I have a memsql ciab cluster I use for local development.
I’ve used the latest instructions to spin up this cluster from github page
docker run -i --init \
--name memsql-ciab \
-e SINGLESTORE_LICENSE=$MEMSQL_LICENSE_KEY \
-e ROOT_PASSWORD=singlestore_doesnt_work \
-p 3306:3306 -p 8080:8080 \
--platform linux/amd64 \
ghcr.io/singlestore-labs/singlestoredb-dev:latest
The cluster is healthy when there are no queries running.
The moment I run a query, I start getting these errors.
uncategorized SQLException for SQL [SELECT col1, col2, col3, DOT_PRODUCT(vector, JSON_ARRAY_PACK(?)) AS score from table where vector IS NOT NULL and vector !='' order by score desc limit ?]; SQL state [HY000]; error code [1777]; (conn=659) Partition local_db:0 has no master instance. This is likely because the node or nodes that hold a copy of the partition are down. Check for offline leaf nodes by running SHOW LEAVES and bring them back online to restore access to the partition; nested exception is java.sql.SQLException: (conn=659) Partition local_db:0 has no master instance. This is likely because the node or nodes that hold a copy of the partition are down. Check for offline leaf nodes by running SHOW LEAVES and bring them back online to restore access to the partition
The query fails with an error, the node suddenly disconnects.
However within seconds of running that query, the node is able to re-attach and the cluster becomes healthy again.
Happens every time I try to run this query
Not sure why this doesn’t work and how to debug this.
Details of my setup
M1 Mac
jdk 11
singlestore-jdbc-client v1.2.2