Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Help! All my plugin server modules have stopped working after a failure at AWS
#11
(02-03-2019, 08:40 PM)JonRussell Wrote: Hi Alberto,

So, I have reviewed the log files. I'm happy to attach them if it will help.
I removed lots of repetitive entries, and noticed some things.

I have this in pandora_server.log :

Code:
2019-02-03 18:09:28 pandora [V10] Decoding json macros from # 234 plugin command '/usr/share/pandora_server/util/plugin/sslcheck.sh'
2019-02-03 18:09:28 pandora [V9] Executing AM # 234 plugin command '/usr/share/pandora_server/util/plugin/sslcheck.sh pandora.xxx.co.uk'
2019-02-03 18:09:28 pandora [V10] Processing module 'SSL Expiry - pandora.xxx.co.uk' for agent ID 12.

2019-02-03 18:14:33 pandora [V10] Decoding json macros from # 234 plugin command '/usr/share/pandora_server/util/plugin/sslcheck.sh'
2019-02-03 18:14:33 pandora [V9] Executing AM # 234 plugin command '/usr/share/pandora_server/util/plugin/sslcheck.sh pandora.xxx.co.uk'
2019-02-03 18:14:33 pandora [V10] Processing module 'SSL Expiry - pandora.xxx.co.uk' for agent ID 12.

So, whatever runs the server modules *is* running. One of my modules is running successfully every 5 minutes.
However, the same module configured for other agents is not, and my other modules are not.

There is nothing in the pandora_server.error log. It is full of

Code:
Use of uninitialized value in concatenation (.) or string at /usr/lib/perl5/PandoraFMS/Core.pm line 4877.

Other than that, there is a single line :

Code:
2019-01-22 22:53:31 - pandora Starting Pandora FMS Server. Error logging activated.

So, I guess the question is, why are my modules not running, if the engine is running and one other module is running. I have tried deleting them and recreating them, but they don't work either.

Is there a query I can run against the DB to list all the modules configured to run ? Maybe my DB is corrupt ?

Thanks.

Jon.

Good morning JonRussell,

Can you please attach me the configuration of the module that works, and one that doesn't work please.

Here is a query to see what modules are in you database: select * from tagente_modulo, use in in the Interface of Pandora FMS.

You can see the status of your database in Diagnostic tool.

Alberto
 Reply
#12
Code:
select * from tagente_modulo where descripcion LIKE '%SSL%'

Results from query attached.
(hostnames have been changed to protect the innocent :-)

I cant see any differences between them ?
module #234 is running fine, and I have live data.
All the others are "Unknown" for 23 days.
module #239 is me creating a new module. Thats not running either.

I edited the script :

Code:
[[email protected]]# cat /usr/share/pandora_server/util/plugin/sslcheck.sh
#!/bin/bash

echo `date` $1 >> /tmp/sshcheck.log

if (($# != 1));
then
        echo "Syntax:  <https_host_name>"
        exit -1
else
...

and the log just contains the one that's running (#234). So, the script is definitely not being called for the other 5 instances ?

Code:
[[email protected]]# cat /tmp/sslcheck.log
Tue Feb 5 12:26:19 UTC 2019 pandora.xxxx.co.uk
Tue Feb 5 12:31:24 UTC 2019 pandora.xxxx.co.uk
Tue Feb 5 12:36:29 UTC 2019 pandora.xxxx.co.uk

Thanks.

Jon.


Attached Files


.xlsx   SQL Results.xlsx (Size: 13.02 KB / Downloads: 4)
 Reply
#13
(02-05-2019, 12:31 PM)JonRussell Wrote:
Code:
select * from tagente_modulo where descripcion LIKE '%SSL%'

Results from query attached.
(hostnames have been changed to protect the innocent :-)

I cant see any differences between them ?
module #234 is running fine, and I have live data.
All the others are "Unknown" for 23 days.
module #239 is me creating a new module. Thats not running either.

I edited the script :

Code:
[[email protected]]# cat /usr/share/pandora_server/util/plugin/sslcheck.sh
#!/bin/bash

echo `date` $1 >> /tmp/sshcheck.log

if (($# != 1));
then
       echo "Syntax:  <https_host_name>"
       exit -1
else
...

and the log just contains the one that's running (#234). So, the script is definitely not being called for the other 5 instances ?

Code:
[[email protected]]# cat /tmp/sslcheck.log
Tue Feb 5 12:26:19 UTC 2019 pandora.xxxx.co.uk
Tue Feb 5 12:31:24 UTC 2019 pandora.xxxx.co.uk
Tue Feb 5 12:36:29 UTC 2019 pandora.xxxx.co.uk

Thanks.

Jon.

Good morning JonRussell,

As you said, it seams to be all Ok with the same data as the one thats working. The only thing I can think about is an error in the configuration of the modules that are not the 234.

If its not to much, can yoy attach me:
1- Configuration of the module working and 1 not working.
2- Configuration of the agent of the module working and 1 of the not working.
It must be a little difference between one and the other that makes that failure..

Alberto
 Reply
#14
Hi Alberto.

Attached are the configs.
However, I'm not convinced its a config error (although I do hope it is) because when I create a new module they don't work either ?

But I hope you find something wrong in the configs.

Thanks.

Jon.


Attached Files


.png   SSLCheck - Working.png (Size: 175.98 KB / Downloads: 3)
.png   SSLCheck - Not Working.png (Size: 176.45 KB / Downloads: 3)
.png   Agent - Working.png (Size: 61.51 KB / Downloads: 6)
.png   Agent - Not Working.png (Size: 62.46 KB / Downloads: 5)
 Reply
#15
(02-11-2019, 10:38 AM)JonRussell Wrote: Hi Alberto.

Attached are the configs.
However, I'm not convinced its a config error (although I do hope it is) because when I create a new module they don't work either ?

But I hope you find something wrong in the configs.

Thanks.

Jon.

Good morning JonRussell,

In the Agent that is not working you aren't pointing to any server to do the checks, so it doesn't know how to do the monitoring.

Please try to put on the pandora server on the agent and please tell us the results.

Alberto
 Reply
#16
HA !
That's fixed it !
Thank you so much... :-)

All my agents and modules are now green !

Any idea why this might have happened suddenly after the AWS network failure ?
I certainly didn't go and uncheck them all !
:-)

Thanks anyway.

Regards,

Jon.
 Reply
#17
(02-12-2019, 04:08 PM)JonRussell Wrote: HA !
That's fixed it !
Thank you so much... :-)

All my agents and modules are now green !

Any idea why this might have happened suddenly after the AWS network failure ?
I certainly didn't go and uncheck them all !
:-)

Thanks anyway.

Regards,

Jon.

Hello JonRussell,

The only thing it occurs to me it is the server name was changed after AWS failure, and the agents didn't recognize the new one again. Or the agents were uploaded manually without the Server in their parameters.

Regards
 Reply


Users browsing this thread: 1 Guest(s)


(c) 2006-2018 Artica Soluciones Tecnológicas. Contents of this wiki are under Create Common Attribution v3 licence. | pandorafms.com | pandorafms.org

Theme © MyBB Themes