Good Morning,
I am hoping someone can provide me with some direction as I am a little stumped. I am using zabbix 5.0. I am using the DBMS SQL by ODBC template collections to monitor an always on configuration. One of the triggers (redo_queue_size) that is preconfigured fires all the time and is cluttering the dashboard. This trigger is created by a trigger prototype inside a Non-local database discovery. The trigger prototype is
I am hoping someone can provide me with some direction as I am a little stumped. I am using zabbix 5.0. I am using the DBMS SQL by ODBC template collections to monitor an always on configuration. One of the triggers (redo_queue_size) that is preconfigured fires all the time and is cluttering the dashboard. This trigger is created by a trigger prototype inside a Non-local database discovery. The trigger prototype is
| {Template DB MSSQL by ODBC Always On:mssql.non-local_db.redo_queue_size["{#GROUP_NAME}*{#REPLICA_NAME}*{#DBNAME}"].last(#1)}>{Template DB MSSQL by ODBC Always On:mssql.non-local_db.redo_queue_size["{#GROUP_NAME}*{#REPLICA_NAME}*{#DBNAME}"].last(#2)} and {Template DB MSSQL by ODBC Always On:mssql.non-local_db.redo_queue_size["{#GROUP_NAME}*{#REPLICA_NAME}*{#DBNAME}"].last(#2)}>{Template DB MSSQL by ODBC Always On:mssql.non-local_db.redo_queue_size["{#GROUP_NAME}*{#REPLICA_NAME}*{#DBNAME}"].last(#3)} and I added this final line, to try and cut down on random triggerings if the queue size goes from 2 to 5 to 6 ms (which are perfectly acceptable and frequent) and {Template DB MSSQL by ODBC Always On:mssql.non-local_db.redo_queue_size["{#GROUP_NAME}*{#REPLICA_NAME}*{#DBNAME}"].last(#3)}>10 I have tried deleting the existing triggers, and have them regenerated.. and it LOOKS like it takes as I can see the triggers on the host reflective of my change. However it's still triggering when we get a sequence like this (pulled from history of the Get non-local DB state) 2021-10-14 09:50:01 [{"log_send_queue_size":"0","redo_queue_size":"1 2", "group_name":"AppNameProdAVG","replica_name":" sql-AppName19-p1","dbname":"AppNameDMNSite"}] 2021-10-14 09:49:01 [{"log_send_queue_size":"0","redo_queue_size":"7 "," group_name":"AppNameProdAVG","replica_name":"sql-AppName19-p1","dbname":"AppNameDMNSite"}] 2021-10-14 09:48:01 ["log_send_queue_size":"0","redo_queue_size":"4 ","g roup_name":"AppNameProdAVG","replica_name":"sql-AppName19-p1","dbname":"AppNameDMNSite"] 2021-10-14 09:47:00 ["log_send_queue_size":"0","redo_queue_size":"9 ","g roup_name":"AppNameProdAVG","replica_name":"sql-AppName19-p1","dbname":"AppNameDMNSite"] With the old logic.. this fires as 12 > 7 > 4 with the new logic.. this should not fire.. as 4 !> 10 yet it still does fire. It also doesn't help that these triggers take HOURS to recreate.. when it should be every hour that the discovery happens as far as my understanding goes, and the create enabled triggers button doesn't appear to do anything. Thank you in advance if anyone could provide some sort of guidance. I am just a user on this application so I don't have access to logs on the backend, but if the thought it that something should be generated somewhere that would give me a clue, I can hit up my system admins to provide me with specific log files. |