-
Type: Task
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: None
-
None
-
Cluster Scalability
-
Fully Compatible
-
0
We can get caught in an infinite loop in the autoRetryTxn function by not incrementing the txnNumber when we receive a SnapShotError.
One solution would be to update the exception handling to properly handle SnapshotErrors by starting a new transaction instead of continuing to run in the do-while loop.
A better long term solution would be to use withTxnAndAutoRetry instead of autoRetryTxn in multi_statement_transaction_all_commands.