Thanks for reaching Q/A and sharing your details. As per details you shared this happens because the system can't convert the ISO8601 format into a Timespan type, which is what Azure Service Fabric expects.
Even though PT5M is a valid ISO8601 duration, it seems that Azure Service Fabric isn’t able to handle it correctly. So, we need to use a different format that Azure can understand.
- Double-check that you're using the correct format for the timeout value. For example, instead of PT5M you should use 00:05:00 (which is a
TimeSpanformat). - Go back to your Bicep or ARM template and make sure you're setting the right field for the upgrade policy. For example, check that
rollingUpgradeMonitoringPolicy.healthCheckRetryTimeoutis the correct property to set for this timeout value.