[jira] [Assigned] (GEODE-2123) Replace static members in ParallelgatewaySenderQueues with instance variables
[ https://issues.apache.org/jira/browse/GEODE-2123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Huynh reassigned GEODE-2123: -- Assignee: Jason Huynh > Replace static members in ParallelgatewaySenderQueues with instance variables > - > > Key: GEODE-2123 > URL: https://issues.apache.org/jira/browse/GEODE-2123 > Project: Geode > Issue Type: Bug > Components: wan >Reporter: nabarun >Assignee: Jason Huynh > > All parallel sender queues shared one BatchRemovalThread, which led to the > case that if BatchRemovalThread thread was killed by one sender, all > secondary sender queues will keep growing as no prq remove messages will be > sent to them. > Also, the static variables in ParallelGatewaySenderQueue needs to be > converted in to instance variables, so that the senders don't share these > variables and work independent of each other. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Assigned] (GEODE-2182) Possible race condition when estimating index size and an entry is destroyed
[ https://issues.apache.org/jira/browse/GEODE-2182?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Jason Huynh reassigned GEODE-2182: -- Assignee: Jason Huynh > Possible race condition when estimating index size and an entry is destroyed > > > Key: GEODE-2182 > URL: https://issues.apache.org/jira/browse/GEODE-2182 > Project: Geode > Issue Type: Bug >Reporter: Jason Huynh >Assignee: Jason Huynh > > There is a possible race condition when estimating size of an index and an > entry is destroyed. This doesnt get caught and instead percolates to the top > as an EntryDestroyedException. > A quick work around would be to catch the exception and return 0, this may > lead to an inefficient lookup but the results would be correct. > The better alternative would be to continue iterating if possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (GEODE-2182) Possible race condition when estimating index size and an entry is destroyed
Jason Huynh created GEODE-2182: -- Summary: Possible race condition when estimating index size and an entry is destroyed Key: GEODE-2182 URL: https://issues.apache.org/jira/browse/GEODE-2182 Project: Geode Issue Type: Bug Reporter: Jason Huynh There is a possible race condition when estimating size of an index and an entry is destroyed. This doesnt get caught and instead percolates to the top as an EntryDestroyedException. A quick work around would be to catch the exception and return 0, this may lead to an inefficient lookup but the results would be correct. The better alternative would be to continue iterating if possible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Comment Edited] (GEODE-523) PartitionListener requires tests and clean up
[ https://issues.apache.org/jira/browse/GEODE-523?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15723272#comment-15723272 ] Sai Boorlagadda edited comment on GEODE-523 at 12/5/16 8:33 PM: [~dschneider] I am thinking of removing existing example and add some new ones? Are there any ideas? Or Some other ways to create primaries if PartitionManager.createPrimaryBucket is deprecated? was (Author: sboorlagadda): [~dschneider] Are there other ways to create primaries if PartitionManager.createPrimaryBucket is deprecated? What are other user-cases we can showcase that this Listener can be used for? > PartitionListener requires tests and clean up > - > > Key: GEODE-523 > URL: https://issues.apache.org/jira/browse/GEODE-523 > Project: Geode > Issue Type: Test > Components: regions >Reporter: Kirk Lund >Assignee: Sai Boorlagadda > > Many users are using PartitionListener. This API requires tests, proper > javadocs and general cleanup. > See GEODE-390 and GEODE-72 -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Assigned] (GEODE-523) PartitionListener requires tests and clean up
[ https://issues.apache.org/jira/browse/GEODE-523?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sai Boorlagadda reassigned GEODE-523: - Assignee: Sai Boorlagadda > PartitionListener requires tests and clean up > - > > Key: GEODE-523 > URL: https://issues.apache.org/jira/browse/GEODE-523 > Project: Geode > Issue Type: Test > Components: regions >Reporter: Kirk Lund >Assignee: Sai Boorlagadda > > Many users are using PartitionListener. This API requires tests, proper > javadocs and general cleanup. > See GEODE-390 and GEODE-72 -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Resolved] (GEODE-2178) Docs: Update HTTP session management version numbers
[ https://issues.apache.org/jira/browse/GEODE-2178?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Dave Barnes resolved GEODE-2178. Resolution: Fixed > Docs: Update HTTP session management version numbers > > > Key: GEODE-2178 > URL: https://issues.apache.org/jira/browse/GEODE-2178 > Project: Geode > Issue Type: Improvement > Components: docs >Reporter: Dave Barnes >Assignee: Dave Barnes >Priority: Minor > > User Manual section Tools andModules -> HTTP Session Management contains > outdated information regarding supported application servers and their > version numbers. Bring these up to date. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (GEODE-2169) Subclasses of OperationContext should also be deprecated
[ https://issues.apache.org/jira/browse/GEODE-2169?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Anilkumar Gingade updated GEODE-2169: - Component/s: (was: core) security > Subclasses of OperationContext should also be deprecated > > > Key: GEODE-2169 > URL: https://issues.apache.org/jira/browse/GEODE-2169 > Project: Geode > Issue Type: Task > Components: docs, management, security >Reporter: Kevin Duling > Fix For: 1.1.0 > > > Because org.apache.geode.cache.operations.OperationContext is deprecated, the > subclasses should also be. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Created] (GEODE-2181) I'm evaluating geode for our product. I started wi...
Gregory Chase created GEODE-2181: Summary: I'm evaluating geode for our product. I started wi... Key: GEODE-2181 URL: https://issues.apache.org/jira/browse/GEODE-2181 Project: Geode Issue Type: Improvement Reporter: Gregory Chase I'm evaluating geode for our product. I started with a simple test to check the behavior of geode on low-mem: - launch a locator and a single server with initial-heap==max-heap==4GB, critical-heap-percentage=90%, eviction-heap-percentage=80% - I see that gfsh passes in addition the following jvm params to server: -Xms4000m -Xmx4000m -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=60 - create a partitioned region - run client which connects to locator and get the region as PROXY (i.e. no local cache) - the client fills the server with elements until it gets ServerOperationException due to low memory (as expected) - now, I remove all elements from the region and try run the client again - the client fails on the first put operation (although server is empty!) My question is: why doesn't the GC collect removed items (I see on server status that currentHeapSize=3151MB, maximumHeapSize=3866)?. I'm expecting to be able to put data in the server if its empty... Thanks. *Reporter*: Assaf *E-mail*: [mailto:assaf_waiz...@amat.com] -- This message was sent by Atlassian JIRA (v6.3.4#6332)
