Diff for /loncom/loncnew between versions 1.29 and 1.80

version 1.29, 2003/10/21 14:24:42 version 1.80, 2007/03/28 00:23:46
Line 35 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 1.29  2003/10/21 14:24:42  foxr  
 #    Fix little typo that may explain growth of connections  
 #  
 #    Revision 1.28  2003/10/14 15:36:21  albertel  
 #    - making it easier to run loncnew,  
 #       /etc/init.d/loncontrol startnew  
 #       /etc/init.d/loncontrol restartnew  
 #      will now start loncnew in place of lonc  
 #  
 #    Revision 1.27  2003/10/07 11:23:03  foxr  
 #    Installed and tested code to process reinit in parent server.  
 #  
 #    Revision 1.26  2003/09/30 11:11:17  foxr  
 #    Add book-keeping hashes to support the re-init procedure.  
 #  
 #    Revision 1.25  2003/09/23 11:22:14  foxr  
 #    Tested ability to receive sigusr2  This is now logged and must be  
 #    properly implemented as a re-read of hosts and re-init of appropriate  
 #    children.  
 #  
 #    Revision 1.24  2003/09/16 09:46:42  foxr  
 #    Added skeletal infrastructure to support SIGUSR2 update hosts request.  
 #  
 #    Revision 1.23  2003/09/15 09:24:49  foxr  
 #    Add use strict and fix all the fallout from that.  
 #  
 #    Revision 1.22  2003/09/02 10:34:47  foxr  
 #    - Fix errors in host dead detection logic (too many cases where the  
 #      retries left were not getting incremented or just not checked).  
 #    - Added some additional status to the ps axuww display:  
 #      o Remaining retries on a host.  
 #      o >>> DEAD <<< indicator if I've given up on a host.  
 #    - Tested the SIGHUP will reset the retries remaining count (thanks to  
 #      the above status stuff, and get allow the loncnew to re-try again  
 #      on the host (thanks to the log).  
 #  
 #    Revision 1.21  2003/08/26 09:19:51  foxr  
 #    How embarrassing... put in the SocketTimeout function in loncnew and forgot  
 #    to actually hook it into the LondTransaction.  Added this to MakeLondConnection  
 #    where it belongs... hopefully transactions (not just connection attempts) will  
 #    timeout more speedily than the socket errors will catch it.  
 #  
 #    Revision 1.20  2003/08/25 18:48:11  albertel  
 #    - fixing a forgotten ;  
 #  
 #    Revision 1.19  2003/08/19 09:31:46  foxr  
 #    Get socket directory from configuration rather than the old hard coded test  
 #    way that I forgot to un-hard code.  
 #  
 #    Revision 1.18  2003/08/06 09:52:29  foxr  
 #    Also needed to remember to fail in-flight transactions if their sends fail.  
 #  
 #    Revision 1.17  2003/08/03 00:44:31  foxr  
 #    1. Correct handling of connection failure: Assume it means the host is  
 #       unreachable and fail all of the queued transactions.  Note that the  
 #       inflight transactions should fail on their own time due either to timeout  
 #       or send/receive failures.  
 #    2. Correct handling of logs for forced death signals.  Pull the signal  
 #       from the event watcher.  
 #  
 #    Revision 1.16  2003/07/29 02:33:05  foxr  
 #    Add SIGINT processing to child processes to toggle annoying trace mode  
 #    on/off.. will try to use this to isolate the compute boud process issue.  
 #  
 #    Revision 1.15  2003/07/15 02:07:05  foxr  
 #    Added code for lonc/lond transaction timeouts.  Who knows if it works right.  
 #    The intent is for a timeout to fail any transaction in progress and kill  
 #    off the sockt that timed out.  
 #  
 #    Revision 1.14  2003/07/03 02:10:18  foxr  
 #    Get all of the signals to work correctly.  
 #  
 #    Revision 1.13  2003/07/02 01:31:55  foxr  
 #    Added kill -HUP logic (restart).  
 #  
 #    Revision 1.11  2003/06/25 01:54:44  foxr  
 #    Fix more problems with transaction failure.  
 #  
 #    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 the 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 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 POSIX;
Line 159  use LONCAPA::LondConnection; Line 61  use LONCAPA::LondConnection;
 use LONCAPA::LondTransaction;  use LONCAPA::LondTransaction;
 use LONCAPA::Configuration;  use LONCAPA::Configuration;
 use LONCAPA::HashIterator;  use LONCAPA::HashIterator;
   use Fcntl qw(:flock);
   
 #  
 #   Disable all signals we might receive from outside for now.  
 #  
   
   
 # Read the httpd configuration file to get perl variables  # Read the httpd configuration file to get perl variables
Line 176  my %perlvar    = %{$perlvarref}; Line 74  my %perlvar    = %{$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 %listening_to; # Socket->host table for who the parent
 my %HostHash; # by loncapaname -> IP.                                  # is listening to.
   my %parent_dispatchers;         # host-> listener watcher events. 
   
   my %parent_handlers; # Parent signal handlers...
   
 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 $NextDebugLevel= 2; # So Sigint can toggle this.  my $NextDebugLevel= 2; # So Sigint can toggle this.
 my $IdleTimeout= 3600; # Wait an hour before pruning connections.  my $IdleTimeout= 600; # Wait 10 minutes before pruning connections.
   
   my $LogTransactions = 0; # When True, all transactions/replies get logged.
   my $executable      = $0; # Get the full path to me.
   
 #  #
 #  The variables below are only used by the child processes.  #  The variables below are only used by the child processes.
Line 203  my $Status          = ""; # Current stat Line 106  my $Status          = ""; # Current stat
 my $RecentLogEntry  = "";  my $RecentLogEntry  = "";
 my $ConnectionRetries=5; # Number of connection retries allowed.  my $ConnectionRetries=5; # Number of connection retries allowed.
 my $ConnectionRetriesLeft=5; # Number of connection retries remaining.  my $ConnectionRetriesLeft=5; # 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 $hosts_tab       = 0;        # True if we are using a static hosts.tab
   my $I_am_child      = 0; # True if this is the child process.
   
