Bringing together the Apache Cassandra experts from the community and DataStax.

Want to learn? Have a question? Want to share your expertise? You are in the right place!

Not sure where to begin? Getting Started

 

question

vkayanala_42513 avatar image
vkayanala_42513 asked Erick Ramirez answered

What is causing "Backpressure rejection while receiving WRITES.WRITE" ?

Hello,

We are running with 3 DC's in a cluster (all are running on core Cassandra) and couple of keyspaces replicating over all 3 DC's and some only 2 DC's.

During the load tests we are noticing huge spike on Mutations and Backpressure rejections, resulting read and write timeouts for couple seconds. And this is happening once in entire load test (we use to do 2 hours load test) so that's little concerning us because if servers are overloaded this should happen very frequently right! during 2 hours load test.

DSE Version: 6.7.3
Workloads: Cassandra

I would like to understand, what is backpressure and how is this causing read and write timeouts?

Thanks in Advance.

cassandra
10 |1000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.

1 Answer

Erick Ramirez avatar image
Erick Ramirez answered

@vkayanala_42513 The backpressure mechanism in DSE 6 is the server's attempt to prevent being overloaded by by read/write requests. For the purposes of what you're asking, the backpressure is a symptom and not the cause.

The problem with your tests is that it is hitting the throughput limit of your cluster and it is overloaded, hence the dropped mutations. Cheers!

Share
10 |1000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.