Cacti 1.1.28 + Autom8 duplicate poller

[Atreides]

Hello,
I've installed Cacti 1.1.28 + SPINE and I'm having a weird issue with Autom8. After I made some templates for network discover(everything went well), and added all hosts to cacti I've notice that every time the poller is running, there's a duplicate process wich is killing my server :

30226 96.4 1.5 412324 31516 ? R 18:11 1:21 /usr/bin/php -q /var/www/html/cacti/poller_automation.php -M

This process is multiplied by 1 every 5 minute(when my cron is doing the job), and I cant seem to find out how to stop it. I've disabled everything in the Automation Tab

This are the rest of the processes regarding poller.

cacti 25759 0.0 0.0 9200 1176 ? Ss 17:40 0:00 /bin/sh -c php /var/www/html/cacti/poller.php > /dev/null 2>&1
cacti 25761 0.3 1.6 415216 34352 ? S 17:40 0:00 php /var/www/html/cacti/poller.php
cacti 25764 0.3 0.2 141772 5740 ? Sl 17:40 0:00 /usr/local/spine/bin/spine -C /usr/local/spine/etc/spine.conf --poller 1 --first 0 --last 178 --mibs
cacti 25775 0.0 0.1 61376 2104 ? S 17:40 0:00 /usr/bin/rrdtool -

From logs all this started when the mysql crashed.

2017-11-20 13:05:07 - SPINE: Poller[1] WARNING: MySQL is NOT Thread Safe!
2017/11/20 13:05:07 - POLLER: Poller[1] WARNING: There are '1' detected as overrunning a polling process, please investigate
2017/11/20 13:05:05 - SYSTEM STATS: Time:298.5614 Method:spine Processes:1 Threads:1 Hosts:178 HostsPerProcess:178 DataSources:2266 RRDsProcessed:1157
2017/11/20 13:05:05 - POLLER: Poller[1] Maximum runtime of 298 seconds exceeded. Exiting.
2017-11-20 13:03:14 - SPINE: Poller[1] ERROR: SQL Failed! Error:'2006', Message:'MySQL server has gone away', SQL Fragment:'UPDATE poller_time SET end_time=NOW() WHERE poller_id=1 AND pid=23534'
2017-11-20 13:03:14 - SPINE: Poller[1] ERROR: SQL Failed! Error:'2006', Message:'MySQL server has gone away', SQL Fragment:'replace into settings (name,value) values ('date',NOW())'
2017-11-20 13:00:07 - SPINE: Poller[1] WARNING: MySQL is NOT Thread Safe!
2017-11-20 13:00:07 - SPINE: Poller[1] WARNING: MySQL is NOT Thread Safe!
2017/11/20 12:58:57 - SYSTEM TOPX STATS: Time:46.1111 DT:1 DS:1099 CYCLES: 5699
2017/11/20 12:58:16 - SYSTEM MONITOR STATS: Time:2.45 Reboots:0 DownDevices:0 Notifications:0 Purges:0
2017/11/20 12:58:00 - SYSTEM STATS: Time:173.3131 Method:spine Processes:1 Threads:1 Hosts:178 HostsPerProcess:178 DataSources:2266 RRDsProcessed:1157

PS : I've tried repair database, rebuild poller cache. Now I'm killing everything related to php every 10 minutes, otherwise the server is crashing.

[phalek]

Why have you set spine to use 1 process and 1 thread only ? that seems to be pretty low ...

[Atreides]

Didnt use SPINE, untill lately and I thought its good to use more threads when you have performance issues. I can increase it, but that doesnt solve my problem and why the
/usr/bin/php -q /var/www/html/cacti/poller_automation.php -M is starting a new process very 5 minutes.

[cigamit]

That process is always going to start, as that is what performs the the automation scans. If you want to stop it, then you will need to disable all your Automation Networks. It will still launch, but will exit pretty quickly. You most likely have a Network set to be scanned every polling.

[Atreides]

That's what I tought, but I stopped every network(status is disabled for all). I'll try and delete them to see if it helps.


EDIT : Seems that deleting all the Networks solved the problem. I'll add a network and see if it happens again.