I'm facing an issue with my Meta nodes failing with an s3 error: dispatch failure
in RisingWave v1.7.0 deployed on k8s using the operator. I was advised to set identity_resolution_timeout_s
to a longer time. How can I set this configuration in the RisingWave operator for Meta nodes? Do I need to set it for other nodes as well?
Neil
Asked on Apr 01, 2024
The identity_resolution_timeout_s
configuration can be set in the RisingWave operator for Meta nodes by updating the risingwave.toml
ConfigMap.
To set the identity_resolution_timeout_s
configuration, you can include it in the risingwave.toml
ConfigMap under the [storage.object_store.s3]
section.
It is recommended to set the identity_resolution_timeout_s
for all nodes, not just the Meta nodes, as this issue is related to S3 connectivity and can affect all components.
Ensure that the ConfigMap is connected to all nodes by including it at the beginning of the RisingWave spec.
If the identity_resolution_timeout_s
parameter doesn't take effect due to a bug, consider using the image ghcr.io/risingwavelabs/risingwave:v1.7.3-676e214
where the fix is implemented.