 #  #
 #   The hash below gives the HTML format for log messages  #   The hash below gives the HTML format for log messages
Line 210  my $ConnectionRetriesLeft=5; # Number of Line 121  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 225  $LogFormats{"DEFAULT"}  = " %s "; Line 150  $LogFormats{"DEFAULT"}  = " %s ";
 Makes an entry into the permanent log file.  Makes an entry into the permanent log file.
   
 =cut  =cut
   
 sub LogPerm {  sub LogPerm {
     my $message=shift;      my $message=shift;
     my $execdir=$perlvar{'lonDaemons'};      my $execdir=$perlvar{'lonDaemons'};
Line 256  host and the time will be formatted into Line 182  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 270  sub Log { Line 196  sub Log {
     my $now   = time;      my $now   = time;
     my $local = localtime($now);      my $local = localtime($now);
     my $finalformat = "$local ($$) [$RemoteHost] [$Status] ";      my $finalformat = "$local ($$) [$RemoteHost] [$Status] ";
     my $finalformat = $finalformat.$format."\n";      $finalformat = $finalformat.$format."\n";
   
     # open the file and put the result.      # open the file and put the result.
   
Line 293  Returns the name of the host that a sock Line 219  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;
Line 308  sub GetPeername { Line 236  sub GetPeername {
  return $peerfile;   return $peerfile;
     }      }
 }  }
 #----------------------------- Timer management ------------------------  
 =pod  =pod
   
 =head2 Debug  =head2 Debug
Line 318  Invoked to issue a debug message. Line 245  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) {
  Log("INFO", "-Debug- $message host = $RemoteHost");   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 341  sub SocketDump { Line 270  sub SocketDump {
  and as what we return in a SIGUSR1   and as what we return in a SIGUSR1
   
 =cut  =cut
   
 sub ShowStatus {  sub ShowStatus {
     my $state = shift;      my $state = shift;
     my $now = time;      my $now = time;
Line 351  sub ShowStatus { Line 281  sub ShowStatus {
   
 =pod  =pod
   
 =head 2 SocketTimeout  =head2 SocketTimeout
   
     Called when an action on the socket times out.  The socket is       Called when an action on the socket times out.  The socket is 
    destroyed and any active transaction is failed.     destroyed and any active transaction is failed.
   
   
 =cut  =cut
   
 sub SocketTimeout {  sub SocketTimeout {
     my $Socket = shift;      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      KillSocket($Socket); # A transaction timeout also counts as
                                 # a connection failure:                                  # a connection failure:
     $ConnectionRetriesLeft--;      $ConnectionRetriesLeft--;
       if($ConnectionRetriesLeft <= 0) {
    Log("CRITICAL", "Host marked DEAD: ".GetServerHost());
    $LondConnecting = 0;
       }
   
   }
   
   #
   #   This function should be called by the child in all cases where it must
   #   exit.  The child process must create a lock file for the AF_UNIX socket
   #   in order to prevent connection requests from lonnet in the time between
   #   process exit and the parent picking up the listen again.
   #
   # Parameters:
   #     exit_code           - Exit status value, however see the next parameter.
   #     message             - If this optional parameter is supplied, the exit
   #                           is via a die with this message.
   #
   sub child_exit {
       my ($exit_code, $message) = @_;
   
       # Regardless of how we exit, we may need to do the lock thing:
   
       #
       #  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);
       if ($hosts_tab) {
    unlink(&GetLoncSocketPath());
       }
   
       if ($message) {
    die($message);
       } else {
    exit($exit_code);
       }
 }  }
   #----------------------------- Timer management ------------------------
   
 =pod  =pod
   
Line 376  Invoked  each timer tick. Line 355  Invoked  each timer tick.
   
   
 sub Tick {  sub Tick {
       my ($Event)       = @_;
       my $clock_watcher = $Event->w;
   
     my $client;      my $client;
     if($ConnectionRetriesLeft > 0) {      UpdateStatus();
  ShowStatus(GetServerHost()." Connection count: ".$ConnectionCount  
    ." Retries remaining: ".$ConnectionRetriesLeft);  
     } else {  
  ShowStatus(GetServerHost()." >> DEAD <<");  
     }  
     # Is it time to prune connection count:      # Is it time to prune connection count:
   
   
Line 392  sub Tick { Line 370  sub Tick {
  if($IdleSeconds > $IdleTimeout) { # Prune a connection...   if($IdleSeconds > $IdleTimeout) { # Prune a connection...
     my $Socket = $IdleConnections->pop();      my $Socket = $IdleConnections->pop();
     KillSocket($Socket);      KillSocket($Socket);
       $IdleSeconds = 0; # Otherwise all connections get trimmed to fast.
       UpdateStatus();
       if(($ConnectionCount == 0)) {
    &child_exit(0);
   
       }
  }   }
     } else {      } else {
  $IdleSeconds = 0; # Reset idle count if not idle.   $IdleSeconds = 0; # Reset idle count if not idle.
Line 399  sub Tick { Line 383  sub Tick {
     #      #
     #  For each inflight transaction, tick down its timeout counter.      #  For each inflight transaction, tick down its timeout counter.
     #      #
     foreach my $item (keys %ActiveTransactions) {  
  my $Socket = $ActiveTransactions{$item}->getServer();      foreach my $item (keys %ActiveConnections) {
  $Socket->Tick();   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 $success;
     Debug(5,"Work but no connections, start ".$Connections." of them");      $success    = &MakeLondConnection;
     my $successCount = 0;      if($success == 0) { # All connections failed:
     for (my $i =0; $i < $Connections; $i++) {  
  $successCount += MakeLondConnection();  
     }  
     if($successCount == 0) { # All connections failed:  
  Debug(5,"Work in queue failed to make any connectiouns\n");   Debug(5,"Work in queue failed to make any connectiouns\n");
  EmptyQueue(); # Fail pending transactions with con_lost.   EmptyQueue(); # Fail pending transactions with con_lost.
    CloseAllLondConnections(); # Should all be closed but....
     }      }
  } else {   } else {
       $LondConnecting = 0;
     ShowStatus(GetServerHost()." >>> DEAD!!! <<<");      ShowStatus(GetServerHost()." >>> DEAD!!! <<<");
     Debug(5,"Work in queue, but gave up on connections..flushing\n");      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();
       }
       &UpdateStatus();
 }  }
   
