Open asoriano-lightbox opened 3 years ago
@apc999, this is something alluxio could help to fix?
Hi, Any update?
any update?
@apc999 how can use this feature ?
cc @imjalpreet
Sorry, missed this thread. I will have a look into it and submit a PR to fix the issue.
We are trying to deploy latest (0.250) PrestoDB on EKS but with the RaptorX feature on. We built a docker image using the the below Dockerfile:
This works great and we are able to create a cluster with 1 coordinator and multiple workers. We mount any config/properties files and catalogs at run time and we are able to access and query our hive metastore within our EKS cluster or within an EC2 instance (as a docker container) that has access to our s3/hive endpoints. However, when we try to enable the RaptorX feature the cluster starts as normal without any error but when we query the s3 parquet files tables that worked before without RaptorX we now get s3 is not supported as the external filesystem
Im following this article on how to enable RaptorX: https://github.com/prestodb/presto/issues/13205
I verified that we have all the jars necessary ( I think)
and also verified that wherever we deployed the cluster has access to our data such as the s3 buckets and the hive metastore endpoint. I tried many hive s3 parameters configurations but nothing seems to help with this error. Last thing i tried was giving it the aws access key and secret on the hive properties that we know has access to our s3. Hive.properties
config.properties
jvm.config
node.properties
I also tried the same thing with using ahanio's image and same error https://hub.docker.com/r/ahanaio/prestodb-sandbox
Not sure if we are misconfigured or is this a bug but i don't know what to try next.
Any help will be appreciated, thank you.