Holes in Graphs

[fragbait]

I have an excessive amount of activity in Cacti:

10/19/2006 11:24:20 AM - CMDPHP: Poller[0] DEBUG: SQL Assoc: "select poller_id,end_time from poller_time where poller_id = 0"
10/19/2006 11:24:20 AM - CMDPHP: Poller[0] DEBUG: SQL Assoc: "select poller_output.output, poller_output.time, poller_output.local_data_id, poller_item.rrd_path, poller_item.rrd_name, poller_item.rrd_num from (poller_output,poller_item) where (poller_output.local_data_id=poller_item.local_data_id and poller_output.rrd_name=poller_item.rrd_name) limit 10000"


It just goes on and on
Sometimes I get data in the graphs and sometimes I don't.

Have looked though the forums and still not luck

Changes made to data:
cat /etc/my.cnf
[mysqld]
set-variable=max_connections=2000

/etc/php.ini
memory_limit = 128M

Using cactid

VERSIONS
cacti-0.8.6i-1.el4.rf
cacti-cactid-0.8.6g-1.2.el4.rf
php-domxml-4.3.9-3.18
php-mysql-4.3.9-3.18
php-imap-4.3.9-3.18
php-4.3.9-3.18
php-odbc-4.3.9-3.18
php-snmp-4.3.9-3.18
php-ldap-4.3.9-3.18
php-pear-4.3.9-3.18
httpd-2.0.52-19.ent.centos4
mysqlclient10-3.23.58-4.RHEL4.1
mysql-4.1.20-1.RHEL4.1
mysqlclient10-devel-3.23.58-4.RHEL4.1
mysql-server-4.1.20-1.RHEL4.1
mysql-4.1.20-1.RHEL4.1
mysql-devel-4.1.20-1.RHEL4.1
mysqlclient10-3.23.58-4.RHEL4.1
libdbi-dbd-mysql-0.6.5-10.RHEL4.1
mysql-bench-4.1.20-1.RHEL4.1

thanks in advance

[TheWitness]

1) Do you have 2 versions of the Cron running (aka did you install from RPM and then follow setup instructions).
2) Is Cactid Segfaulting (post or pm your log to me).

It is easy to tell if 1) is the issue by looking to a graph that only graphs 1 element, if it's ok, then you have two CRON tasks.

TheWitness

[fragbait]

I did set up from RPMs and I did add a cron job, so I see this in cron

Oct 19 13:25:01 mail crond[10917]: (cacti) CMD (php /var/www/cacti/poller.php &>/dev/null)
Oct 19 13:25:01 mail crond[10915]: (root) CMD (su cactiuser -c "php /var/www/cacti/poller.php" > /dev/null 2>&1)


So if you set up from rpm then DO NOT manually add an entry to crons?
This this right?

tx

[TheWitness]

Right, most RPM's perform the setup for you and therefore, when you do it, you get two, they essentially cause updating to fail on any RRD file with more than 1 DS.

TheWitness

[fragbait]

I'll take my self out the back and put myself out of misery!

After getting rid of the cron job I had in place. Everything began to look much better. Thanks TW.

Because the rpm created a user called cacti and I had created a user called cactiuser, I had to make a couple of other configuration changes to line it all up. Now it