritable,	cb => \&LondWritable,
data => $Connection,	data => $Connection,
desc => 'Connection to lond server');	desc => 'Connection to lond server');
$ActiveConnections{$Connection} = $event;	$ActiveConnections{$Connection} = $event;
	if ($ConnectionCount == 0) {
	&SetupTimer; # Need to handle timeouts with connections...
	}
$ConnectionCount++;	$ConnectionCount++;
Debug(4, "Connection count = ".$ConnectionCount);	Debug(4, "Connection count = ".$ConnectionCount);
if($ConnectionCount == 1) { # First Connection:	if($ConnectionCount == 1) { # First Connection:
Line 1039 sub MakeLondConnection {	Line 1194 sub MakeLondConnection {
}	}
Log("SUCESS", "Created connection ".$ConnectionCount	Log("SUCESS", "Created connection ".$ConnectionCount
." to host ".GetServerHost());	." to host ".GetServerHost());
	$LondConnecting = 1; # Connection in progress.
return 1; # Return success.	return 1; # Return success.
}	}

Line 1070 The text of the request to send.	Line 1226 The text of the request to send.
=cut	=cut

sub StartRequest {	sub StartRequest {
my $Lond = shift;
my $Request = shift; # This is a LondTransaction.	my ($Lond, $Request) = @_;

Debug(6, "StartRequest: ".$Request->getRequest());	Debug(6, "StartRequest: ".$Request->getRequest());

Line 1081 sub StartRequest {	Line 1237 sub StartRequest {
$ActiveTransactions{$Lond} = $Request;	$ActiveTransactions{$Lond} = $Request;

$Lond->InitiateTransaction($Request->getRequest());	$Lond->InitiateTransaction($Request->getRequest());
$event = Event->io(fd => $Socket,	my $event = Event->io(fd => $Socket,
poll => "w",	poll => "w",
cb => \&LondWritable,	cb => \&LondWritable,
data => $Lond,	data => $Lond,
Line 1120 sub QueueTransaction {	Line 1276 sub QueueTransaction {

my $LondSocket = $IdleConnections->pop();	my $LondSocket = $IdleConnections->pop();
if(!defined $LondSocket) { # Need to queue request.	if(!defined $LondSocket) { # Need to queue request.
Debug(8,"Must queue...");	Debug(5,"Must queue...");
$WorkQueue->enqueue($requestData);	$WorkQueue->enqueue($requestData);
if($ConnectionCount < $MaxConnectionCount) {	Debug(5, "Queue Transaction startnew $ConnectionCount $LondConnecting");
Debug(4,"Starting additional lond connection");	if(($ConnectionCount < $MaxConnectionCount) && (! $LondConnecting)) {
MakeLondConnection();
	if($ConnectionRetriesLeft > 0) {
	Debug(5,"Starting additional lond connection");
	if(&MakeLondConnection() == 0) {
	EmptyQueue(); # Fail transactions, can't make connection.
	CloseAllLondConnections; # Should all be closed but...
	}
	} else {
	ShowStatus(GetServerHost()." >>> DEAD !!!! <<<");
	$LondConnecting = 0;
	EmptyQueue(); # It's worse than that ... he's dead Jim.
	CloseAllLondConnections; # Should all be closed but..
	}
}	}
} else { # Can start the request:	} else { # Can start the request:
Debug(8,"Can start...");	Debug(8,"Can start...");
Line 1155 sub ClientRequest {	Line 1323 sub ClientRequest {
my $rv = $socket->recv($thisread, POSIX::BUFSIZ, 0);	my $rv = $socket->recv($thisread, POSIX::BUFSIZ, 0);
Debug(8, "rcv: data length = ".length($thisread)	Debug(8, "rcv: data length = ".length($thisread)
." read =".$thisread);	." read =".$thisread);
unless (defined $rv && length($thisread)) {	unless (defined $rv && length($thisread)) {
# Likely eof on socket.	# Likely eof on socket.
Debug(5,"Client Socket closed on lonc for ".$RemoteHost);	Debug(5,"Client Socket closed on lonc for ".$RemoteHost);
close($socket);	close($socket);
Line 1166 sub ClientRequest {	Line 1334 sub ClientRequest {
Debug(8,"Data: ".$data." this read: ".$thisread);	Debug(8,"Data: ".$data." this read: ".$thisread);
$data = $data.$thisread; # Append new data.	$data = $data.$thisread; # Append new data.
$watcher->data($data);	$watcher->data($data);
if($data =~ /(.*\n)/) { # Request entirely read.	if($data =~ /\n$/) { # Request entirely read.
if($data eq "close_connection_exit\n") {	if($data eq "close_connection_exit\n") {
Log("CRITICAL",	Log("CRITICAL",
"Request Close Connection ... exiting");	"Request Close Connection ... exiting");
Line 1174 sub ClientRequest {	Line 1342 sub ClientRequest {
exit;	exit;
}	}
Debug(8, "Complete transaction received: ".$data);	Debug(8, "Complete transaction received: ".$data);
	if($LogTransactions) {
	Log("SUCCESS", "Transaction: '$data'"); # Transaction has \n.
	}
my $Transaction = LondTransaction->new($data);	my $Transaction = LondTransaction->new($data);
$Transaction->SetClient($socket);	$Transaction->SetClient($socket);
QueueTransaction($Transaction);	QueueTransaction($Transaction);
Line 1272 sub SetupLoncListener {	Line 1443 sub SetupLoncListener {
my $SocketName = GetLoncSocketPath();	my $SocketName = GetLoncSocketPath();
unlink($SocketName);	unlink($SocketName);
unless ($socket =IO::Socket::UNIX->new(Local => $SocketName,	unless ($socket =IO::Socket::UNIX->new(Local => $SocketName,
Listen => 10,	Listen => 250,
Type => SOCK_STREAM)) {	Type => SOCK_STREAM)) {
die "Failed to create a lonc listner socket";	die "Failed to create a lonc listner socket";
}	}
Line 1282 sub SetupLoncListener {	Line 1453 sub SetupLoncListener {
fd => $socket);	fd => $socket);
}	}

