How to fix negative number for latency on check_vmware_api

Question Copied

Every few minutes, a check will return CRITICAL as it appears to be getting a negative number for latency:

CHECK_VMWARE_API.PL CRITICAL - io read latency=-1 ms | io_read=-1ms;40;90

The same check will then return OK a few seconds later:

CHECK_VMWARE_API.PL OK - io read latency=0 ms | io_read=0ms;40;90

Any idea why it might be getting a negative number?

Answer Copied

This problem can be fixed by collecting multiple samples of the data, described in the help text for the plugin:

-M, --maxsamples=<max sample count> Maximum number of samples to retrieve. Max sample number is ignored for historic intervals. Default value is 1 (latest available sample).

Note that check_vmware_api is already deprecated. Consider switching to check_vmware_v2 instead.