We have a template that has 15 Discovery Item Prototypes configured. 14 of those items have been working correctly for a few years now. We added a new item on Thursday which the zabbix_sender refuses to upload data for.
We have a bash script on the hosts that runs every 10 minutes and outputs it's values into a zabbix_input file that the zabbix_sender is uploading using the '-i' flag.
The current Items are configured as such:
Name: Query timestamp: {#TABLE.NAME}
Type: Zabbix trapper
Key: query.time[{#TABLE.NAME}]
Type of info: Numeric (unsigned)
Data Type: Decimal
Units: unixtime
Example output in the zabbix_input file that's getting uploaded:
This item is getting it's value from a trapper item that monitors a timestamp of a file that gets updated every time a query succeeds. There are multiple table names which are discovered daily. This item has been working great for years.
The NEW item that refuses to work is configured almost identically:
Name: Query Success timestamp: {#TABLE.NAME}
Type: Zabbix trapper
Key: query-success.time[{#TABLE.NAME}]
Type of info: Numeric (unsigned)
Data Type: Decimal
Units: unixtime
Example output in the zabbix_input file that's getting uploaded:
This item is getting it's value from a trapper item also, it monitors a separate file that's timestamp is updated every time the query succeeds (sometimes the query fails, so we want to know when the last time it succeeded was).
So more about the problem:
When invoking the zabbix_sender command to upload the input file, we get successes for all items OTHER THAN the new item. So let's say it's monitoring 10 table name files, I will get 10 successes (for the working items) and 10 failures (for the new items). The items are the same besides the name, the input is the same, the data types are the same, and it's just not working.
Example error output:
Any and all help would be appreciated. I gave it a few days to see if the sync configuration would update and it would work, but that's not the case 4 days later.
We have a bash script on the hosts that runs every 10 minutes and outputs it's values into a zabbix_input file that the zabbix_sender is uploading using the '-i' flag.
The current Items are configured as such:
Name: Query timestamp: {#TABLE.NAME}
Type: Zabbix trapper
Key: query.time[{#TABLE.NAME}]
Type of info: Numeric (unsigned)
Data Type: Decimal
Units: unixtime
Example output in the zabbix_input file that's getting uploaded:
Code:
- query.time[example table] 1457032243
The NEW item that refuses to work is configured almost identically:
Name: Query Success timestamp: {#TABLE.NAME}
Type: Zabbix trapper
Key: query-success.time[{#TABLE.NAME}]
Type of info: Numeric (unsigned)
Data Type: Decimal
Units: unixtime
Example output in the zabbix_input file that's getting uploaded:
Code:
- query-success.time[example table] 1457032243
So more about the problem:
When invoking the zabbix_sender command to upload the input file, we get successes for all items OTHER THAN the new item. So let's say it's monitoring 10 table name files, I will get 10 successes (for the working items) and 10 failures (for the new items). The items are the same besides the name, the input is the same, the data types are the same, and it's just not working.
Example error output:
Code:
16:54:16: # zabbix_input.sh
zabbix_sender [32540]: DEBUG: answer [{
"response":"success",
info from server: "processed: 10; [B]failed: 10[/B]; total: 20; seconds spent: 0.005430"
sent: 20; skipped: 0; total: 20
Comment