 =pod  =pod
Line 448  Trigger disconnections of idle sockets. Line 441  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 468  long enough, it will be shut down and re Line 461  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(5, "Server to idle");      &Debug(5, "Server to idle");
Line 547  sub ClientWritable { Line 541  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!!   if($result == 0) {    # client hung up on us!!
     Log("INFO", "lonc pipe client hung up on us!");      # Log("INFO", "lonc pipe client hung up on us!");
     $Watcher->cancel;      $Watcher->cancel;
     $Socket->shutdown(2);      $Socket->shutdown(2);
     $Socket->close();      $Socket->close();
Line 600  The transaction that is being completed. Line 594  The transaction that is being completed.
   
 sub CompleteTransaction {  sub CompleteTransaction {
     &Debug(5,"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");
Line 612  sub CompleteTransaction { Line 609  sub CompleteTransaction {
  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 628  sub CompleteTransaction { Line 627  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();
   
Line 645  sub StartClientReply { Line 644  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 655  sub StartClientReply { Line 656  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 664  Parameters: Line 664  Parameters:
     
    The LondTransaction we are failing.     The LondTransaction we are failing.
     
   
 =cut  =cut
   
 sub FailTransaction {  sub FailTransaction {
     my $transaction = shift;      my $transaction = shift;
     Log("WARNING", "Failing transaction ".$transaction->getRequest());      
     Debug(5, "Failing transaction: ".$transaction->getRequest());      #  If the socket is dead, that's already logged.
   
       if ($ConnectionRetriesLeft > 0) {
    Log("WARNING", "Failing transaction "
       .$transaction->getLoggableRequest());
       }
       Debug(1, "Failing transaction: ".$transaction->getLoggableRequest());
     if (!$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  = $transaction->getClient();   my $client  = $transaction->getClient();
  Debug(5," Replying con_lost to ".$transaction->getRequest());   Debug(1," Replying con_lost to ".$transaction->getRequest());
  StartClientReply($transaction, "con_lost\n");   StartClientReply($transaction, "con_lost\n");
     }      }
     if($ConnectionRetriesLeft <= 0) {  
  Log("CRITICAL", "Host marked dead: ".GetServerHost());  
     }  
   
 }  }
   
 =pod  =pod
   
 =head1  EmptyQueue  =head1  EmptyQueue
   
   Fails all items in the work queue with con_lost.    Fails all items in the work queue with con_lost.
   Note that each item in the work queue is a transaction.    Note that each item in the work queue is a transaction.
   
 =cut  =cut
   
 sub EmptyQueue {  sub EmptyQueue {
     $ConnectionRetriesLeft--; # Counts as connection failure too.      $ConnectionRetriesLeft--; # Counts as connection failure too.
     while($WorkQueue->Count()) {      while($WorkQueue->Count()) {
Line 703  sub EmptyQueue { Line 709  sub EmptyQueue {
 Close all connections open on lond prior to exit e.g.  Close all connections open on lond prior to exit e.g.
   
 =cut  =cut
   
 sub CloseAllLondConnections {  sub CloseAllLondConnections {
     foreach my $Socket (keys %ActiveConnections) {      foreach my $Socket (keys %ActiveConnections) {
  KillSocket($Socket);        if(exists($ActiveTransactions{$Socket})) {
    FailTransaction($ActiveTransactions{$Socket});
         }
         KillSocket($Socket);
     }      }
 }  }
 =cut  
   
 =pod  =pod
   
Line 730  Parameters: Line 739  Parameters:
   
 nonzero if we are allowed to create a new connection.  nonzero if we are allowed to create a new connection.
   
   
 =cut  =cut
   
 sub KillSocket {  sub KillSocket {
     my $Socket = shift;      my $Socket = shift;
   
Line 749  sub KillSocket { Line 758  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 782  The connection must echo the challenge b Line 792  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 836  sub LondReadable { Line 857  sub LondReadable {
     &Debug(2, "Socket->Readable returned: $status");      &Debug(2, "Socket->Readable returned: $status");
   
     if($status != 0) {      if($status != 0) {
  # bad return from socket read. Currently this means that   # 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",   Log("WARNING",
     "Lond connection lost.");      "Lond connection lost.");
  if(exists($ActiveTransactions{$Socket})) {   if(exists($ActiveTransactions{$Socket})) {
     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);
Line 854  sub LondReadable { Line 880  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 878  sub LondReadable { Line 919  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: $LondVersion");
    }
  # If necessary, complete a transaction and then go into the   # If necessary, complete a transaction and then go into the
  # idle queue.   # idle queue.
  #  Note that a trasition to idle indicates a live lond   #  Note that a trasition to idle indicates a live lond
Line 894  sub LondReadable { Line 943  sub LondReadable {
  .$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
  #  here inthe first place.   #  here inthe first place.
   
  Deubg(6, "SendingRequest state encountered in readable");   Debug(6, "SendingRequest state encountered in readable");
  $Watcher->poll("w");   $Watcher->poll("w");
  $Watcher->cb(\&LondWritable);   $Watcher->cb(\&LondWritable);
   
Line 907  sub LondReadable { Line 960  sub LondReadable {
   
   
     } else {      } else {
  # Invalid state.   # Invalid state.
  Debug(4, "Invalid state in LondReadable");   Debug(4, "Invalid state in LondReadable");
     }      }
 }  }
Line 989  sub LondWritable { Line 1042  sub LondWritable {
   
     SocketDump(6,$Socket);      SocketDump(6,$Socket);
   
     if      ($State eq "Connected")         {      #  If the socket is writable, we must always write.
       # Only by writing will we undergo state transitions.
  if ($Socket->Writable() != 0) {      # Old logic wrote in state specific code below, however
     #  The write resulted in an error.      # That forces us at least through another invocation of
     # We'll treat this as if the socket got disconnected:      # this function after writability is possible again.
     Log("WARNING", "Connection to ".$RemoteHost.      # This logic also factors out common code for handling
  " has been disconnected");      # 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});      FailTransaction($ActiveTransactions{$Socket});
     $Watcher->cancel();   } else {
     KillSocket($Socket);      #  In the process of conneting, so need to turn that off.
     return;      
       $LondConnecting = 0;
  }   }
  #  "init" is being sent...   $Watcher->cancel();
    KillSocket($Socket);
    return;
       }
   
