Cacti-1.2.25 not showing devices down

[eholz1]

Hello members,
A true anomaly. I have a new install of cacit-1.2.25 running on Oracle 9.
I copied the database from an existing running cacti-1.2.19. this preserved the hosts, graphs, etc.
I have a query that modifies the poller rra updates to correct graphs are produced. I do not
copy the rrd files from old to new. the cacti-1.2.25 correctly recreates the graphs in the rra directory (assuming I get the permissons correct!)

I use "SNMP or Ping" to determine device status. A device can be down, and cacti show, when I click the device
in "Manage/Devices" this is down, the screen shows inred - there is no snmp response (or ping if device is shut down).
BUT, the device does not show as down in the cacti log, or the Devices summay, and does not send email, and does not show up in the "thold,Device Status" tab.

Any ideas here?
Thanks
eholz1

[TheWitness]

This is a Thold bug then. What version of Thold?

[eholz1]

Ah, mister Witness,

I will check, and get the LATEST thold I can find.

Thanks,

Eholz1

[TheWitness]

Keep us posted. I hope it's better now.

[eholz1]

Hello Again, Thanks for the help.

No disce. No Joy. I downladed the thold plugin from git hub, ver 1.8.
Still have the same issue. I run the poller every minute, and cron every minute.

I have cacti 1-2.19 running with thold ver 1.5.2 - which works fine.

Perhaps there is a version one back from 1.8>

Thanks Again,
Eholz1

[eholz1]

Hello Members,

I installed the thold ver 1.5.2, still not luck. The permissions are set to root.
Do the permissions need to be "apache"?

Thanks Again,
eholz1

[TheWitness]

Log a bug to GitHub at https://github.com/cacti/plugin_thold

[eholz1]

Hello Again,

Not sure it is a bug. It might be "pilot error". I copied a cacti database from cacti-1.2.19
using mysqldump to get a sql file to create the new database, and corrected the poller_item table to use the
correct rra_path.

This seemed to work fine, and graphs are being created in the rra path. I did not copy any *rrd files.
After this I noticed that graphs were there, but no device would detect as down.
I ran a test, an added a device that was not in the "old" database, I could change this device to
respond or not respond to snmp command, etc.
This device when "down" shows up in the thold tab, and shows as "down" in the :"Manage/Devices" area.

I used the msyqldump to create an sql file since I have 500 hosts to add!
Perhaps this "show cut" does not work??
Thanks for support on this, Let me know your thoughts on this.
eholz1

[TheWitness]

Pilot error in life happens. We are all thankful when it does not cost us our lives and the lives of our passengers. Be well.

[salinois]

hello guys,
I have the same issue, my devices don't respond to ping but cacti show "UP". When the device don't respond to ping, no graph displaying.
cacti version: 1.2.26
thod version : 1.8

this is a new install and new database; i have only ten devices for the moment.

is it necessary to have plugin thold for the status of UP or Down devices?

salinois

[salinois]

hello again,
it works now, I had incorrectly configured the "Downed Device Detection" box in device.

salinois