[ANNOUNCEMENT] Apache Commons JCS 3.2 Released

[Thomas Vandahl]

The Apache Commons JCS team is pleased to announce the Apache Commons JCS 3.2 
release!

Apache Commons JCS is a distributed, versatile caching system.

Changes in this version include:

Fixed Bugs:
o Fix OSGi manifest  Issue: JCS-234. Thanks to Chris Lake. 
o Handle ServiceConfigurationError when using a different logging framework  
Issue: JCS-232. Thanks to Jeremy Long. 
o Fix Lateral TCP cache recovery loop for empty TCP servers  Issue: JCS-231. 
Thanks to Andrea Vettori. 
o UDPDiscoveryReceiver with EncryptingSerializer fails #88.  Issue: JCS-230. 
Thanks to Dorota Oeknigk-Urbanska. 
o LateralTCPListener class uses always StandardSerializer #89.  Issue: JCS-229. 
Thanks to Dorota Oeknigk-Urbanska. 

Changes:
o Bump actions/cache from 2.1.7 to 3.0.9 #86, #111. Thanks to Dependabot, Gary 
Gregory. 
o Bump actions/checkout from 2 to 3.0.2 #93. Thanks to Dependabot, Gary 
Gregory. 
o Bump actions/setup-java from 2 to 3.5.1. Thanks to Gary Gregory. 
o Bump commons-parent from 52 to 58 #90, #130, #154. Thanks to Dependabot. 
o Bump log4j-api from 2.17.1 to 2.20.0 #85, #100, #107, #145. Thanks to 
Dependabot. 
o Bump log4j-core from 2.17.1 to 2.20.0 #87, #99, #108, #144. Thanks to 
Dependabot. 
o Bump commons-text from 1.9 to 1.10.0 #110. Thanks to Dependabot. 
o Bump commons-io from 2.11.0 to 2.12.0. Thanks to Gary Gregory. 

For complete information on Apache Commons JCS, including instructions on how 
to submit bug reports, patches, or suggestions for improvement, see the Apache 
Commons JCS website:

https://commons.apache.org/proper/commons-jcs/

Download it from
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

Thomas Vandahl 
Apache Commons Team

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 3.1

[Thomas Vandahl]

The Apache Commons project announces the release of Apache Commons JCS 3.1.

Apache Commons JCS is a distributed caching system written in Java. It is 
intended to speed up 
applications by providing a means to manage cached data of various dynamic 
natures.

JCS 3.0 and onwards require at least Java 8.

JCS 3.1 uses log4j as an optional component only. Nevertheless, the dependency 
has been updated to address the Log4Shell issues.

This is a feature and bug fix release.

For a list of changes in this release see:
https://commons.apache.org/proper/commons-jcs/changes-report.html#a3.1

For complete information on Apache Commons JCS, including instructions on
how to submit bug reports, patches, or suggestions for improvement, see the
Apache Apache Commons JCS website:
https://commons.apache.org/proper/commons-jcs/

Download page:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

Have fun!


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [jcs] Invalid Class Exception help needed

[Thomas Vandahl]

Hi Stuart,

On 22.01.21 08:26, Stuart Maclean wrote:
I have a cache region whose key type is say String and whose value type 
is some class C.  C is Serializable.


I use the cache, so objects of class C go into that region. The cache is 
backed by disk, so objects of class C end up in files.


C may then change, such that its serialVersionUID changes. If so, what I 
want to happen is that when I do a


I guess what you are looking for happens in e.g. 
org.apache.commons.jcs.auxiliary.disk.indexed.IndexedDiskCache.readElement(K)


In this method, the ClassNotFoundException from deserialization is 
wrapped into an IOException. This exception is then caught by 
org.apache.commons.jcs.auxiliary.disk.indexed.IndexedDiskCache.processGet(K) 



In the catch clause, reset() is called, which clears the cache. So 
actually the behavior you are asking for actually should be there.


If you need more detailed control, you may provide your own 
org.apache.commons.jcs.engine.behavior.IElementSerializer and use the 
object validation features described in 
java.io.ObjectInputStream.registerValidation(ObjectInputValidation, int)


You will have to create the disk cache programmatically then, as the 
IElementSerializer cannot be set via configuration.


HTH
Bye, Thomas

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [jcs] Maven - Commons JCS 3

[Thomas Vandahl]

On 30.10.20 21:57, tolai...@ups.com.INVALID wrote:

Is the Commons JCS 3 library going to be published to Maven?


