Current transaction is aborted, commands ignored until end of transaction block

[rahul-bansal-sumo]

Getting current transaction is aborted SQLException, whenever there is upscale/downscale of worker pods in kubernetes. Although we have implemented graceful shutdown for worker pods but still issue persist.

Current transaction is aborted SQLException also happening because of another reasons like timeout Interrupted exception and network problem. But we are concerned about queries which are getting aborted after few milliseconds of query start.
Exception code pointer -

trino/core/trino-main/src/main/java/io/trino/transaction/InMemoryTransactionManager.java

Line 407 in d2c414a

throw new TrinoException(TRANSACTION_ALREADY_ABORTED, "Current transaction is aborted, commands ignored until end of transaction block");

As per our understanding io.airlift.concurrent.BoundedExecutor is used to queue the tasks and then runs it on trino worker pod. But due to unavailable worker pods, I think trino used fail-fast mechanism and aborted such queries initially itself.

We are using trino 469 version.