fix: prevent tx_executor reading from sock if operation is canceled #4704
Conversation
BorysTheDev
changed the title
| @@ -67,8 +67,6 @@ class ClusterShardMigration { | |||
|
|
|||
| auto tx_data = tx_reader.NextTxData(&reader, cntx); | |||
| if (!tx_data) { | |||
| in_migration_->ReportError(GenericError("No tx data")); | |||
There was a problem hiding this comment.
why do you need this?
if there is no transaction data you need to report error? if its not running doesnt it mean that error was already reported?
There was a problem hiding this comment.
we do report inside tx_reader.NextTxData(&reader, cntx); If the cntx is canceled we don't need report. If we have error it is already reported
There was a problem hiding this comment.
So why do you need this?
There was a problem hiding this comment.
I didn't get the question. I'v removed it, because we don't need it
There was a problem hiding this comment.
ops sorry my eys are playing with me
| break; | ||
|
|
||
| std::optional<TransactionData> tx_data; | ||
| while ((tx_data = tx_reader.NextTxData(&reader, cntx))) { |
There was a problem hiding this comment.
here it looks like we did not call the NextTxData if the cntx was closed..
so the problem is in cluster code only where you do not have such if as there was here?
There was a problem hiding this comment.
There is no problem here. just decided to make it shorter
BorysTheDev
deleted the
prevent_tx_executor_read_from_socket_if_op_is_canceled
branch
fixes: #4689
problem: we read from the socket after it was shutdown