Monitor Tab not auto updating devices

[dpoore]

I have installed Cacti 1.1.37 on RedHat 7 with thold and monitor plugins. When I go to the monitor tab in cacti the devices are not updating I have to run
" php -q cmd.php --first=1 --last=6 " from the command line. When I do this the time on the down device does update but only for 5 minutes even thou the device has been for 2 days.

Capture.JPG (107.1 KiB) Viewed 1955 times

Thanks for the help.

[dpoore]

I have also looked at the Cacti log and it says the following:

2018/05/17 14:26:00 - POLLER: Poller[1] Maximum runtime of 58 seconds exceeded. Exiting.
2018/05/17 14:26:00 - SYSTEM STATS: Time:59.2891 Method:spine Processes:1 Threads:1 Hosts:6 HostsPerProcess:6 DataSources:13 RRDsProcessed:0
2018/05/17 14:26:01 - SYSTEM THOLD STATS: Time:0.0738 Tholds:0 TotalDevices:6 DownDevices:1 NewDownDevices:0
2018/05/17 14:26:01 - SYSTEM MONITOR STATS: Time:0.17 Reboots:0 DownDevices:0 Notifications:0 Purges:0

[netniV]

What are your poller settings?

[dpoore]

My cron is the following:

*/5 * * * * <username> /usr/bin/php /usr/share/cacti/poller.php > /dev/null 2>&1

Settings for the Poller is as follows:
General
Data Collection Enabled = checked
Poller Type = spine
(If I change this to cmd.php it still does not update)
Poller Interval = 5 min
Cron Interval = 5 min
Maximum Concurrent Poller Processes = 1
Balance Process Load = checked
Disable increasing OID Check = not checked
SNMP Bulkwalk Fetch Size = 10

Spine Specific Execution Parameters
Invalid Data Logging = none
Maximum Threads per Process = 1
Number of PHP Script Servers = 1
Script and Script Server Timeout Value = 25
The Maximum SNMP OIDs Per SNMP Get Request = 10

Thanks

[cigamit]

Your real problem is this.
2018/05/17 14:26:00 - POLLER: Poller[1] Maximum runtime of 58 seconds exceeded. Exiting.

You are running over your polling time, so something is hanging it up. You need to figure that out otherwise the plugins ever execute at the end.
For spine, only having 1 process and 1 thread is bad. Change the number of processes to be equal to the number of CPU cores, and change the number of threads to 12. Change the number of PHP Script Servers to something higher than 1 (try 4 or 5) and change the timeout for it to 5 seconds.

Then lets see what your poller times are like.

[dpoore]

Ok will let you know.

Thanks

[dpoore]

Ok I made the changes you requested and here are the logs

2018/05/31 07:30:01 - SYSTEM MONITOR STATS: Time:0.24 Reboots:0 DownDevices:0 Notifications:0 Purges:0
2018/05/31 07:30:00 - SYSTEM THOLD STATS: Time:0.1004 Tholds:0 TotalDevices:6 DownDevices:1 NewDownDevices:0
2018/05/31 07:30:00 - SYSTEM STATS: Time:298.7413 Method:spine Processes:2 Threads:12 Hosts:6 HostsPerProcess:3 DataSources:61 RRDsProcessed:0
2018/05/31 07:30:00 - POLLER: Poller[Main Poller] Maximum runtime of 298 seconds exceeded. Exiting.

The monitor status still has not changed.

[netniV]

Again, your run time has exceed and the poller is exiting.

[dpoore]

Ok I think I got it.

I had to copy spine.conf.dist to /etc and I renamed it to spine.conf. everything seems happy now.

[netniV]

You need to have the spine.conf in either the current directory, /etc or specify the path when running spine. There is an option within Cacti to say where the configuration sits under Settings.