   
   
       if      ($State eq "Connected")         {
   
    #  "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 1011  sub LondWritable { Line 1083  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 1053  sub LondWritable { Line 1128  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 1088  sub LondWritable { Line 1153  sub LondWritable {
 =pod  =pod
           
 =cut  =cut
   
 sub QueueDelayed {  sub QueueDelayed {
     Debug(3,"QueueDelayed called");      Debug(3,"QueueDelayed called");
   
Line 1095  sub QueueDelayed { Line 1161  sub QueueDelayed {
   
     Debug(4, "Delayed path: ".$path);      Debug(4, "Delayed path: ".$path);
     opendir(DIRHANDLE, $path);      opendir(DIRHANDLE, $path);
       
     my @alldelayed = grep /\.$RemoteHost$/, readdir DIRHANDLE;      my @all_host_ids;
       my $host_iterator =   &LondConnection::GetHostIterator();
       while (!$host_iterator->end()) {
    my ($host_id,$host_name) = @{$host_iterator->get()}[0,3];
    if ($host_name eq $RemoteHost) {
       push(@all_host_ids, $host_id);
    }
    $host_iterator->next();
       }
       my $host_id_re = '(?:'.join('|',@all_host_ids).')';
       my @alldelayed = grep(/\.$host_id_re$/, readdir(DIRHANDLE));
     closedir(DIRHANDLE);      closedir(DIRHANDLE);
     my $dfname;      foreach my $dfname (sort(@alldelayed)) {
     my $reqfile;   my $reqfile = "$path/$dfname";
     foreach $dfname (sort  @alldelayed) {   my ($host_id) = ($dfname =~ /\.([^.]*)$/);
  $reqfile = "$path/$dfname";   Debug(4, "queueing ".$reqfile." for $host_id");
  Debug(4, "queueing ".$reqfile);  
  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."\n"; # 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 $Transaction = LondTransaction->new("sethost:$host_id:$cmd");
  $Transaction->SetDeferred($reqfile);   $Transaction->SetDeferred($reqfile);
  QueueTransaction($Transaction);   QueueTransaction($Transaction);
     }      }
Line 1133  sub MakeLondConnection { Line 1208  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.
Line 1145  sub MakeLondConnection { Line 1220  sub MakeLondConnection {
  # exchange underway.   # exchange underway.
  #   #
  my $Socket = $Connection->GetSocket();   my $Socket = $Connection->GetSocket();
  if($Socket == undef) {   if($Socket eq undef) {
     die "did not get a socket from the connection";      &child_exit(-1, "did not get a socket from the connection");
  } else {   } else {
     &Debug(9,"MakeLondConnection got socket: ".$Socket);      &Debug(9,"MakeLondConnection got socket: ".$Socket);
  }   }
Line 1159  sub MakeLondConnection { Line 1234  sub MakeLondConnection {
    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($ConnectionCount == 1) { # First Connection:
Line 1167  sub MakeLondConnection { Line 1244  sub MakeLondConnection {
  }   }
  Log("SUCESS", "Created connection ".$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 1198  The text of the request to send. Line 1276  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 1250  sub QueueTransaction { Line 1328  sub QueueTransaction {
     if(!defined $LondSocket) { # Need to queue request.      if(!defined $LondSocket) { # Need to queue request.
  Debug(5,"Must queue...");   Debug(5,"Must queue...");
  $WorkQueue->enqueue($requestData);   $WorkQueue->enqueue($requestData);
  if($ConnectionCount < $MaxConnectionCount) {   Debug(5, "Queue Transaction startnew $ConnectionCount $LondConnecting");
    if(($ConnectionCount < $MaxConnectionCount)   && (! $LondConnecting)) {
   
     if($ConnectionRetriesLeft > 0) {      if($ConnectionRetriesLeft > 0) {
  Debug(5,"Starting additional lond connection");   Debug(5,"Starting additional lond connection");
  if(MakeLondConnection() == 0) {   if(&MakeLondConnection() == 0) {
     EmptyQueue(); # Fail transactions, can't make connection.      EmptyQueue(); # Fail transactions, can't make connection.
       CloseAllLondConnections; # Should all be closed but...
  }   }
     } else {      } else {
  ShowStatus(GetServerHost()." >>> DEAD !!!! <<<");   ShowStatus(GetServerHost()." >>> DEAD !!!! <<<");
    $LondConnecting = 0;
  EmptyQueue(); # It's worse than that ... he's dead Jim.   EmptyQueue(); # It's worse than that ... he's dead Jim.
    CloseAllLondConnections; # Should all be closed but..
     }      }
  }   }
     } else { # Can start the request:      } else { # Can start the request:
Line 1301  sub ClientRequest { Line 1384  sub ClientRequest {
     Debug(8,"Data: ".$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($data eq "close_connection_exit\n") {
     Log("CRITICAL",      Log("CRITICAL",
  "Request Close Connection ... exiting");   "Request Close Connection ... exiting");
Line 1309  sub ClientRequest { Line 1392  sub ClientRequest {
     exit;      exit;
  }   }
  Debug(8, "Complete transaction received: ".$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 1317  sub ClientRequest { Line 1403  sub ClientRequest {
   
 }  }
   
   #
   #     Accept a connection request for a client (lonc child) and
   #    start up an event watcher to keep an eye on input from that 
   #    Event.  This can be called both from NewClient and from
   #    ChildProcess.
   # Parameters:
   #    $socket       - The listener socket.
   # Returns:
   #   NONE
   # Side Effects:
   #    An event is made to watch the accepted connection.
   #    Active clients hash is updated to reflect the new connection.
   #    The client connection count is incremented.
   #
   sub accept_client {
       my ($socket) = @_;
   
