Reputation: 103
I have setup an EMR in AWS with PrestoDB installed on it, Earlier I was able to query with PrestoDB but somehow after a restart it stopped working and started giving following error "Error running command: Server refused connection: http://ip-*---.us-west-2.compute.internal:8889/v1/statement" I have looked into all configuration files and nothing seems to be wrong. I have also cross check Hive configuration files but could not get any success. Could anyone who has encountered similar issue can help me.
Upvotes: 5
Views: 891
Reputation: 338
Yes , You will have to restart presto on all machines . Adding to the note I would like you to mention that give it a shot installing open source presto on EMR using presto admin. It has a lot of functions which will help you to avoid such issues . Updating and maintaining the cluster is easy using presto admin
Upvotes: 2
Reputation: 5294
I know this is an old question, but I've run into this as well.
The likely reason is that you only restarted the Presto server on the coordinating node. You have to ssh into each core node and restart the Presto server there as well.
If this is a persistent Presto cluster, you would probably benefit from installing presto-admin. It's kind of a pain to set up the first time, but it makes this stuff much easier once it's in place.
Upvotes: 1