What is wrong with 
https://repo1.maven.org/maven2/org/apache/commons/commons-jcs3-core/3.0/commons-jcs3-core-3.0.jar 
?


Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 3.0 Released

[Thomas Vandahl]

The Apache Commons Team is pleased to announce the availability of
Apache Commons JCS 3.0

Apache Commons JCS is a distributed, versatile caching system.

This has been a major overhaul of JCS with many adjustments for JDK 8+,
better concurrency and logging.

Commons JCS 3.0 requires Java 8 or later.

The release notes can be reviewed at:
http://www.apache.org/dist/commons/jcs/RELEASE-NOTES.txt

Distribution packages can be downloaded from:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

When downloading, please verify signatures using the KEYS file available
at: http://www.apache.org/dist/commons

More information and comprehensive documentation is available at:
https://commons.apache.org/proper/commons-jcs/

Maven artifacts are also available in the central Maven repository:
http://repo1.maven.org/maven2/org/apache/commons/jcs3

The Apache Commons Team


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [csv] Slicing through CSVRecords

[Thomas Vandahl]

Hi Siegfried,

Happy New Year btw.!

On 31.12.18 15:34, Siegfried Goeschl wrote:
> somehow I’m working a lot with ad-hoc scripting and commons-csv  and it would 
> be helpful to have additional functionality slicing through CSV records, e.g.
> 
> * group CSVRecord by the value of column 
> * get a list of distinct values from a column
> 
> No rocket science but to avoid typing the same code over and over 
> 
> * has anyone implemented that already and I’m just unaware of it
> * if not - could this be a valuable contribution?

I prefer SuperCSV (http://super-csv.github.io/super-csv/) for tasks like
this. It supports bean mapping and cell-processors which makes record
handling super easy.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 2.2 Released

[Thomas Vandahl]

The Apache Commons Team is pleased to announce the availability of:

Apache Commons JCS 2.2

Apache Commons JCS is a distributed, versatile caching system.

The release notes can be reviewed at:
http://www.apache.org/dist/commons/jcs/RELEASE-NOTES.txt

Distribution packages can be downloaded from:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

When downloading, please verify signatures using the KEYS file available
at:
http://www.apache.org/dist/commons

More information and comprehensive documentation is available at:
https://commons.apache.org/proper/commons-jcs/

Maven artifacts are also available in the central Maven repository:
http://repo1.maven.org/maven2/org/apache/commons/

The Apache Commons Team


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[jcs] Re: Can JCS have update events?

[Thomas Vandahl]

On 08.04.17 14:40, robertlazarski . wrote:
> We need update events because our code already uses them - we are
> replacing our current cache libs (ehcache) with another lib -
> preferably JCS.
> 
> More specifically, our local cache is filled with stock prices via a
> "HTTP server push" stream and on updates we need to perform some
> additional calculations using events and listeners.

Although I strongly believe that this should be handled outside the
scope of the cache, I think that update events could be added to JCS
quite easily.

JCS does not support this at the moment. The best place to add such a
feature probably would be CompositeCache.update() and remove(),
respectively. Please be sure to take the asynchronous nature of event
handling into account.

> I looked a lot at the JCS code and didn't find update events, which is
> why I looked at the JCS JCache implementation. Apparently JSR-107
> didn't define persistence though. I notice some other projects have a
> JSR-107 "CacheStore" for disk persistence but to be clear I definitely
> want to use JCS.

The JCache implementation actually is just a wrapper around JCS, so all
features of JCS, including the disk auxiliary should be available from
JCache, too. I'm only maintaining the core of JCS, so I cannot speak
from JCache experiences here.

> Saying this as humbly as possible as clearly I am not a Cache expert ,
> I am an Apache Axis2 committer plus I have done several Linux related
> contributions and I would be willing to submit patches if I end up
> having to "scratch my own itch" . Of course though I would rather use
> an API that already exists.

Patches are always welcome.
However I still suggest to reconsider putting data manipulations into
the scope of the cache, independently of the library used.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: Can JCS have update events?

[Thomas Vandahl]

Please use [jcs] as a prefix to the subject when posting to
us...@commons.apache.org.

On 06.04.17 18:21, robertlazarski . wrote:
> I am coding a simple local "Indexed Disk Auxiliary Cache" and
> IsEternal=true using JCS, and I can't figure out how to have an
> onUpdate() event. Is there update events in JCS?

Why are you coding the cache? This should only need configuration. Why
do you think you need update events?

> I looked at the JCS implementation of JCache events and it looks like
> what I need, but I can't figure out how to do disk persistence with
> JCS / JCache as I can't find any code examples.

