[CASSANDRA-8621] For streaming operations, when a socket is closed/reset, we should retry/reinitiate that stream

• Type: Improvement
• Status: Resolved
• Priority: Normal
• Resolution: Won't Fix
• Component/s: Legacy/Streaming and Messaging
• Assignee: Unassigned
• Reporter: Jeremy Hanna
• Created: 14/Jan/15 18:03
• Updated: 16/Apr/19 09:31
• Resolved: 25/May/16 17:30

Description

Currently we have a setting (streaming_socket_timeout_in_ms) that will timeout and retry the stream operation in the case where tcp is idle for a period of time. However in the case where the socket is closed or reset, we do not retry the operation. This can happen for a number of reasons, including when a firewall sends a reset message on a socket during a streaming operation, such as nodetool rebuild necessarily across DCs or repairs.

Doing a retry would make the streaming operations more resilient. It would be good to log the retry clearly as well (with the stream session ID and node address).

Issue Links

• is superceded by: CASSANDRA-11841 Add keep-alive to stream protocol
• relates to: CASSANDRA-9294 Streaming errors should log the root cause
• links to: 0.1 patch for 2.1