	#
	# Toggle transaction logging.
	# Implicit inputs:
	# LogTransactions
	# Implicit Outputs:
	# LogTransactions
	sub ToggleTransactionLogging {
	print STDERR "Toggle transaction logging...\n";
	if(!$LogTransactions) {
	$LogTransactions = 1;
	} else {
	$LogTransactions = 0;
	}


	Log("SUCCESS", "Toggled transaction logging: $LogTransactions \n");
	}

	=pod

	=head2 ChildStatus

	Child USR1 signal handler to report the most recent status
	into the status file.

	We also use this to reset the retries count in order to allow the
	client to retry connections with a previously dead server.
	=cut

	sub ChildStatus {
	my $event = shift;
	my $watcher = $event->w;

	Debug(2, "Reporting child status because : ".$watcher->data);
	my $docdir = $perlvar{'lonDocRoot'};
	my $fh = IO::File->new(">>$docdir/lon-status/loncstatus.txt");
	print $fh $$."\t".$RemoteHost."\t".$Status."\t".
	$RecentLogEntry."\n";
	#
	# Write out information about each of the connections:
	#
	if ($DebugLevel > 2) {
	print $fh "Active connection statuses: \n";
	my $i = 1;
	print STDERR "================================= Socket Status Dump:\n";
	foreach my $item (keys %ActiveConnections) {
	my $Socket = $ActiveConnections{$item}->data;
	my $state = $Socket->GetState();
	print $fh "Connection $i State: $state\n";
	print STDERR "---------------------- Connection $i \n";
	$Socket->Dump(-1); # Ensure it gets dumped..
	$i++;
	}
	}
	$ConnectionRetriesLeft = $ConnectionRetries;
	}

=pod	=pod

=head2 SignalledToDeath	=head2 SignalledToDeath
Line 1290 Called in response to a signal that caus	Line 1518 Called in response to a signal that caus

=cut	=cut

=pod

sub SignalledToDeath {	sub SignalledToDeath {
my ($signal) = @_;	my $event = shift;
	my $watcher= $event->w;

	Debug(2,"Signalled to death! via ".$watcher->data);
	my ($signal) = $watcher->data;
chomp($signal);	chomp($signal);
Log("CRITICAL", "Abnormal exit. Child $$ for $RemoteHost "	Log("CRITICAL", "Abnormal exit. Child $$ for $RemoteHost "
."died through "."\"$signal\"");	."died through "."\"$signal\"");
LogPerm("F:lonc: $$ on $RemoteHost signalled to death: "	LogPerm("F:lonc: $$ on $RemoteHost signalled to death: "
."\"$signal\"");	."\"$signal\"");
die("Signal abnormal end");	exit 0;

}	}

	=head2 ToggleDebug

	This sub toggles trace debugging on and off.

