September 2008 - jbosscache-issues

[JBoss JIRA] Created: (JBCACHE-1414) more efficient exists() in JDBCCacheLoader and FileCacheLoader

by Manik Surtani (JIRA)

more efficient exists() in JDBCCacheLoader and FileCacheLoader -------------------------------------------------------------- Key: JBCACHE-1414 URL: https://jira.jboss.org/jira/browse/JBCACHE-1414 Project: JBoss Cache Issue Type: Feature Request Security Level: Public (Everyone can see) Components: Cache loaders Affects Versions: 2.2.0.GA Reporter: Manik Surtani Assignee: Manik Surtani Fix For: 3.0.0.CR1, 3.0.0.GA, 2.X exists() in both these cache loaders loads the node and then tests if the result is null. Hugely inefficient since it would involve byte stream deserialization. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 2 months

1
3
0 / 0

[JBoss JIRA] Commented: (JBCACHE-1331) Clustering between data centres

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1331?page=com.atlassian.jira.p... ] Manik Surtani commented on JBCACHE-1331: ---------------------------------------- Also see JBCACHE-816 > Clustering between data centres > ------------------------------- > > Key: JBCACHE-1331 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1331 > Project: JBoss Cache > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Components: Replication > Reporter: Manik Surtani > Assignee: Manik Surtani > Fix For: 3.2.0 > > > A mechanism for efficiently replicating state between data centres. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Commented: (JBCACHE-816) Create a CacheProxy to allow cache replication across WANs

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-816?page=com.atlassian.jira.pl... ] Manik Surtani commented on JBCACHE-816: --------------------------------------- Also see JBCACHE-1331 > Create a CacheProxy to allow cache replication across WANs > ---------------------------------------------------------- > > Key: JBCACHE-816 > URL: https://jira.jboss.org/jira/browse/JBCACHE-816 > Project: JBoss Cache > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Components: Clustering > Affects Versions: 1.4.0.GA, 2.0.0.GA > Reporter: Manik Surtani > Assignee: Manik Surtani > Fix For: 3.1.0 > > Original Estimate: 3 weeks > Remaining Estimate: 3 weeks > > This has to do with multiple JBoss Cache clusters deployed on LANs as islands, requiring a bridge between such islands. An ideal use case is where a cluster spans multiple data centres and would need to replicate across a WAN. > The ideal strategy would be to allow for async bridging, since most calls would be colocated to a single LAN, but may fail over to a remote LAN (pretty much the same idea of async repl within a LAN, except that now we treat an entire LAN as a host. Sort of a meta-cluster). > For more details and some brainstorming ideas, see: > http://wiki.jboss.org/wiki/Wiki.jsp?page=JBossCacheAcrossDataCentres -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Created: (JBCACHE-1411) Marshaller has no special treatment of arrays, even arrays of primitives or known types

by Manik Surtani (JIRA)

Marshaller has no special treatment of arrays, even arrays of primitives or known types --------------------------------------------------------------------------------------- Key: JBCACHE-1411 URL: https://jira.jboss.org/jira/browse/JBCACHE-1411 Project: JBoss Cache Issue Type: Bug Security Level: Public (Everyone can see) Components: Replication Affects Versions: 3.0.0.BETA1, 2.2.0.GA Reporter: Manik Surtani Assignee: Manik Surtani Fix For: 3.0.0.GA This is an inefficiency that causes, for example, the marshalling of a byte[] to fall back to Serializable. E.g., cache.put(fqn, key, new byte[]{1, 2, 3, 4}) would result in the byte[] being serialized as a Serializable. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
2
0 / 0

[JBoss JIRA] Created: (JBCACHE-1415) allow to put jboss cache lib in the WEB-INF/lib dir.The Cache#shutdown and Cache#stop methods are not blocking an provok errors when undeploying a webapp

by Laurent Mimoun (JIRA)

allow to put jboss cache lib in the WEB-INF/lib dir.The Cache#shutdown and Cache#stop methods are not blocking an provok errors when undeploying a webapp --------------------------------------------------------------------------------------------------------------------------------------------------------- Key: JBCACHE-1415 URL: https://jira.jboss.org/jira/browse/JBCACHE-1415 Project: JBoss Cache Issue Type: Feature Request Security Level: Public (Everyone can see) Components: Clustering Affects Versions: 2.2.0.GA Environment: tomcat 6 Reporter: Laurent Mimoun Assignee: Manik Surtani Priority: Critical -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
1
0 / 0

[JBoss JIRA] Resolved: (JBCACHE-1368) Optimize the CacheLoader.putAll() method so reads are not necessary