       Debug(8, "Entering accept for lonc UNIX socket\n");
       my $connection = $socket->accept(); # Accept the client connection.
       Debug(8,"Connection request accepted from "
     .GetPeername($connection, AF_UNIX));
   
   
       my $description = sprintf("Connection to lonc client %d",
         $ClientConnection);
       Debug(9, "Creating event named: ".$description);
       Event->io(cb      => \&ClientRequest,
         poll    => 'r',
         desc    => $description,
         data    => "",
         fd      => $connection);
       $ActiveClients{$connection} = $ClientConnection;
       $ClientConnection++;
   }
   
 =pod  =pod
   
Line 1335  sub NewClient { Line 1455  sub NewClient {
     my $event      = shift; # Get the event parameters.      my $event      = shift; # Get the event parameters.
     my $watcher    = $event->w;       my $watcher    = $event->w; 
     my $socket     = $watcher->fd; # Get the event' socket.      my $socket     = $watcher->fd; # Get the event' socket.
     my $connection = $socket->accept(); # Accept the client connection.  
     Debug(8,"Connection request accepted from "  
   .GetPeername($connection, AF_UNIX));  
   
   
     my $description = sprintf("Connection to lonc client %d",      &accept_client($socket);
       $ClientConnection);  
     Debug(9, "Creating event named: ".$description);  
     Event->io(cb      => \&ClientRequest,  
       poll    => 'r',  
       desc    => $description,  
       data    => "",  
       fd      => $connection);  
     $ActiveClients{$connection} = $ClientConnection;  
     $ClientConnection++;  
 }  }
   
 =pod  =pod
Line 1359  sub NewClient { Line 1466  sub NewClient {
 Returns the name of the UNIX socket on which to listen for client  Returns the name of the UNIX socket on which to listen for client
 connections.  connections.
   
   =head2 Parameters:
   
       host (optional)  - Name of the host socket to return.. defaults to
                          the return from GetServerHost().
   
 =cut  =cut
   
 sub GetLoncSocketPath {  sub GetLoncSocketPath {
     return $UnixSocketDir."/".GetServerHost();  
       my $host = GetServerHost(); # Default host.
       if (@_) {
    ($host)  = @_; # Override if supplied.
       }
       return $UnixSocketDir."/".$host;
 }  }
   
 =pod  =pod
Line 1399  connection.  The event handler establish Line 1516  connection.  The event handler establish
 (creating a communcations channel), that int turn will establish  (creating a communcations channel), that int turn will establish
 another event handler to subess requests.  another event handler to subess requests.
   
   =head2  Parameters:
   
      host (optional)   Name of the host to set up a unix socket to.
   
 =cut  =cut
   
 sub SetupLoncListener {  sub SetupLoncListener {
       my ($host,$SocketName) = @_;
       if (!$host) { $host = &GetServerHost(); }
       if (!$SocketName) { $SocketName = &GetLoncSocketPath($host); }
   
   
     my $socket;  
     my $SocketName = GetLoncSocketPath();  
     unlink($SocketName);      unlink($SocketName);
   
       my $socket;
     unless ($socket =IO::Socket::UNIX->new(Local  => $SocketName,      unless ($socket =IO::Socket::UNIX->new(Local  => $SocketName,
     Listen => 10,       Listen => 250, 
     Type   => SOCK_STREAM)) {      Type   => SOCK_STREAM)) {
  die "Failed to create a lonc listner socket";   if($I_am_child) {
       &child_exit(-1, "Failed to create a lonc listener socket");
    } else {
       die "Failed to create a lonc listner socket";
    }
     }      }
     Event->io(cb     => \&NewClient,      return $socket;
       poll   => 'r',  }
       desc   => 'Lonc listener Unix 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   =pod 
Line 1426  into the status file. Line 1570  into the status file.
   
 We also use this to reset the retries count in order to allow the  We also use this to reset the retries count in order to allow the
 client to retry connections with a previously dead server.  client to retry connections with a previously dead server.
   
 =cut  =cut
   
 sub ChildStatus {  sub ChildStatus {
     my $event = shift;      my $event = shift;
     my $watcher = $event->w;      my $watcher = $event->w;
   
     Debug(2, "Reporting child status because : ".$watcher->data);      Debug(2, "Reporting child status because : ".$watcher->data);
     my $docdir = $perlvar{'lonDocRoot'};      my $docdir = $perlvar{'lonDocRoot'};
     my $fh = IO::File->new(">>$docdir/lon-status/loncstatus.txt");      
     print $fh $$."\t".$RemoteHost."\t".$Status."\t".      open(LOG,">>$docdir/lon-status/loncstatus.txt");
       flock(LOG,LOCK_EX);
       print LOG $$."\t".$RemoteHost."\t".$Status."\t".
  $RecentLogEntry."\n";   $RecentLogEntry."\n";
       #
       #  Write out information about each of the connections:
       #
       if ($DebugLevel > 2) {
    print LOG "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 LOG "Connection $i State: $state\n";
       print STDERR "---------------------- Connection $i \n";
       $Socket->Dump(-1); # Ensure it gets dumped..
       $i++;
    }
       }
       flock(LOG,LOCK_UN);
       close(LOG);
     $ConnectionRetriesLeft = $ConnectionRetries;      $ConnectionRetriesLeft = $ConnectionRetries;
       UpdateStatus();
 }  }
   
 =pod  =pod
Line 1457  sub SignalledToDeath { Line 1624  sub SignalledToDeath {
     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\"");
     exit 0;      exit 0;
   
 }  }
   
   =pod
   
 =head2 ToggleDebug  =head2 ToggleDebug
   
 This sub toggles trace debugging on and off.  This sub toggles trace debugging on and off.
Line 1478  sub ToggleDebug { Line 1647  sub ToggleDebug {
   
 }  }
   
   =pod
   
 =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.
   Optional parameter:
      $socket  - if provided, this is a socket already open for listen
                 on the client socket. Otherwise, a new listen is set up.
   