	=cut

	sub ToggleDebug {
	my $Current = $DebugLevel;
	$DebugLevel = $NextDebugLevel;
	$NextDebugLevel = $Current;

	Log("SUCCESS", "New debugging level for $RemoteHost now $DebugLevel");

	}

=head2 ChildProcess	=head2 ChildProcess

This sub implements a child process for a single lonc daemon.	This sub implements a child process for a single lonc daemon.
Line 1311 This sub implements a child process for	Line 1558 This sub implements a child process for
sub ChildProcess {	sub ChildProcess {


# For now turn off signals.	#
	# Signals must be handled by the Event framework...
$SIG{QUIT} = \&SignalledToDeath;	#
$SIG{HUP} = IGNORE;
$SIG{USR1} = IGNORE;	Event->signal(signal => "QUIT",
$SIG{INT} = IGNORE;	cb => \&SignalledToDeath,
$SIG{CHLD} = IGNORE;	data => "QUIT");
$SIG{__DIE__} = \&SignalledToDeath;	Event->signal(signal => "HUP",
	cb => \&ChildStatus,
	data => "HUP");
	Event->signal(signal => "USR1",
	cb => \&ChildStatus,
	data => "USR1");
	Event->signal(signal => "USR2",
	cb => \&ToggleTransactionLogging);
	Event->signal(signal => "INT",
	cb => \&ToggleDebug,
	data => "INT");

SetupTimer();

SetupLoncListener();	SetupLoncListener();

Line 1330 sub ChildProcess {	Line 1586 sub ChildProcess {

# Setup the initial server connection:	# Setup the initial server connection:

# &MakeLondConnection(); // let first work requirest do it.	# &MakeLondConnection(); // let first work requirest do it.


Debug(9,"Entering event loop");	Debug(9,"Entering event loop");
Line 1344 sub ChildProcess {	Line 1600 sub ChildProcess {

sub CreateChild {	sub CreateChild {
my $host = shift;	my $host = shift;

	my $sigset = POSIX::SigSet->new(SIGINT);
	sigprocmask(SIG_BLOCK, $sigset);
$RemoteHost = $host;	$RemoteHost = $host;
Log("CRITICAL", "Forking server for ".$host);	Log("CRITICAL", "Forking server for ".$host);
$pid = fork;	my $pid = fork;
if($pid) { # Parent	if($pid) { # Parent
$ChildHash{$pid} = $RemoteHost;	$RemoteHost = "Parent";
	$ChildHash{$pid} = $host;
	$HostToPid{$host}= $pid;
	sigprocmask(SIG_UNBLOCK, $sigset);

} else { # child.	} else { # child.
ShowStatus("Connected to ".$RemoteHost);	ShowStatus("Connected to ".$RemoteHost);
ChildProcess;	$SIG{INT} = 'DEFAULT';
	sigprocmask(SIG_UNBLOCK, $sigset);
	ChildProcess; # Does not return.
}	}

}	}
Line 1386 if ($childpid != 0) {	Line 1651 if ($childpid != 0) {
#	#

ShowStatus("Parent writing pid file:");	ShowStatus("Parent writing pid file:");
$execdir = $perlvar{'lonDaemons'};	my $execdir = $perlvar{'lonDaemons'};
open (PIDSAVE, ">$execdir/logs/lonc.pid");	open (PIDSAVE, ">$execdir/logs/lonc.pid");
print PIDSAVE "$$\n";	print PIDSAVE "$$\n";
close(PIDSAVE);	close(PIDSAVE);



if (POSIX::setsid() < 0) {	if (POSIX::setsid() < 0) {
print "Could not create new session\n";	print "Could not create new session\n";
exit -1;	exit -1;
Line 1400 ShowStatus("Forking node servers");	Line 1667 ShowStatus("Forking node servers");

