geode-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <>
Subject [jira] [Commented] (GEODE-4033) TxFailover doesn't throw exception when tx state is not found on the server
Date Sat, 02 Dec 2017 00:37:00 GMT


ASF GitHub Bot commented on GEODE-4033:

pivotal-eshu commented on a change in pull request #1115: GEODE-4033: Exception is thrown
when Tx Id is not found on server dur…

 File path: geode-core/src/main/java/org/apache/geode/cache/client/internal/
 @@ -251,21 +251,8 @@ private Object executeWithServerAffinity(ServerLocation loc, Op op)
     int transactionId = absOp.getMessage().getTransactionId();
     // for CommitOp we do not have transactionId in AbstractOp
     // so set it explicitly for TXFailoverOp
-    try {
 Review comment:
   The TXFailover is used when a client lost connection to a server. The client will fail
over to another server, and the TXFailover is for the new server to find if the original server
who hosting the transaction still exist. If it exist, it will let client continue the transaction.

   This fix only eliminates the scenario when the new server does not find any server hosting
the transaction but client finds it is the first operation of the transaction and does the
retry instead of failing the transaction.

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:

> TxFailover doesn't throw exception when tx state is not found on the server
> ---------------------------------------------------------------------------
>                 Key: GEODE-4033
>                 URL:
>             Project: Geode
>          Issue Type: Bug
>          Components: transactions
>            Reporter: Anilkumar Gingade
>            Assignee: Anilkumar Gingade
> When a client transaction gets an exception due to connection failure; it tries to execute
the transaction on other available server. This failover transaction is processed, if the
tx.state is available on any of the servers; if not an exception thrown back to the client.

> Currently the client catches this exception and tries to re-create the exception instead
of returning back to the application. This could cause multiple transaction state to be created
for the same transaction id on the cluster (if the first attempt to the server is still in
process), causing unexpected behavior.

This message was sent by Atlassian JIRA

View raw message