Did you read the docs at
https://commons.apache.org/proper/commons-jcs/IndexedDiskAuxCache.html?
The configuration example on that page is basically all you need. See
https://commons.apache.org/proper/commons-jcs/JCSPlugins.html for a
description of JCS' concept of auxiliaries.

Bye, Thomas

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [jcs] Migrating from Ehcache to JCS, need help getting the right config

[Thomas Vandahl]

On 23.03.17 15:01, robertlazarski . wrote:
> Thanks for the response! That helped a lot. I am about to start coding
> but I am trying to figure out #3. Please see below.

The cache warming itself is not a big deal. Just call getKeySet() on the
memory cache and call get() for all keys. This will load the values from
disk into the memory cache. The question is just where to insert the
loop into the application initialization.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [jcs] Migrating from Ehcache to JCS, need help getting the right config

[Thomas Vandahl]

Hi there,

On 21.03.17 15:50, robertlazarski . wrote:
> Hello! We have a production system running Ehcache and we'd like to
> use JCS instead. What I need to do with JCS is:
> 
> 1) Load all Objects into memory, unlimited with infinite "time to
> live". In practice over many years, its not been a problem for us.
> Currently using LRU.

Yes, this is easy, just define a memory cache big enough to hold all
your objects. Set

jcs.default.elementattributes.IsEternal=true

> 2) Store Objects from memory to disk on JVM shutdown, or at least
> periodically flush to disk - while keeping the entire cache in memory.

See
https://commons.apache.org/proper/commons-jcs/IndexedDiskAuxCache.html
for the configuration of an auxiliary disk cache. If you have configured
your memory cache big enough, no "overflow" of items to disk will happen
in normal operation.

You need to shut down the cache orderly
(CompositeCacheManager.getInstance().shutDown()) to trigger the "dump to
disk" process. There is no simple method to do this periodically.

> 
> 3) Most importantly, on server startup we need the entire disk
> persisted cache loaded into memory, i.e. "cache warming" . I am
> wondering if the CacheLoaderAdapter is the right class for this?
> Currently we use a class that Ehcache calls on startup that just uses
> an Iterator over all the cache Elements, effectively loading
> everything into memory.

This can be achieved by setting ClearDiskOnStartup=false. This happens
to be the default. JCS will load all the keys into memory then but not
the values. So if you want the values to be loaded, just modify the
class you used for EHCache.

How to integrate this into your application depends on your application
lifecycle. There is no CacheLoaderAdapter class in JCS. If you tell me a
bit more about your application, I might be able to give you a hint.

HTH
Bye, Thomas

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 2.1 Released

[Thomas Vandahl]

The Apache Commons Team is pleased to announce the availability of:

Apache Commons JCS 2.1

Apache Commons JCS is a distributed, versatile caching system.

The release notes can be reviewed at:
http://www.apache.org/dist/commons/jcs/RELEASE-NOTES.txt

Distribution packages can be downloaded from:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

When downloading, please verify signatures using the KEYS file available
at:
http://www.apache.org/dist/commons

More information and comprehensive documentation is available at:
https://commons.apache.org/proper/commons-jcs/

Maven artifacts are also available in the central Maven repository:
http://repo1.maven.org/maven2/org/apache/commons/

The Apache Commons Team


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS : element expires sooner than configured

[Thomas Vandahl]

Ouch, that hurt. I'm sorry, this is a problem with ElementAttributes.clone() 
not updating createTime. I fixed it in trunk. Could you try if it works for you?

Bye, Thomas 

-- tv

> Am 23.01.2017 um 10:48 schrieb Sahib Bin Mahboob Ron :
> 
> Hi,
> 
> I am trying to configure a JCS using the following configuration:
> 
> # DEFAULT CACHE REGION
> 
> jcs.default=
> jcs.default.cacheattributes=
>org.apache.commons.jcs.engine.CompositeCacheAttributes
> jcs.default.cacheattributes.MaxObjects=1000
> jcs.default.cacheattributes.MemoryCacheName=
>org.apache.commons.jcs.engine.memory.lru.LRUMemoryCache
> 
> jcs.default.elementattributes=org.apache.commons.jcs.engine.ElementAttributes
> jcs.default.elementattributes.IsEternal=false
> jcs.default.elementattributes.MaxLife=300
> jcs.default.elementattributes.IsSpool=false
> 
> 
> While it is expected to expire the item in 300 seconds as per MaxLife
> configuration, it does not work properly in following steps:
> 
> 1. Putting a key "aKey" for the first time in cache: It works perfectly.
> 2. "aKey" expires after 300 seconds.
> 3. Put "aKey" again on the cache.
> 4. It expires immediately.
> 
> Am I missing any configuration or this is the way it should behave?
> 
> - Sahib


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 2.0 Released

[Thomas Vandahl]

The Apache Commons Team is pleased to announce the availability of:

Apache Commons JCS 2.0

Apache Commons JCS is a distributed, versatile caching system.