Log("CRITICAL", "--------------- Starting children ---------------");	Log("CRITICAL", "--------------- Starting children ---------------");

	LondConnection::ReadConfig; # Read standard config files.
my $HostIterator = LondConnection::GetHostIterator;	my $HostIterator = LondConnection::GetHostIterator;
while (! $HostIterator->end()) {	while (! $HostIterator->end()) {

$hostentryref = $HostIterator->get();	my $hostentryref = $HostIterator->get();
CreateChild($hostentryref->[0]);	CreateChild($hostentryref->[0]);
	$HostHash{$hostentryref->[0]} = $hostentryref->[4];
$HostIterator->next();	$HostIterator->next();
}	}
	$RemoteHost = "Parent Server";

# Maintain the population:	# Maintain the population:

Line 1415 ShowStatus("Parent keeping the flock");	Line 1685 ShowStatus("Parent keeping the flock");
#	#
# Set up parent signals:	# Set up parent signals:
#	#
$SIG{INT} = &KillThemAll;
$SIG{TERM} = &KillThemAll;	$SIG{INT} = \&Terminate;
	$SIG{TERM} = \&Terminate;
	$SIG{HUP} = \&Restart;
	$SIG{USR1} = \&CheckKids;
	$SIG{USR2} = \&UpdateKids; # LonManage update request.

while(1) {	while(1) {
$deadchild = wait();	my $deadchild = wait();
if(exists $ChildHash{$deadchild}) { # need to restart.	if(exists $ChildHash{$deadchild}) { # need to restart.
$deadhost = $ChildHash{$deadchild};	my $deadhost = $ChildHash{$deadchild};
	delete($HostToPid{$deadhost});
delete($ChildHash{$deadchild});	delete($ChildHash{$deadchild});
Log("WARNING","Lost child pid= ".$deadchild.	Log("WARNING","Lost child pid= ".$deadchild.
"Connected to host ".$deadhost);	"Connected to host ".$deadhost);
Line 1429 while(1) {	Line 1704 while(1) {
CreateChild($deadhost);	CreateChild($deadhost);
}	}
}	}



	=pod

	=head1 CheckKids

	Since kids do not die as easily in this implementation
	as the previous one, there is no need to restart the
	dead ones (all dead kids get restarted when they die!!)
	The only thing this function does is to pass USR1 to the
	kids so that they report their status.

	=cut

	sub CheckKids {
	Debug(2, "Checking status of children");
	my $docdir = $perlvar{'lonDocRoot'};
	my $fh = IO::File->new(">$docdir/lon-status/loncstatus.txt");
	my $now=time;
	my $local=localtime($now);
	print $fh "LONC status $local - parent $$ \n\n";
	foreach my $pid (keys %ChildHash) {
	Debug(2, "Sending USR1 -> $pid");
	kill 'USR1' => $pid; # Tell Child to report status.
	sleep 1; # Wait so file doesn't intermix.
	}
	}

	=pod

	=head1 UpdateKids

	parent's SIGUSR2 handler. This handler:

	=item

	Rereads the hosts file.

	=item

	Kills off (via sigint) children for hosts that have disappeared.

	=item

	QUITs children for hosts that already exist (this just forces a status display
	and resets the connection retry count for that host.

	=item

	Starts new children for hosts that have been added to the hosts.tab file since
	the start of the master program and maintains them.

	=cut

	sub UpdateKids {

	Log("INFO", "Updating connections via SIGUSR2");

	# Just in case we need to kill our own lonc, we wait a few seconds to
	# give it a chance to receive and relay lond's response to the
	# re-init command.
	#

	sleep(2); # Wait a couple of seconds.

	my %hosts; # Indexed by loncapa hostname, value=ip.

