Hi All
What is the result of setting read-timeout=0 in gemfire client cache xml pool
For ex:
<pool name="testPool" read-timeout="0" retry-attempts="5">
<server host="test" port="1000" />
</pool>
I ask because of this bug note from http://www.vmware.com/support/vfabric-gemfire/doc/BugsFixedGemFire66.html
| 02/12/07 | #36349 | Bridge client region.put() completes without exception, but entry value is not updated at the server | 5.0 | closed | Server's entry value is not updated although client region.put completes without exception | This happens when operations are performed out of order on the server. The problem arises from this sequence of events: 1. A client attempts to put value X one or more times, but each attempt times out. 2. Each failed attempt "orphans" a thread on the server. 3. The client picks a new connection (and its associated server thread) and continues to perform its sequential updates (X+1, X+2, ... X+n). 4. The orphan threads are eventually scheduled and successfully perform the put with value X, overwriting the previous values (X+1 or X+2 or X+n). |
Disable timeout behavior for the BridgeLoader, BridgeWriter, or BridgeClient by setting its "readTimeout" parameter to zero. This causes all Region operations supported by the client to block until the server has finished with the operation, preserving client ordering. The "retryAttempts" configuration will still be used when there are communication failures with the server or when the server cache closes in the midst of the operation.
Does this issue still exist with gemfire 6.5.1 ?
Thanks