The release notes can be reviewed at:
http://www.apache.org/dist/commons/jcs/RELEASE-NOTES.txt

Distribution packages can be downloaded from:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

When downloading, please verify signatures using the KEYS file available
at:
http://www.apache.org/dist/commons

More information and comprehensive documentation is available at:
https://commons.apache.org/proper/commons-jcs/

Maven artifacts are also available in the central Maven repository:
http://repo1.maven.org/maven2/org/apache/commons/

The Apache Commons Team


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS 2.0-beta-2: ShrinkerThread: Unexpected trouble in shrink cycle

[Thomas Vandahl]

On 26.11.16 16:07, Kirys wrote:
> [C]2016-11-26 15:56:00,070|JCS-Scheduler-Thread-21|INFO
> o.a.c.j.e.m.s.ShrinkerThread: Unexpected trouble in shrink cycle
> java.lang.NoClassDefFoundError:
> org/apache/commons/jcs/engine/control/group/GroupAttrName

This looks like a class-loader issue between threads. The class
mentioned is included in the commons-jcs-core jar and the application
should have no problems loading it. I'm not familiar with the TomEE
environment. Can you try to reproduce the problem outside this context?

Bye, Thomas



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [ANNOUNCE] Apache Commons JCS 2.0-beta-2 Released

[Thomas Vandahl]

On 16.11.16 14:56, Tim Cronin wrote:
> when will it be accessible via maven/ivy repo?

It already is:
http://repo1.maven.org/maven2/org/apache/commons/commons-jcs-core/2.0-beta-2/

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[ANNOUNCE] Apache Commons JCS 2.0-beta-2 Released

[Thomas Vandahl]

The Apache Commons Team is pleased to announce the availability of:

Apache Commons JCS 2.0-beta-2

Apache Commons JCS is a distributed, versatile caching system.

The release notes can be reviewed at:
http://www.apache.org/dist/commons/jcs/RELEASE-NOTES.txt

Distribution packages can be downloaded from:
https://commons.apache.org/proper/commons-jcs/download_jcs.cgi

When downloading, please verify signatures using the KEYS file available at:
http://www.apache.org/dist/commons

Maven artifacts are also available in the central Maven repository:

http://repo1.maven.org/maven2/org/apache/commons/

The Apache Commons Team


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: Apache JCS 2.0 beta 2?

[Thomas Vandahl]

Hi Ryan,

On 23.09.16 22:54, Ryan Fong wrote:
> G'day, all. I've submitted a couple of patches towards Apache JCS 2.0 beta-2 
> a few months back. I've also been seeing some errors like JCS-146 which have 
> already been fixed but are not available in a download.  However, I don't see 
> any activity or intent to build a new beta-2 or a release candidate. Is this 
> project still active? What are the release expectations?
> 

your contributions are very much appreciated. I'm currently working on
JCS-130 to clean up the inter-dependencies of managers, factories,
services and stuff. Before releasing 2.0, I wanted to get this finished.
Unfortunately, I have very little time for open source development at
the moment so this takes longer than I expected.

In the meantime you can use a fairly recent build (currently 2016-08-07)
from the snapshot-repository at
https://repository.apache.org/content/groups/snapshots/org/apache/commons/commons-jcs-core/2.0-SNAPSHOT/

HTH
Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS]

[Thomas Vandahl]

Hi Mat,

On 03.02.16 17:44, Mat Jaggard wrote:
> I have three patches that might be helpful...
> 
> 1. Javadoc changes to make it compile on my machine
> 2. Server port changes which may (or may not) fix the continuum unit
> test issue
> 3. Change to the way existing tables are detected (I think I got a
> different SQL error code)

Thanks again for contributing. I applied patches 1 and 2. I'm not so
sure about 3.
The check for SQL state 23000 is quite common. We use the same check in
DB Torque for several different database systems with good success.
What error message did you get?

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS]

[Thomas Vandahl]

On 02.02.16 11:22, Mat Jaggard wrote:
> I've submitted my pull request.
> 
> https://github.com/apache/commons-jcs/pull/2
> 
> On Mon, 1 Feb 2016 at 16:32 Gary Gregory  wrote:
> 
>> It is, github is just a mirror. You submit a patch with the Jira if you
>> prefer.

Thanks for your contribution. How do I merge this into SVN now? I'm not
familiar with github.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [configuration] How to read an INI Configuration with brackets as list separator

[Thomas Vandahl]

On 25.08.15 07:55, Patrick Brunmayr wrote:
 Hello
 
 How can i read a configuration like that ?
 
 [RESPONSE]
 code = 200
 description = Command completed successfully
 queuetime = 0
 runtime = 0.006
 property[count][0] = 164
 property[domain][0] = xxx
 property[domain][1] = xxx
 property[domain][2] = xxx
 

Hi Patrick,

list separators are for values, not keys. See the JavaDoc for
HierarchicalINIConfiguration for a way to handle files like this. You
may want to check out the section
http://commons.apache.org/proper/commons-configuration/userguide_v1.10/howto_xml.html#Accessing_structured_properties
of the users guide for possibilities to access your data. Using the
XPathExpressionEngine might cover your case but that's just a wild guess.

HTH
Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Region No longer alive so returning false for key

[Thomas Vandahl]

On 06.07.15 15:37, Tim Cronin wrote:
 is there any documentation on migrating from 1.3 to 2?

Not yet. I started to write something, but I have yet to finish it.

 
 the JCSAdmin.jsp the FAQ link is broken. moved due to package change
 http://svn.apache.org/viewvc/commons/proper/jcs/trunk/commons-jcs-core/src/main/java/org/apache/commons/jcs/admin/JCSAdmin.jsp?view=log

Thank you for the hint I will fix this.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Region No longer alive so returning false for key

[Thomas Vandahl]

On 06.07.15 17:06, Tim Cronin wrote:
 I added the following before calling the composite cache shutdown.
 this should be called via the composite cache shutdown code i think
 
 ThreadPoolManager poolMgr = ThreadPoolManager.getInstance();
 
 for(String poolName : poolMgr.getPoolNames()) {
 try {
 poolMgr.getPool(poolName).shutdown();
 } catch (Throwable t) {
 LOGGER.warn(Failed to close pool  + poolName, t);
 }
 }

Thank you for your contributions I'll try to take care of these.
Please not that these problems also have been present in 1.3.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Region No longer alive so returning false for key

[Thomas Vandahl]

On 06.07.15 16:33, Tim Cronin wrote:
 MaxLifeSeconds has been changed to MaxLife, didn't find anything in the
 docs about this change.

See http://commons.apache.org/proper/commons-jcs/RegionProperties.html

Bye, Thomas



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Region No longer alive so returning false for key

[Thomas Vandahl]

On 02.07.15 15:13, Tim Cronin wrote:
 JCS 1.3
 
 i periodically get the above error.
 
 what do i need to do to recover from this?
 

First of all I'd suggest to try out commons-jcs 2.0-beta1. It contains
numerous fixes and cleans up the lifecycle of most of the JCS manager
classes. If that doesn't help, come back and ask again as this would
require deeper digging.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Error starting web service with JCS on WAS 8.5.5 (Linux only)

[Thomas Vandahl]

Hi Barry,

On 01.06.15 02:09, Barry Lulas wrote:
 Hello,
 
 I have created a web service for WebSphere that uses a basic JCS 
 configuration.  When I package and deploy for WAS under Windows on my local 
 machine I have no issues.  When I deploy that same .ear file to my WAS 
 running on Linux I get the following exception when I try to start my service 
 in the WAS console:
 
