* Changed 'database::X::ping_before_connect' to 'database::X::ping' and made the value be the actual timeout to wait for pings when trying to connect to a database.

Signed-off-by: Digimer <digimer@alteeve.ca>

Anvil/Tools/Database.pm

@@ -551,19 +551,21 @@ sub configure_pgsql
 
 This method tries to connect to all databases it knows of. To define databases for a machine to connect to, load a configuration file with the following parameters;
 
- database::1::host			=	an-striker01.alteeve.com
+ database::1::host		=	an-striker01.alteeve.com
- database::1::port			=	5432
+ database::1::port		=	5432
- database::1::name			=	scancore
+ database::1::name		=	scancore
- database::1::user			=	admin
+ database::1::user		=	admin
- database::1::password			=	Initial1
+ database::1::password		=	Initial1
- database::1::ping_before_connect	=	1
+ database::1::ping		=	0.25
- 
+ 
- database::2::host			=	an-striker02.alteeve.com
+ database::2::host		=	an-striker02.alteeve.com
- database::2::port			=	5432
+ database::2::port		=	5432
- database::2::name			=	scancore
+ database::2::name		=	scancore
- database::2::user			=	admin
+ database::2::user		=	admin
- database::2::password			=	Initial1
+ database::2::password		=	Initial1
- database::2::ping_before_connect	=	1
+ database::2::ping		=	0.25
+
+Please see the comments in /etc/anvil/anvil.conf for more information on how to use the above variables.
  
 The C<< 1 >> and C<< 2 >> are the IDs of the given databases. They can be any number and do not need to be sequential, they just need to be unique. 
 
@@ -679,9 +681,9 @@ sub connect
 		}});
 		
 		# If not set, we will always ping before connecting.
-		if ((not exists $anvil->data->{database}{$id}{ping_before_connect}) or (not defined $anvil->data->{database}{$id}{ping_before_connect}))
+		if ((not exists $anvil->data->{database}{$id}{ping}) or (not defined $anvil->data->{database}{$id}{ping}))
 		{
-			$anvil->data->{database}{$id}{ping_before_connect} = 1;
+			$anvil->data->{database}{$id}{ping} = 1;
 		}
 		
 		# Make sure the user didn't specify the same target twice.
@@ -713,17 +715,22 @@ sub connect
 			password => $anvil->Log->secure ? $password : "--",
 		}});
 		
+		### TODO: Can we do a telnet port ping with a short timeout instead of a shell ping call?
+		
 		# Assemble my connection string
 		my $db_connect_string = "$driver:dbname=$name;host=$host;port=$port";
 		$anvil->Log->variables({source => $THIS_FILE, line => __LINE__, level => 3, list => { 
-			db_connect_string                      => $db_connect_string, 
+			db_connect_string       => $db_connect_string, 
-			"database::${id}::ping_before_connect" => $anvil->data->{database}{$id}{ping_before_connect},
+			"database::${id}::ping" => $anvil->data->{database}{$id}{ping},
 		}});
-		#if ($anvil->data->{database}{$id}{ping_before_connect})
+		if ($anvil->data->{database}{$id}{ping})
-		if (0)
 		{
 			# Can I ping?
-			my ($pinged) = $anvil->System->ping({ping => $host, count => 1});
+			my ($pinged) = $anvil->System->ping({
+				ping    => $host, 
+				count   => 1,
+				timeout => $anvil->data->{database}{$id}{ping},
+			});
 			
 			my $ping_time = tv_interval ($start_time, [gettimeofday]);
 			print "[".$ping_time."] - Pinged: [$host:$port:$name:$user]\n";
@@ -731,7 +738,7 @@ sub connect
 			$anvil->Log->variables({source => $THIS_FILE, line => __LINE__, level => 3, list => { pinged => $pinged }});
 			if (not $pinged)
 			{
-				# Didn't ping and 'database::<id>::ping_before_connect' not set. Record this 
+				# Didn't ping and 'database::<id>::ping' not set. Record this 
 				# in the failed connections array.
 				$anvil->Log->entry({source => $THIS_FILE, line => __LINE__, level => 1, priority => "alert", key => "log_0063", variables => { 
 					host => $port ? $host.":".$port : $host,

Anvil/Tools/System.pm

@@ -840,6 +840,10 @@ B<NOTE>: See C<< System->remote_call >> for additional information on specifying
 
 This is the host name or IP address of a remote machine that you want to run the ping on. This is used to test a remote machine's access to a given ping target.
 
+=head3 timeout (optional, default '1')
+
+This is how long we will wait for a ping to return, in seconds. Any real number is allowed (C<< 1 >> (one second), C<< 0.25 >> (1/4 second), etc). If set to C<< 0 >>, we will wait for the ping command to exit without limit.
+
 =cut
 sub ping
 {
@@ -862,6 +866,7 @@ sub ping
 	my $ping     = $parameter->{ping}     ? $parameter->{ping}     : "";
 	my $port     = $parameter->{port}     ? $parameter->{port}     : "";
 	my $target   = $parameter->{target}   ? $parameter->{target}   : "";
+	my $timeout  = $parameter->{timeout}  ? $parameter->{timeout}  : 1;	# This sets the 'timeout' delay.
 	$anvil->Log->variables({source => $THIS_FILE, line => __LINE__, level => $debug, list => { 
 		count    => $count, 
 		fragment => $fragment, 
@@ -872,6 +877,14 @@ sub ping
 		target   => $target, 
 	}});
 	
