StarGate | mcpc proxy ) that allows commuinication | Game Engine library

The REST API does not currently support ALLOW FILTERING but you have some other options.

• If usernames are unique in your system and you’re able to alter the data model you could make user_username the partition key. This will allow you to query by that column without needing filtering.
• Another option if you can’t change the data model would be to add an index on user_username. The index will enable you to maintain your existing access patterns while also allowing you to query by the username.

Stargate does not currently support advanced workloads like Search and Graph. I think you might have to drop and recreate that keyspace without the Solr index for it to work again since the schema still exists on the other nodes.

This issue has been documented here. There has also been a request made to support Solr here.

So when you run your query for the initial 20, you'll get a property in the returned JSON called pageState. Here's an example with page-size=5:

At this time no. I've been meaning to log a Github issue on this, it should be a pretty simple change. I'll get an issue logged and update this answer with the link.

EDIT: Here's the issue.

Found the error in this typo:

The Stargate REST API equivalent of LIMIT is the page-size query parameter documented here

https://stargate.io/docs/stargate/1.0/attachments/restv2.html#operation/getRows

By default this token will persist for 30 minutes on a sliding window. This means that if you create a token and then wait 29 minutes before using it to authenticate a request to any of the APIs the time to live will be reset to 30 minutes. Likewise, if you wait 31 minutes after creating the token to use it then you'll receive an error because the token no longer exists. The length of time that the token persists can be configured using the stargate.auth_tokenttl system property. As example start up configuration that would set the tokens to persist for 100 seconds would be as follows.

To respawn your service when it fails, add the following to the [Service] block:

Yes, Stargate can handle high traffic.

Stargate isn't going to be the bottleneck since it's really just another client. What matters is the throughput of your cluster.

You need to size your cluster to handle the throughput you require. For example, if your 4-node cluster can handle 100K operations per second but your app requires 200K ops/s then you need to double the number of nodes in your cluster. Cheers!