replication general question
I have two mysql boxes setup. Fast machines, I think they are dual 3ghz with boat loads of ram. They are not real busy servers but they have some pretty good sized tables, one of them with a few million rows. My question is, I have Nagios setup to monitor the seconds behind master on the backup server. Usually the boxes are pretty current, within ten or twenty seconds. Other times though they seem to get way behind, like I just bumped the nagios warning email level up to 600 seconds. It doesn't seem to have anything to do really with usage because it usually happens in the middle of the night. Does Mysql do re-indexing or something? What could I do to figure out why the replicatant box is getting so far behind? --ja -- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: replication general question
What type of drives to you have on your system? That is often more important than CPU speed. My guess is that there are nightly maintenance crons slowing down disk access. I have never monitored replication via the seconds-behind-master function as we do not use 4.1, so I can't speak for how accurate it is.. Atle - Flying Crocodile Inc, Unix Systems Administrator On Thu, 23 Jun 2005 [EMAIL PROTECTED] wrote: I have two mysql boxes setup. Fast machines, I think they are dual 3ghz with boat loads of ram. They are not real busy servers but they have some pretty good sized tables, one of them with a few million rows. My question is, I have Nagios setup to monitor the seconds behind master on the backup server. Usually the boxes are pretty current, within ten or twenty seconds. Other times though they seem to get way behind, like I just bumped the nagios warning email level up to 600 seconds. It doesn't seem to have anything to do really with usage because it usually happens in the middle of the night. Does Mysql do re-indexing or something? What could I do to figure out why the replicatant box is getting so far behind? --ja -- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: replication general question
I don't think network latency would be an issue. This is within a protected
network dmz so it has it's own switch.
Here is the nagios script stuff. Might be more than what you need but let me
know if you are able to use some of it.
I have two on the server. first, I have:
[EMAIL PROTECTED] jabbott]# more secondsBehind.sh
#!/bin/sh
mysql -pMYSECRET -e show slave status\G | grep Seconds
Then I have this that I run in the rc.local. This sets up a port that listens
for a connection on port 5151. I have hole open in my firewall into my dmz for
port 5151
[EMAIL PROTECTED] jabbott]# more socket.pl
#!/usr/bin/perl
use IO::Socket;
$server_port = 5151;
$server = IO::Socket::INET-new(LocalPort = $server_port,
Type = SOCK_STREAM,
Reuse = 1,
Listen= 10)
or die Could not be a tcp server on port $server_port : [EMAIL PROTECTED];
while ($client = $server-accept ()) {
my $sysArg = `/home/jabbott/secondsBehind.sh`;
# uncomment the next line for debugging
print $client is the new connection\n\n;
print $client $sysArg\n;
print connect \n;
close ($client);
}
close ($server);
Then, on the Nagios side I have this:
$ cat /usr/lib/nagios/plugins/mysql-replication-lag.pl
#!/usr/bin/perl -w
use strict;
use lib nagios/plugins ;
use utils qw($TIMEOUT %ERRORS);
use IO::Socket;
$ENV{'PATH'}='';
$ENV{'BASH_ENV'}='';
$ENV{'ENV'}='';
my ($ip_address,$port,$warn,$critical) = @ARGV;
# Just in case of problems, let's not hang Nagios
$SIG{'ALRM'} = sub {
print No Answer from Client\n;
exit $ERRORS{UNKNOWN};
};
alarm($TIMEOUT);
my $sock = new IO::Socket::INET(
PeerAddr = $ip_address,
PeerPort = $port,
Proto= 'tcp',
);
unless ($sock) {
print Socket could not be created. Reason: $!\n;
exit $ERRORS{'UNKNOWN'};
}
my $result = $sock || Could not read socket\n;
close($sock);
alarm(0);
print $result;
unless ($result =~ /^\s*Seconds_Behind_Master:\s*/i) {
exit $ERRORS{'UNKNOWN'};
}
$result =~ s/\D//g;
exit $ERRORS{'CRITICAL'} if ($result$critical);
exit $ERRORS{'WARNING'} if ($result$warn);
exit $ERRORS{'OK'}
1;
__END__
On Thu, 23 Jun 2005, James Green wrote:
Checked for network latency? I have replication running on similar
hardware hooked up to the same switch, and have never seen it rise above
0 seconds behind.
Not that I check often, I have no need to...
Is your nagios script open for public use - I was about to have to write
something for this task myself.
Thanks,
James
On 23/6/2005, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
I have two mysql boxes setup. Fast machines, I think they are dual 3ghz
with boat loads of ram. They are not real busy servers but they have some
pretty good sized tables, one of them with a few million rows.
My question is, I have Nagios setup to monitor the seconds behind master on
the backup server. Usually the boxes are pretty current, within ten or
twenty seconds. Other times though they seem to get way behind, like I just
bumped the nagios warning email level up to 600 seconds. It doesn't seem to
have anything to do really with usage because it usually happens in the
middle of the night. Does Mysql do re-indexing or something? What could I
do to figure out why the replicatant box is getting so far behind?
--ja
--
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Re: replication general question
Hello. I haven't heard about periodical MySQL jobs, but Unix boxes usually have some midnight cron jobs (updating of locate database for example). [EMAIL PROTECTED] wrote: I have two mysql boxes setup. Fast machines, I think they are dual 3ghz with boat loads of ram. They are not real busy servers but they have some pretty good sized tables, one of them with a few million rows. My question is, I have Nagios setup to monitor the seconds behind master on the backup server. Usually the boxes are pretty current, within ten or twenty seconds. Other times though they seem to get way behind, like I just bumped the nagios warning email level up to 600 seconds. It doesn't seem to have anything to do really with usage because it usually happens in the middle of the night. Does Mysql do re-indexing or something? What could I do to figure out why the replicatant box is getting so far behind? --ja -- For technical support contracts, goto https://order.mysql.com/?ref=ensita This email is sponsored by Ensita.NET http://www.ensita.net/ __ ___ ___ __ / |/ /_ __/ __/ __ \/ /Gleb Paharenko / /|_/ / // /\ \/ /_/ / /__ [EMAIL PROTECTED] /_/ /_/\_, /___/\___\_\___/ MySQL AB / Ensita.NET ___/ www.mysql.com -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