+	# Was timeout specified as a simple integer?
+	if (($timeout !~ /^\d+$/) && ($timeout !~ /^\d+\.\d+$/))
+	{
+		# The timeout was invalid, switch it to 1
+		$anvil->Log->variables({source => $THIS_FILE, line => __LINE__, level => $debug, list => { timeout => $timeout }});
+		$timeout = 1;
+	}
+	
 	# If the payload was set, take 28 bytes off to account for ICMP overhead.
 	if ($payload)
 	{
@@ -881,7 +894,12 @@ sub ping
 	
 	# Build the call. Note that we use 'timeout' because if there is no connection and the hostname is 
 	# used to ping and DNS is not available, it could take upwards of 30 seconds time timeout otherwise.
-	my $shell_call = $anvil->data->{path}{exe}{timeout}." 1 ".$anvil->data->{path}{exe}{'ping'}." -W 1 -n $ping -c 1";
+	my $shell_call = "";
+	if ($timeout)
+	{
+		$shell_call = $anvil->data->{path}{exe}{timeout}." $timeout ";
+	}
+	$shell_call .= $anvil->data->{path}{exe}{'ping'}." -W 1 -n $ping -c 1";
 	$anvil->Log->variables({source => $THIS_FILE, line => __LINE__, level => $debug, list => { shell_call => $shell_call }});
 	if (not $fragment)
 	{

anvil.conf

@@ -1,19 +1,44 @@
-# This is the main Striker (and ScanCore) configuration file.
+### This is the main Striker (and ScanCore) configuration file.
 # 
+# Database connections;
+# 
+# Each Anvil! database is defined below using an incrementing counter as the second variable. The value of 
+# the second variable is not important, so long as it is unique. Generally, it's a simple incrementing 
+# intger. 
+# 
+# There are six variables for each database definition;
+# host     = This is the (resolvable) host name or IP address of the peer database.
+# port     = This is the TCP port used to connect to the database. By default, it is 5432
+# name     = This is the name of the database itself. By default (and almost always), it is 'scancore'.
+# user     = This is the user that owns the database. By default (and almost always), it is 'admin'.
+# password = This is the password used to connect to the database.
+#            NOTE: Do not directly change the database password. Please use 'anvil-update-password' so that
+#                  the WebUI, database, nodes and peers are all kept in sync.
+# ping     = If set, this will cause a ping to be made against the database server before the actual 
+#            connection is attempted. This can be useful when peer dashboards are offline for extended 
+#            periods of time. The value can be any real number and will be used as the timeout for the actual
+#            ping. If the peer dashboard is on the same subnet, a value of '0.25' (250ms) should be a good 
+#            balance between giving the peer a chance to reply and not waiting too long when it is offline. 
+#            If the peer is remote, you will want to set this to '1' (1000ms) or higher. 
+#            
+#            Setting this to '0' disables pinging before connecting entirely. In this case, if the peer is
+#            offline, it will take about 3 seconds on average for the connection attempt to timeout. This 
+#            might be necessary if the peer dashboard is behind a firewall/router or otherwise can't respond
+#            to ICMP pings.
 
 database::1::host			=	192.168.122.201
 database::1::port			=	5432
 database::1::name			=	scancore
 database::1::user			=	admin
 database::1::password			=	Initial1
-database::1::ping_before_connect	=	1
+database::1::ping			=	1
 
 database::2::host			=	192.168.122.202
 database::2::port			=	5432
 database::2::name			=	scancore
 database::2::user			=	admin
 database::2::password			=	Initial1
-database::2::ping_before_connect	=	1
+database::2::ping			=	1
 
 # This is the schema for the ScanCore database.
 sys::database::schema			=	/usr/sbin/anvil/anvil.sql

notes

@@ -479,3 +479,9 @@ pcs resource manage srv01-c7
 ====
 
 Unrelated Notes;
+
+
+17:03 <@kgaillot> digimer: quick skim looks good. for metadata, <version> refers to OCF version, so 1.0; your RA version goes in <resource-agent ... version=...>. also on-fail/interval/depth are pacemaker properties, not part of the ocf 
+                  metadata
+17:05 <@kgaillot> digimer: also be aware that probes (monitor w/0 interval) can be done on nodes that aren't intended to run the resource, so it should return not running in that case (eg if necessary files don't exist). really this is 
+                  an issue that should be handled on the pacemaker side but it isn't currently

share/words.xml

@@ -114,7 +114,7 @@ Connecting to Database with configuration ID: [#!variable!id!#]
 		<key name="log_0060">Database user: [#!variable!user!#] already exists with ID: [#!variable!id!#].</key>
 		<key name="log_0061"><![CDATA[[ Error ] - The method Get->users_home() was asked to find the home directory for the user: [#!variable!user!#], but was unable to do so.]]></key>
 		<key name="log_0062">SSH session opened without a password to: [#!variable!target!#].</key>
-		<key name="log_0063">The database: [#!variable!host!# -> #!variable!name!#] with the ID: [#!variable!id!#] did not respond to pings and 'database::#!variable!id!#::ping_before_connect' is not set to '0' in '#!data!path::configs::striker.conf!#', skipping it.</key>
+		<key name="log_0063">The database: [#!variable!host!# -> #!variable!name!#] with the ID: [#!variable!id!#] did not respond to pings and 'database::#!variable!id!#::ping' is not set to '0' in '#!data!path::configs::striker.conf!#', skipping it.</key>
 		<key name="log_0064">[ Warning ] - The database: [#!variable!name!#] on host: [#!variable!host!#] with ID: [#!variable!id!#] can not be used, skipping it.</key>
 		<key name="log_0065">
 The database connection error was: