Pandora FMS Community

[SOLVED] Lots of modules are showing as non initialized and status unknown

CyclopsElectronics · 2 oct 2018, 20:08 · 16 mensajes · 85 visitas

  1. Inicio del debate

    Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    [

    ](//pandorafms.com/community/media/forum/10546/26527=2466-uknown.PNG)

  2. Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

  3. Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

    Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    [

    ](//pandorafms.com/community/media/forum/10546/26537=2469-lag.PNG)

  4. Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

    Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

  5. Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

    Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

    Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

  6. Hi I have the problem where lots of modules are showing as non initialized and status unknown. Even modules from the Pandora server itself. Some of the modules not working on one server are working on another server. I've checked pandora_server.log and it looks fine. "2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Launching 7.0NG.727 180925 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] Config options: Logfile at C:PandoraFMSPandora_Serverlogspandora_server.log, Basepath is C:PandoraFMSPandora_Serverpandora_server.conf, Checksum is 0, Master is 1, SNMP Console is 1, Server Threshold at 5 sec, verbosity at 1, Alert Threshold at 60, ServerName is 'GBYOR-VSMGMT1' 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Pandora FMS Enterprise module not available.
    2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] No SNMP trap daemon configured. Start snmptrapd manually. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] SNMP log file C:PandoraFMSPandora_Serverlogspandora_snmptrap.log not found. 2018-10-02 16:41:08 GBYOR-VSMGMT1 [V1] - Netflow daemon disabled.
    2018-10-02 16:41:09 GBYOR-VSMGMT1 [V1] Server GBYOR-VSMGMT1 is the current master." I've tried restarting the Pandora server, and making sure the Pandora and tentacle services are running. I can see that files are coming in to data_in and going out, but not all of them. For example there are files in there from the last couple of months, I don't think this is expected behaviour? I've also made sure that the Pandora server, console and database have all been updated. I've checked that the credentials for the db both match in pandora_server.conf and config.php. I've also ran mysqldump.exe with the same credentials to make sure they work, and they do. In addition (possible red herring), I have data files from an agent on a server that are stuck in the temp folder. Though if I run a manual tentacle_client.exe test from this agent server to the Pandora server it transfers a file fine. Any hints on what might be the cause, or what I can examine? Again there are modules on the Pandora server itself that are showing as unknown. I've attached a screenshot from the web console of the module list for the Pandora server.

    Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

    Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

    Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

    Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

  7. Good morning CyclopsElectronics, Can you please attach me the log file to see the possible errors. Have you seen if all the servers are correct¿? It could be caused because the data server is fallen. Please look that all servers are correct: Alberto

    Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

    Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

    Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

    Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

  8. Hi Alberto The Pandora server and the data/database for it are on the same server. I've attached the log for you. Maybe it's the lag? Here is a screenshot and it shows 20 minutes lag. I've attached the screenshot, didn't know how to post it in the forum message.

    Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

    Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

    Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

    Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

    Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

  9. Good morning CyclopsElectronics, Have you got many agents/modules in your instalation? Can you try to put the dataserver_threads to 3 for example and see if the lag goes out, or at least less. Tell me the result. Alberto

    Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

    Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

    Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

    Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

    Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

  10. Hi Alberto Thank you for the advice, this has reduced the lag from 20 minutes to a few seconds. However I still have 145 monitors unknown and 117 monitors initialised. 190 monitors are working.

    Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

    Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

    Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

    Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

    Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

  11. Good morning CyclopsElectronics, Try the to increase yo 5 the threads, no more than 5, and try to increased some other threads as the icmp and snmp. Alberto

    Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

    Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

    Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

    Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

    Hi Alberto I've not increased the threads anymore, and surely but slowly the number of unknowns are going down and the normal status is going up. Only 68 unknowns remain. For the non initialised ones I'm only using the built in modules, I haven't added any custom ones. I've attached the server conf for the Pandora server. I've pm'ed you the password for the archive. Do you need any more info to help troubleshoot?- Pandora.zip

  12. Hi Alberto Before making any further changes, I checked the status and had the following: 227 normal, 110 unknown, 117 non-initialised I think the number of threads is related to the problem, because the 227 showing as normal, is higher than it has been since I opened this community post, which was 190 at the start. I've changed the threads from 3 to 5 (and from 1 to 5 for all other thread options) and restarted the Pandora Server service. Immediately I have 228 normal (increased from 1). Should I try setting them to 10?

    Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

    Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

    Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

    Hi Alberto I've not increased the threads anymore, and surely but slowly the number of unknowns are going down and the normal status is going up. Only 68 unknowns remain. For the non initialised ones I'm only using the built in modules, I haven't added any custom ones. I've attached the server conf for the Pandora server. I've pm'ed you the password for the archive. Do you need any more info to help troubleshoot?

    Good morniing CyclopsElectronics, Your .conf is ok. We don´t understand why do you have that non-iniciated modules. Please tell me whta type of modules are non-iniciated to try to replicate it on our system. Alberto

  13. Good morning CyclopsElectronics, Usually the threads are really important on the way the agents and monitors work. Also a way of looking if something went wrong is the sincronization of the time between the console and the server, please check its ok. Never increase more than 5 the threads of any server, because this will caused many problems on pandora, going to crush the software. Alberto

    Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

    Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

    Hi Alberto I've not increased the threads anymore, and surely but slowly the number of unknowns are going down and the normal status is going up. Only 68 unknowns remain. For the non initialised ones I'm only using the built in modules, I haven't added any custom ones. I've attached the server conf for the Pandora server. I've pm'ed you the password for the archive. Do you need any more info to help troubleshoot?

    Good morniing CyclopsElectronics, Your .conf is ok. We don´t understand why do you have that non-iniciated modules. Please tell me whta type of modules are non-iniciated to try to replicate it on our system. Alberto

    Hi Alberto Today the number of unknown modules is 33, number of known modules is 308 (again this is going in the right direction). However the number of non initialised modules has gone up 1 from 117 to 118. The modules that are non initialised are: CPU Load TCP_Connection DiskUsed_C: DiskUsed_D: Memory_Used Network_Usa...e_Bytes System_Load_AVG Maybe there's a backlog? Maybe if I leave it long enough the number of unknown modules will keep decreasing to 0 and then the number of non initialised modules will start going down?

  14. Hi Alberto The time on the server and the console is the same, as the console and the Pandora server are installed on the same server. Also, for all of our servers we synchronise the time with a central time server, so all of the clocks of all servers are the same. Increasing the threads seems to be helping, after I increased them yesterday, I've come in this morning and now have 246 modules showing as normal, 91 unknown and 117 uninitialised. What I have noticed is that increasing the threads increases the number of modules showing as normal, and decreases the number of modules showing as unknown. However the number of non initialised modules has remained at 117. Is there anything I can do to force them to reinitialise/troubleshoot these non initialised modules? Also if increasing the threads is helping, why is that? Have I done something wrong in the configuration? Thank you.

    Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

    Hi Alberto I've not increased the threads anymore, and surely but slowly the number of unknowns are going down and the normal status is going up. Only 68 unknowns remain. For the non initialised ones I'm only using the built in modules, I haven't added any custom ones. I've attached the server conf for the Pandora server. I've pm'ed you the password for the archive. Do you need any more info to help troubleshoot?

    Good morniing CyclopsElectronics, Your .conf is ok. We don´t understand why do you have that non-iniciated modules. Please tell me whta type of modules are non-iniciated to try to replicate it on our system. Alberto

    Hi Alberto Today the number of unknown modules is 33, number of known modules is 308 (again this is going in the right direction). However the number of non initialised modules has gone up 1 from 117 to 118. The modules that are non initialised are: CPU Load TCP_Connection DiskUsed_C: DiskUsed_D: Memory_Used Network_Usa...e_Bytes System_Load_AVG Maybe there's a backlog? Maybe if I leave it long enough the number of unknown modules will keep decreasing to 0 and then the number of non initialised modules will start going down?

