Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Query on IMap with indexes returns stale data #359

Closed
joeplanisky-temboo opened this issue Nov 20, 2012 · 3 comments

Comments

Projects
None yet
3 participants
@joeplanisky-temboo
Copy link

commented Nov 20, 2012

Doing a predicate query on an IMap that has indexes sometimes returns stale data, while retrieving a value with IMap.get(key) always returns the most current value. Once a node begins returning stale data, it will always return that same stale value for all future queries. The problem seems to be triggered by another node joining the cluster while a query is in progress.

We have seen this issue in Hazelcast 2.1.3 and 2.4.

I will attach code to demonstrate the issue in a comment to this issue.

@joeplanisky-temboo

This comment has been minimized.

Copy link
Author

commented Nov 20, 2012

I've uploaded a program (and associated hazelcast.xml file) to demonstrate the issue at https://gist.github.com/4118374

To see the problem, run multiple (up to 6) instances of the Demo.java program. All instances may be run on the same machine in separate JVMs. When a query begins returning values that do not match the values returned by IMap.get(key), the program will display an error message.

I usually get the error by the time I launch the 4th instance, but sometimes I can get all 6 instances running without error. In that case, I kill (ctrl-c) and restart one instance until the error occurs.

@mdogan

This comment has been minimized.

Copy link
Member

commented Nov 21, 2012

@akommareddi

This comment has been minimized.

Copy link

commented Dec 3, 2012

Hi - Is there an expected release that the fix to this issue is targeted at? Our upgrade from 1.9 to the latest version has been held up by this issue, we would like to plan based on the the eta. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.