 =cut  =cut
   
 sub ChildProcess {  sub ChildProcess {
       #  We've inherited all the
       #  events of our parent and those have to be cancelled or else
       #  all holy bloody chaos will result.. trust me, I already made
       #  >that< mistake.
   
       my $host = GetServerHost();
       foreach my $listener (keys %parent_dispatchers) {
    my $watcher = $parent_dispatchers{$listener};
    my $s       = $watcher->fd;
    if ($listener ne $host) { # Close everyone but me.
       Debug(5, "Closing listen socket for $listener");
       $s->close();
    }
    Debug(5, "Killing watcher for $listener");
   
    $watcher->cancel();
    delete($parent_dispatchers{$listener});
   
       }
   
       #  kill off the parent's signal handlers too!  
       #
   
       for my $handler (keys %parent_handlers) {
    my $watcher = $parent_handlers{$handler};
    $watcher->cancel();
    delete($parent_handlers{$handler});
       }
   
       $I_am_child    = 1; # Seems like in spite of it all I may still getting
                                   # parent event dispatches.. flag I'm a child.
   
   
     #      #
     #  Signals must be handled by the Event framework...      #  Signals must be handled by the Event framework...
 #      #
   
     Event->signal(signal   => "QUIT",      Event->signal(signal   => "QUIT",
   cb       => \&SignalledToDeath,    cb       => \&SignalledToDeath,
Line 1500  sub ChildProcess { Line 1705  sub ChildProcess {
     Event->signal(signal   => "USR1",      Event->signal(signal   => "USR1",
   cb       => \&ChildStatus,    cb       => \&ChildStatus,
   data     => "USR1");    data     => "USR1");
       Event->signal(signal   => "USR2",
     cb       => \&ToggleTransactionLogging);
     Event->signal(signal   => "INT",      Event->signal(signal   => "INT",
   cb       => \&ToggleDebug,    cb       => \&ToggleDebug,
   data     => "INT");    data     => "INT");
   
     SetupTimer();      #  Figure out if we got passed a socket or need to open one to listen for
           #  client requests.
     SetupLoncListener();  
       my ($socket) = @_;
       if (!$socket) {
   
    $socket =  SetupLoncListener();
       }
       #  Establish an event to listen for client connection requests.
   
   
       Event->io(cb   => \&NewClient,
         poll => 'r',
         desc => 'Lonc Listener Unix Socket',
         fd   => $socket);
           
     $Event::Debuglevel = $DebugLevel;      $Event::DebugLevel = $DebugLevel;
           
     Debug(9, "Making initial lond connection for ".$RemoteHost);      Debug(9, "Making initial lond connection for ".$RemoteHost);
   
 # Setup the initial server connection:  # Setup the initial server connection:
           
      # &MakeLondConnection(); // let first work requirest do it.       # &MakeLondConnection(); // let first work request do it.
   
       #  need to accept the connection since the event may  not fire.
   
       &accept_client($socket);
   
     Debug(9,"Entering event loop");      Debug(9,"Entering event loop");
     my $ret = Event::loop(); #  Start the main event loop.      my $ret = Event::loop(); #  Start the main event loop.
           
           
     die "Main event loop exited!!!";      &child_exit (-1,"Main event loop exited!!!");
 }  }
   
 #  Create a new child for host passed in:  #  Create a new child for host passed in:
   
 sub CreateChild {  sub CreateChild {
       my ($host, $socket) = @_;
   
     my $sigset = POSIX::SigSet->new(SIGINT);      my $sigset = POSIX::SigSet->new(SIGINT);
     sigprocmask(SIG_BLOCK, $sigset);      sigprocmask(SIG_BLOCK, $sigset);
     my $host = shift;  
     $RemoteHost = $host;      $RemoteHost = $host;
     Log("CRITICAL", "Forking server for ".$host);      Log("CRITICAL", "Forking server for ".$host);
     my $pid          = fork;      my $pid          = fork;
     if($pid) { # Parent      if($pid) { # Parent
  $RemoteHost = "Parent";   $RemoteHost = "Parent";
  $ChildHash{$pid} = $host;   $ChildHash{$pid} = $host;
  $HostToPid{$host}= $pid;  
  sigprocmask(SIG_UNBLOCK, $sigset);   sigprocmask(SIG_UNBLOCK, $sigset);
   
     } else { # child.      } else { # child.
  ShowStatus("Connected to ".$RemoteHost);   ShowStatus("Connected to ".$RemoteHost);
  $SIG{INT} = 'DEFAULT';   $SIG{INT} = 'DEFAULT';
  sigprocmask(SIG_UNBLOCK, $sigset);   sigprocmask(SIG_UNBLOCK, $sigset);
  ChildProcess; # Does not return.   if(defined $socket) {
       &ChildProcess($socket);
    } else {
       ChildProcess; # Does not return.
    }
     }      }
   }
   
   # parent_client_connection:
   #    Event handler that processes client connections for the parent process.
   #    This sub is called when the parent is listening on a socket and
   #    a connection request arrives.  We must:
   #     Start a child process to accept the connection request.
   #     Kill our listen on the socket.
   # Parameter:
   #    event       - The event object that was created to monitor this socket.
   #                  event->w->fd is the socket.
   # Returns:
   #    NONE
   #
   sub parent_client_connection {
       if ($I_am_child) {
    #  Should not get here, but seem to anyway:
    &Debug(5," Child caught parent client connection event!!");
    my ($event) = @_;
    my $watcher = $event->w;
    $watcher->cancel(); # Try to kill it off again!!
       } else {
    &Debug(9, "parent_client_connection");
    my ($event)   = @_;
    my $watcher   = $event->w;
    my $socket    = $watcher->fd;
    if ($hosts_tab) {
   
       # Lookup the host associated with this socket:
   
       my $host = $listening_to{$socket};
   
       # Start the child:
   
   
   
       &Debug(9,"Creating child for $host (parent_client_connection)");
       &CreateChild($host, $socket);
       
       # Clean up the listen since now the child takes over until it exits.
   
       $watcher->cancel(); # Nolonger listening to this event
       delete($listening_to{$socket});
       delete($parent_dispatchers{$host});
       $socket->close();
   
