Search code examples
snappydata

Starting Snappydata by using inline arguments


On starting snappydata using the command line arguments I am getting the below error

ERROR 38000: (SQLState=38000 Severity=20000) (Server=/X.X.X.157[1528] Thread=ThriftProcessor-0) The exception 'com.gemstone.gemfire.cache.TimeoutException: The SnappyData catalog in hive meta-store is not accessible' was thrown while evaluating an expression.

Below are the scripts I have used to start the cluster.

./bin/snappy locator start -dir=work/localhost-locator-1 -bind-address=X.X.X.157 -client-bind-address=X.X.X.157 -hostname-for-clients=X.X.X.157

./bin/snappy leader start  -dir=work/localhost-lead-1  -locators=X.X.X.157[10334] -spark.executor.cores=1

./bin/snappy server start  -dir=work/localhost-server-1  -locators=X.X.X.157[10334] -bind-address=X.X.X.157 -client-bind-address=X.X.X.157 -J-Dgemfirexd.hostname-for-clients=X.X.X.157 -heap-size=1g

Solution

  • The error that I mentioned in my post was happening because of the sequence in which I starting the Snappydata members. The correct sequence is to start the locator first followed by server and then the lead. I have tested the same in a distributed environment as well.