	# Need to re-read the host table:


	LondConnection::ReadConfig();
	my $I = LondConnection::GetHostIterator;
	while (! $I->end()) {
	my $item = $I->get();
	$hosts{$item->[0]} = $item->[4];
	$I->next();
	}

	# The logic below is written for clarity not for efficiency.
	# Since I anticipate that this function is only rarely called, that's
	# appropriate. There are certainly ways to combine the loops below,
	# and anyone wishing to obscure the logic is welcome to go for it.
	# Note that we don't re-direct sigchild. Instead we do what's needed
	# to the data structures that keep track of children to ensure that
	# when sigchild is honored, no new child is born.
	#

	# For each existing child; if it's host doesn't exist, kill the child.

	foreach my $child (keys %ChildHash) {
	my $oldhost = $ChildHash{$child};
	if (!(exists $hosts{$oldhost})) {
	Log("CRITICAL", "Killing child for $oldhost host no longer exists");
	delete $ChildHash{$child};
	delete $HostToPid{$oldhost};
	kill 'QUIT' => $child;
	}
	}
	# For each remaining existing child; if it's host's ip has changed,
	# Restart the child on the new IP.

	foreach my $child (keys %ChildHash) {
	my $oldhost = $ChildHash{$child};
	my $oldip = $HostHash{$oldhost};
	if ($hosts{$oldhost} ne $oldip) {

	# kill the old child.

	Log("CRITICAL", "Killing child for $oldhost host ip has changed...");
	delete $ChildHash{$child};
	delete $HostToPid{$oldhost};
	kill 'QUIT' => $child;

	# Do the book-keeping needed to start a new child on the
	# new ip.

	$HostHash{$oldhost} = $hosts{$oldhost};
	CreateChild($oldhost);
	}
	}
	# Finally, for each new host, not in the host hash, create a
	# enter the host and create a new child.
	# Force a status display of any existing process.

	foreach my $host (keys %hosts) {
	if(!(exists $HostHash{$host})) {
	Log("INFO", "New host $host discovered in hosts.tab...");
	$HostHash{$host} = $hosts{$host};
	CreateChild($host);
	} else {
	kill 'HUP' => $HostToPid{$host}; # status display.
	}
	}
	}


	=pod

	=head1 Restart

	Signal handler for HUP... all children are killed and
	we self restart. This is an el-cheapo way to re read
	the config file.

	=cut

	sub Restart {
	&KillThemAll; # First kill all the children.
	Log("CRITICAL", "Restarting");
	my $execdir = $perlvar{'lonDaemons'};
	unlink("$execdir/logs/lonc.pid");
	exec("$execdir/loncnew");
	}

	=pod

	=head1 KillThemAll

	Signal handler that kills all children by sending them a
	SIGHUP. Responds to sigint and sigterm.

	=cut

sub KillThemAll {	sub KillThemAll {
	Debug(2, "Kill them all!!");
	local($SIG{CHLD}) = 'IGNORE'; # Our children >will< die.
	foreach my $pid (keys %ChildHash) {
	my $serving = $ChildHash{$pid};
	ShowStatus("Nicely Killing lonc for $serving pid = $pid");
	Log("CRITICAL", "Nicely Killing lonc for $serving pid = $pid");
	kill 'QUIT' => $pid;
	}


}	}


	#
	# Kill all children via KILL. Just in case the
	# first shot didn't get them.

	sub really_kill_them_all_dammit
	{
	Debug(2, "Kill them all Dammit");
	local($SIG{CHLD} = 'IGNORE'); # In case some purist reenabled them.
	foreach my $pid (keys %ChildHash) {
	my $serving = $ChildHash{$pid};
	&ShowStatus("Nastily killing lonc for $serving pid = $pid");
	Log("CRITICAL", "Nastily killing lonc for $serving pid = $pid");
	kill 'KILL' => $pid;
	delete($ChildHash{$pid});
	my $execdir = $perlvar{'lonDaemons'};
	unlink("$execdir/logs/lonc.pid");
	}
	}
	=pod

	=head1 Terminate

	Terminate the system.

	=cut

	sub Terminate {
	&Log("CRITICAL", "Asked to kill children.. first be nice...");
	&KillThemAll;
	#
	# By now they really should all be dead.. but just in case
	# send them all SIGKILL's after a bit of waiting:

	sleep(4);
	&Log("CRITICAL", "Now kill children nasty");
	&really_kill_them_all_dammit;
	Log("CRITICAL","Master process exiting");
	exit 0;

	}
	=pod

=head1 Theory	=head1 Theory

The event class is used to build this as a single process with an	The event class is used to build this as a single process with an

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>

Removed from v.1.10
changed lines
	Added in v.1.57