[
https://issues.apache.org/jira/browse/GEODE-109?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14654558#comment-14654558
]
ASF GitHub Bot commented on GEODE-109:
--------------------------------------
Github user sbawaska commented on a diff in the pull request:
https://github.com/apache/incubator-geode/pull/13#discussion_r36257113
--- Diff:
gemfire-core/src/main/java/com/gemstone/gemfire/internal/redis/ExecutionHandlerContext.java
---
@@ -179,31 +203,21 @@ public void executeCommand(ChannelHandlerContext ctx,
Command command) throws Ex
*
* @param exec Executor to use
* @param command Command to execute
- * @param cache Cache instance
- * @param client Client data associated with client
- * @param n Recursive max depth of retries
* @throws Exception Throws exception if exception is from within
execution and not to be handled
*/
- private void executeWithoutTransaction(final Executor exec, Command
command, int n) throws Exception {
- try {
- exec.executeCommand(command, this);
- } catch (RegionDestroyedException e) {
- if (n > 0)
- executeWithoutTransaction(exec, command, n - 1);
- else
- throw e;
+ private void executeWithoutTransaction(final Executor exec, Command
command) throws Exception {
+ for (int i = 0; i < MAXIMUM_NUM_RETRIES; i++) {
+ try {
+ exec.executeCommand(command, this);
+ return;
+ } catch (Exception e) {
+ if (e instanceof RegionDestroyedException || e.getCause()
instanceof QueryInvocationTargetException)
+ Thread.sleep(WAIT_REGION_DSTRYD_MILLIS);
+ }
}
+ throw new Exception("Could not execute command");
--- End diff --
Can you please re-throw the underlying exception here.
> Bugs in redis adapter when running with multiple vms
> ----------------------------------------------------
>
> Key: GEODE-109
> URL: https://issues.apache.org/jira/browse/GEODE-109
> Project: Geode
> Issue Type: Bug
> Components: core
> Affects Versions: 1.0.0-incubating
> Reporter: Vitaliy Gavrilov
>
> 1) The meta data attached with the redis lists implementation does not always
> stay synchronized with the list.
> 2) Queries run by sorted set requests fail in execution
> 3) Potential deadlock when regions are created simultaneously in different vms
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