    } else {
       my $connection = $socket->accept(); # Accept the client connection.
       Event->io(cb      => \&get_remote_hostname,
         poll    => 'r',
         data    => "",
         fd      => $connection);
    }
       }
 }  }
   
   sub get_remote_hostname {
    my ($event)   = @_;
    my $watcher   = $event->w;
    my $socket    = $watcher->fd;
   
    my $thisread;
    my $rv = $socket->recv($thisread, POSIX::BUFSIZ, 0);
    Debug(8, "rcv:  data length = ".length($thisread)." read =".$thisread);
    if (!defined($rv) || length($thisread) == 0) {
       # Likely eof on socket.
       Debug(5,"Client Socket closed on lonc for p_c_c");
       close($socket);
       $watcher->cancel();
       return;
    }
   
    my $data    = $watcher->data().$thisread;
    $watcher->data($data);
    if($data =~ /\n$/) { # Request entirely read.
       chomp($data);
    } else {
       return;
    }
   
    &Debug(5,"Creating child for $data (parent_client_connection)");
    &CreateChild($data);
   
    # Clean up the listen since now the child takes over until it exits.
    $watcher->cancel(); # Nolonger listening to this event
    $socket->send("done\n");
    $socket->close();
   }
   
   # parent_listen:
   #    Opens a socket and starts a listen for the parent process on a client UNIX
   #    domain socket.
   #
   #    This involves:
   #       Creating a socket for listen.
   #       Removing any socket lock file
   #       Adding an event handler for this socket becoming readable
   #         To the parent's event dispatcher.
   # Parameters:
   #    loncapa_host    - LonCAPA cluster name of the host represented by the client
   #                      socket.
   # Returns:
   #    NONE
   #
   sub parent_listen {
       my ($loncapa_host) = @_;
       Debug(5, "parent_listen: $loncapa_host");
   
       my ($socket,$file);
       if (!$loncapa_host) {
    $loncapa_host = 'common_parent';
    $file         = $perlvar{'lonSockCreate'};
       } else {
    $file         = &GetLoncSocketPath($loncapa_host);
       }
       $socket = &SetupLoncListener($loncapa_host,$file);
   
       $listening_to{$socket} = $loncapa_host;
       if (!$socket) {
    die "Unable to create a listen socket for $loncapa_host";
       }
       
       my $lock_file = $file.".lock";
       unlink($lock_file); # No problem if it doesn't exist yet [startup e.g.]
   
       my $watcher = 
    Event->io(cb    => \&parent_client_connection,
     poll  => 'r',
     desc  => "Parent listener unix socket ($loncapa_host)",
     data => "",
     fd    => $socket);
       $parent_dispatchers{$loncapa_host} = $watcher;
   
   }
   
   sub parent_clean_up {
       my ($loncapa_host) = @_;
       Debug(5, "parent_clean_up: $loncapa_host");
   
       my $socket_file = &GetLoncSocketPath($loncapa_host);
       unlink($socket_file); # No problem if it doesn't exist yet [startup e.g.]
       my $lock_file   = $socket_file.".lock";
       unlink($lock_file); # No problem if it doesn't exist yet [startup e.g.]
   }
   
   
   # listen_on_all_unix_sockets:
   #    This sub initiates a listen on all unix domain lonc client sockets.
   #    This will be called in the case where we are trimming idle processes.
   #    When idle processes are trimmed, loncnew starts up with no children,
   #    and only spawns off children when a connection request occurs on the
   #    client unix socket.  The spawned child continues to run until it has
   #    been idle a while at which point it eventually exits and once more
   #    the parent picks up the listen.
   #
   #  Parameters:
   #      NONE
   #  Implicit Inputs:
   #    The configuration file that has been read in by LondConnection.
   #  Returns:
   #     NONE
   #
   sub listen_on_all_unix_sockets {
       Debug(5, "listen_on_all_unix_sockets");
       my $host_iterator      =   &LondConnection::GetHostIterator();
       while (!$host_iterator->end()) {
    my $host_entry_ref =   $host_iterator->get();
    my $host_name      = $host_entry_ref->[3];
    Debug(9, "Listen for $host_name");
    &parent_listen($host_name);
    $host_iterator->next();
       }
   }
   
   sub listen_on_common_socket {
       Debug(5, "listen_on_common_socket");
       &parent_listen();
   }
   
   #   server_died is called whenever a child process exits.
   #   Since this is dispatched via a signal, we must process all
   #   dead children until there are no more left.  The action
   #   is to:
   #      - Remove the child from the bookeeping hashes
   #      - Re-establish a listen on the unix domain socket associated
   #        with that host.
   # Parameters:
   #    The event, but we don't actually care about it.
   sub server_died {
       &Debug(9, "server_died called...");
       
       while(1) { # Loop until waitpid nowait fails.
    my $pid = waitpid(-1, WNOHANG);
    if($pid <= 0) {
       return; # Nothing left to wait for.
    }
    # need the host to restart:
   
    my $host = $ChildHash{$pid};
    if($host) { # It's for real...
       &Debug(9, "Caught sigchild for $host");
       delete($ChildHash{$pid});
       if ($hosts_tab) {
    &parent_listen($host);
       } else {
    &parent_clean_up($host);
       }
       
    } else {
       &Debug(5, "Caught sigchild for pid not in hosts hash: $pid");
    }
       }
   
   }
   
 #  #
 #  Parent process logic pass 1:  #  Parent process logic pass 1:
 #   For each entry in the hosts table, we will  #   For each entry in the hosts table, we will
Line 1594  ShowStatus("Forking node servers"); Line 2033  ShowStatus("Forking node servers");
   
 Log("CRITICAL", "--------------- Starting children ---------------");  Log("CRITICAL", "--------------- Starting children ---------------");
   
 my $HostIterator = LondConnection::GetHostIterator;  LondConnection::ReadConfig;               # Read standard config files.
 while (! $HostIterator->end()) {  
   
     my $hostentryref = $HostIterator->get();  $RemoteHost = "[parent]";
     CreateChild($hostentryref->[0]);  if ($hosts_tab) {
     $HostHash{$hostentryref->[0]} = $hostentryref->[4];      &listen_on_all_unix_sockets();
     $HostIterator->next();  } else {
       &listen_on_common_socket();
 }  }
   