    Good morning CyclopsElectronics, All that modules goes on the Data server, so its possible, as you said, that it will decrease the number, but is now is having some problems, probably its never going to reach 0. A possible solution i¡s to introduce more RAM memory to your computer to make the data tranfer faster so all modules can reach Pandora FMS before the interval. Another possible solution could be to increased the Interval Time for the modules to, lets say, 15 minutes, so all modules can receive data before the next interval. You can change it with the bulk operations massively. Tell me the new results please. Alberto

  15. Good morning CyclopsElectronics, If you can send me the server.conf to see what could go wrong it would be of great help. Usuarlly the non-iniciated modules can be because of the: luck of threads, wrong "exec" configuration, bad module configuration or even bad agent configuration. When creation, for example, a data module, please try the command directly on the server to see if it goes Ok and then put it on the module. Alberto

    Hi Alberto I've not increased the threads anymore, and surely but slowly the number of unknowns are going down and the normal status is going up. Only 68 unknowns remain. For the non initialised ones I'm only using the built in modules, I haven't added any custom ones. I've attached the server conf for the Pandora server. I've pm'ed you the password for the archive. Do you need any more info to help troubleshoot?

    Good morniing CyclopsElectronics, Your .conf is ok. We don´t understand why do you have that non-iniciated modules. Please tell me whta type of modules are non-iniciated to try to replicate it on our system. Alberto

    Hi Alberto Today the number of unknown modules is 33, number of known modules is 308 (again this is going in the right direction). However the number of non initialised modules has gone up 1 from 117 to 118. The modules that are non initialised are: CPU Load TCP_Connection DiskUsed_C: DiskUsed_D: Memory_Used Network_Usa...e_Bytes System_Load_AVG Maybe there's a backlog? Maybe if I leave it long enough the number of unknown modules will keep decreasing to 0 and then the number of non initialised modules will start going down?

    Good morning CyclopsElectronics, All that modules goes on the Data server, so its possible, as you said, that it will decrease the number, but is now is having some problems, probably its never going to reach 0. A possible solution i¡s to introduce more RAM memory to your computer to make the data tranfer faster so all modules can reach Pandora FMS before the interval. Another possible solution could be to increased the Interval Time for the modules to, lets say, 15 minutes, so all modules can receive data before the next interval. You can change it with the bulk operations massively. Tell me the new results please. Alberto

    Hi Alberto The Pandora server has 8 GB of RAM and 4 CPU cores, these hardly get used and never get maxed out. I have this command that runs once an hour on the Pandora server - c:PandoraFMSPandora_Serverbinpandora_db.exe c:pandoraFMSPandora_Serverpandora_server.conf. If I run this manually it gives good output/doesn't look like it fails. However when I log in to the web console it tells me I have 177,798 events and gives the warning "This system has too many events in the database". Could this be related?

  16. Hi I fixed the issue. I now have 443 normal modules, none unknown and none non-initialised. For the unknown issue there was two fixes. The first was to increase the number of threads as advised by Alberto. This fixed the majority of the unknown modules. For the remainder, it was caused when I upgraded the agent on the servers. During the upgrade process the pandora.conf file is kept and a new one is created called pandora.conf.new. I deleted pandora.conf on some of them, and renamed pandora.conf.new to pandora.conf. I then deleted the new agent that appeared in Pandora as I saw it as a duplicate. This meant the Pandora agent on some servers were reporting to an agent ID that no longer existed. To fix this I copied the agent ID from the agent I didn't delete, and pasted it in to the pandora.conf, and restarted the pandora fms agent. For the non-initialised modules, the issue was that I had uninstalled the pandora agent from these machines. Installing the agent, but not starting the Pandora FMS Agent yet, copying the agent ID for each of these machines in to there pandora.conf, and then starting the Pandora FMS service fixed the issue.

Entra en tu cuenta para publicar.