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1368?page=com.atlassian.jira.p... ] Manik Surtani resolved JBCACHE-1368. ------------------------------------ Fix Version/s: 3.0.0.CR1 Resolution: Done > Optimize the CacheLoader.putAll() method so reads are not necessary > ------------------------------------------------------------------- > > Key: JBCACHE-1368 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1368 > Project: JBoss Cache > Issue Type: Task > Security Level: Public(Everyone can see) > Reporter: Elias Ross > Assignee: Manik Surtani > Fix For: 3.0.0.CR1, 3.0.0.GA > > > Almost all CacheLoader implementations do a read of the existing data map when "putAll" is called but this may be unnecessary as the Cache likely has loaded the cache loader data at some point. > Manik suggests: "[We] need to analyse in the interceptors and see if that can be done away with. No real reason to do a read here. From the cache loader javadoc for the put(Fqn, Map) method: > 'Puts all entries of the map into the existing map of the given node, overwriting existing keys, but not clearing the existing map before insertion. ' > " > One improvement would be to check in the CacheLoader impl. if the Node was already loaded, then the Map merge can happen without the read. However, this logic might better belong in the cache loader interceptor. It might work okay to do this check in the base class for existing CacheLoader impl. as well. > Relates all JBCACHE-1221 - JDBC cache loader optimization. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Updated: (JBCACHE-1082) Optimise nodes for single elements

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1082?page=com.atlassian.jira.p... ] Manik Surtani updated JBCACHE-1082: ----------------------------------- Fix Version/s: 3.1.0 (was: 3.0.0.GA) > Optimise nodes for single elements > ---------------------------------- > > Key: JBCACHE-1082 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1082 > Project: JBoss Cache > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Affects Versions: 2.0.0.CR1 > Reporter: Manik Surtani > Assignee: Mircea Markus > Fix For: 3.1.0 > > Attachments: ReplicatedPojoPerfTest.java, UnversionedNode.patch.txt > > > Many use cases just place a single object in a Node. Rather than going through the memory and performance penalties of using Maps for a single entry, we should: > 1) Create a reference to 2 Objects for key and value, as well as a Map. > 2) Only initialise/construct the Map if Node.putAll() is called, or when Node.put() is called with a second key/value pair > 3) Otherwise just hold the key/value pair as direct object refs. > Would affect UnversionedNode and WorkspaceNodeImpl. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Updated: (JBCACHE-1009) Support preload functionality as part of region activation

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1009?page=com.atlassian.jira.p... ] Manik Surtani updated JBCACHE-1009: ----------------------------------- Fix Version/s: 3.1.0 (was: 3.0.0.GA) > Support preload functionality as part of region activation > ---------------------------------------------------------- > > Key: JBCACHE-1009 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1009 > Project: JBoss Cache > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Components: Cache loaders > Reporter: Brian Stansberry > Assignee: Manik Surtani > Fix For: 3.1.0 > > > Currently a cache loader preload only happens as part of cache start. It could be supported as part of the activateRegion process; i.e. if one of the defined preload regions in the cache loader config encapsulates the region being activated, preload the region. If one of the defined preload regions names a subregion of the region being activated, preload the subregion. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Updated: (JBCACHE-1371) Test for possible bug in TxInterceptor where a long running prepare executes simultaneously with a rollback on a different thread

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1371?page=com.atlassian.jira.p... ] Manik Surtani updated JBCACHE-1371: ----------------------------------- Fix Version/s: 3.1.0 (was: 3.0.0.GA) > Test for possible bug in TxInterceptor where a long running prepare executes simultaneously with a rollback on a different thread > --------------------------------------------------------------------------------------------------------------------------------- > > Key: JBCACHE-1371 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1371 > Project: JBoss Cache > Issue Type: Task > Security Level: Public(Everyone can see) > Components: Transactions > Reporter: Manik Surtani > Assignee: Manik Surtani > Fix For: 3.1.0 > > > The use case is when an async TM times out a transaction and rolls it back. The prepare may be in the process of acquiring locks, what happens to the rollback, and is the prepare thread properly interrupted? -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

[JBoss JIRA] Resolved: (JBCACHE-1372) Update LGPL headers on all src files.

by Manik Surtani (JIRA)

[ https://jira.jboss.org/jira/browse/JBCACHE-1372?page=com.atlassian.jira.p... ] Manik Surtani resolved JBCACHE-1372. ------------------------------------ Resolution: Done > Update LGPL headers on all src files. > ------------------------------------- > > Key: JBCACHE-1372 > URL: https://jira.jboss.org/jira/browse/JBCACHE-1372 > Project: JBoss Cache > Issue Type: Task > Security Level: Public(Everyone can see) > Components: Scripts and processes > Reporter: Manik Surtani > Assignee: Manik Surtani > Priority: Minor > Fix For: 3.0.0.GA > > > Should read as per the wiki: > http://wiki.jboss.org/wiki/JBossLGPLUsage -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

16 years, 3 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jbosscache-issues September 2008