 $RemoteHost = "Parent Server";  $RemoteHost = "Parent Server";
   
 # Maintain the population:  # Maintain the population:
   
 ShowStatus("Parent keeping the flock");  ShowStatus("Parent keeping the flock");
   
 #  
 #   Set up parent signals:  
 #  
   
 $SIG{INT}  = \&Terminate;  
 $SIG{TERM} = \&Terminate;   
 $SIG{HUP}  = \&Restart;  
 $SIG{USR1} = \&CheckKids;   
 $SIG{USR2} = \&UpdateKids; # LonManage update request.  
   
 while(1) {  
     my $deadchild = wait();  
     if(exists $ChildHash{$deadchild}) { # need to restart.  
  my $deadhost = $ChildHash{$deadchild};  
  delete($HostToPid{$deadhost});  
  delete($ChildHash{$deadchild});  
  Log("WARNING","Lost child pid= ".$deadchild.  
       "Connected to host ".$deadhost);  
  Log("INFO", "Restarting child procesing ".$deadhost);  
  CreateChild($deadhost);  
     }  
 }  
   
   # We need to setup a SIGChild event to handle the exit (natural or otherwise)
   # of the children.
   
   Event->signal(cb       => \&server_died,
         desc     => "Child exit handler",
         signal   => "CHLD");
   
   
   # Set up all the other signals we set up.
   
   $parent_handlers{INT} = Event->signal(cb       => \&Terminate,
         desc     => "Parent INT handler",
         signal   => "INT");
   $parent_handlers{TERM} = Event->signal(cb       => \&Terminate,
          desc     => "Parent TERM handler",
          signal   => "TERM");
   if ($hosts_tab) {
       $parent_handlers{HUP}  = Event->signal(cb       => \&Restart,
      desc     => "Parent HUP handler.",
      signal   => "HUP");
   } else {
       $parent_handlers{HUP}  = Event->signal(cb       => \&KillThemAll,
      desc     => "Parent HUP handler.",
      signal   => "HUP");
   }
   $parent_handlers{USR1} = Event->signal(cb       => \&CheckKids,
          desc     => "Parent USR1 handler",
          signal   => "USR1");
   $parent_handlers{USR2} = Event->signal(cb       => \&UpdateKids,
          desc     => "Parent USR2 handler.",
          signal   => "USR2");
   
   #  Start procdesing events.
   
   $Event::DebugLevel = $DebugLevel;
   Debug(9, "Parent entering event loop");
   my $ret = Event::loop();
   die "Main Event loop exited: $ret";
   
 =pod  =pod
   
Line 1652  sub CheckKids { Line 2107  sub CheckKids {
     my $now=time;      my $now=time;
     my $local=localtime($now);      my $local=localtime($now);
     print $fh "LONC status $local - parent $$ \n\n";      print $fh "LONC status $local - parent $$ \n\n";
       foreach my $host (keys %parent_dispatchers) {
    print $fh "LONC Parent process listening for $host\n";
       }
     foreach my $pid (keys %ChildHash) {      foreach my $pid (keys %ChildHash) {
  Debug(2, "Sending USR1 -> $pid");   Debug(2, "Sending USR1 -> $pid");
  kill 'USR1' => $pid; # Tell Child to report status.   kill 'USR1' => $pid; # Tell Child to report status.
  sleep 1; # Wait so file doesn't intermix.  
     }      }
   
 }  }
   
 =pod  =pod
Line 1689  sub UpdateKids { Line 2147  sub UpdateKids {
   
     Log("INFO", "Updating connections via SIGUSR2");      Log("INFO", "Updating connections via SIGUSR2");
   
     #  Just in case we need to kill our own lonc, we wait a few seconds to      #  I'm not sure what I was thinking in the first implementation.
     #  give it a chance to receive and relay lond's response to the       # someone will have to work hard to convince me the effect is any
     #  re-init command.      # different than Restart, especially now that we don't start up 
     #      # per host servers automatically, may as well just restart.
       # The down side is transactions that are in flight will get timed out
     sleep(2); # Wait a couple of seconds.      # (lost unless they are critical).
   
     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      if ($hosts_tab) {
     # new ip.   &Restart();
       } else {
     $HostHash{$oldhost} = $hosts{$oldhost};   &KillThemAll();
     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.  
  }  
     }      }
 }  }
   
Line 1782  sub Restart { Line 2177  sub Restart {
     Log("CRITICAL", "Restarting");      Log("CRITICAL", "Restarting");
     my $execdir = $perlvar{'lonDaemons'};      my $execdir = $perlvar{'lonDaemons'};
     unlink("$execdir/logs/lonc.pid");      unlink("$execdir/logs/lonc.pid");
     exec("$execdir/loncnew");      exec("$executable");
 }  }
   
 =pod  =pod
Line 1799  sub KillThemAll { Line 2194  sub KillThemAll {
     local($SIG{CHLD}) = 'IGNORE';      # Our children >will< die.      local($SIG{CHLD}) = 'IGNORE';      # Our children >will< die.
     foreach my $pid (keys %ChildHash) {      foreach my $pid (keys %ChildHash) {
  my $serving = $ChildHash{$pid};   my $serving = $ChildHash{$pid};
  Debug(2, "Killing lonc for $serving pid = $pid");   ShowStatus("Nicely Killing lonc for $serving pid = $pid");
  ShowStatus("Killing lonc for $serving pid = $pid");   Log("CRITICAL", "Nicely Killing lonc for $serving pid = $pid");
  Log("CRITICAL", "Killing lonc for $serving pid = $pid");  
  kill 'QUIT' => $pid;   kill 'QUIT' => $pid;
  delete($ChildHash{$pid});  
     }      }
     my $execdir = $perlvar{'lonDaemons'};  }
     unlink("$execdir/logs/lonc.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  =pod
Line 1819  Terminate the system. Line 2229  Terminate the system.
 =cut  =cut
   
 sub Terminate {  sub Terminate {
     KillThemAll;      &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");      Log("CRITICAL","Master process exiting");
     exit 0;      exit 0;
   

Removed from v.1.29  
changed lines
  Added in v.1.80


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