 [6/1/15 2:00:51:875 CEST] 0093 BeansDeployer E BeansDeployer deploy 
 java.lang.NullPointerException at 
 java.util.Properties.load(Properties.java:356) at 
 org.apache.commons.jcs.jcache.JCSCachingManager.readConfig(JCSCachingManager.java:95)
  at 
 org.apache.commons.jcs.jcache.JCSCachingManager.init(JCSCachingManager.java:70)
  at 
 org.apache.commons.jcs.jcache.JCSCachingProvider.getCacheManager(JCSCachingProvider.java:56)
  at 
 org.apache.commons.jcs.jcache.extras.cdi.ExtraJCacheExtension.addJCacheBeans(ExtraJCacheExtension.java:81)
  at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at 
 sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:94) 
 at 
 sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:55)
  at java.lang.reflect.Method.invoke(Method.java:619) at 
 org.apache.webbeans.event.ObserverMethodImpl.notify(ObserverMethodImpl.java:285)
  at 
 org.apache.webbeans.event.NotificationManager.fireEvent(NotificationManager.java:455)
  at or
 g
.apache.webbeans.container.BeanManagerImpl.fireEvent(BeanManagerImpl.java:452) 
at 
org.apache.webbeans.config.BeansDeployer.fireAfterBeanDiscoveryEvent(BeansDeployer.java:269)
 at org.apache.webbeans.config.BeansDeployer.deploy(BeansDeployer.java:180) at 
org.apache.webbeans.lifecycle.AbstractLifeCycle.startApplication(AbstractLifeCycle.java:124)
 at 
org.apache.webbeans.web.lifecycle.WebContainerLifecycle.startApplication(WebContainerLifecycle.java:78)
 at 
com.ibm.ws.webbeans.common.CommonLifeCycle.startApplication(CommonLifeCycle.java:106)
 at 
com.ibm.ws.webbeans.services.JCDIServletContainerInitializer.onStartup(JCDIServletContainerInitializer.java:85)
 at 
com.ibm.ws.webcontainer.webapp.WebAppImpl.initializeServletContainerInitializers(WebAppImpl.java:613)
 at com.ibm.ws.webcontainer.webapp.WebAppImpl.initialize(WebAppImpl.java:409) 
at 
com.ibm.ws.webcontainer.webapp.WebGroupImpl.addWebApplication(WebGroupImpl.java:88)
 at com.ibm.ws.webcontainer.VirtualHostImpl.addWebApplication(VirtualHost
 I
mpl.java:169) at 
com.ibm.ws.webcontainer.WSWebContainer.addWebApp(WSWebContainer.java:749) at 
com.ibm.ws.webcontainer.WSWebContainer.addWebApplication(WSWebContainer.java:634)
 at 
com.ibm.ws.webcontainer.component.WebContainerImpl.install(WebContainerImpl.java:426)
 at 
com.ibm.ws.webcontainer.component.WebContainerImpl.start(WebContainerImpl.java:718)
 at 
com.ibm.ws.runtime.component.ApplicationMgrImpl.start(ApplicationMgrImpl.java:1177)
 at 
com.ibm.ws.runtime.component.DeployedApplicationImpl.fireDeployedObjectStart(DeployedApplicationImpl.java:1370)
 at 
com.ibm.ws.runtime.component.DeployedModuleImpl.start(DeployedModuleImpl.java:639)
 at 
com.ibm.ws.runtime.component.DeployedApplicationImpl.start(DeployedApplicationImpl.java:968)
 at 
com.ibm.ws.runtime.component.ApplicationMgrImpl.startApplication(ApplicationMgrImpl.java:776)
 at 
com.ibm.ws.runtime.component.ApplicationMgrImpl.startApplicationDynamically(ApplicationMgrImpl.java:1379)
 at com.ibm.ws.runtime.component.ApplicationMgrImpl.start
 (
ApplicationMgrImpl.java:2189) at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl.start(CompositionUnitMgrImpl.java:435)
 at 
com.ibm.ws.runtime.component.CompositionUnitImpl.start(CompositionUnitImpl.java:123)
 at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl.start(CompositionUnitMgrImpl.java:378)
 at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl.access$500(CompositionUnitMgrImpl.java:126)
 at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl$1.run(CompositionUnitMgrImpl.java:653)
 at 
com.ibm.ws.security.auth.ContextManagerImpl.runAs(ContextManagerImpl.java:5477) 
at 
com.ibm.ws.security.auth.ContextManagerImpl.runAsSystem(ContextManagerImpl.java:5603)
 at 
com.ibm.ws.security.core.SecurityContext.runAsSystem(SecurityContext.java:255) 
at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl.startCompositionUnit(CompositionUnitMgrImpl.java:667)
 at 
com.ibm.ws.runtime.component.CompositionUnitMgrImpl.startCompositionUnit(CompositionUnitMgrImpl.java:611)
 at com.ibm.ws.runtime.compo
 n
ent.ApplicationMgrImpl.startApplication(ApplicationMgrImpl.java:1269) at 
sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at 
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:94) 
at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:55)
 at java.lang.reflect.Method.invoke(Method.java:619) at 
sun.reflect.misc.Trampoline.invoke(MethodUtil.java:87) at 
sun.reflect.GeneratedMethodAccessor73.invoke(Unknown Source) at 

Re: can # sometimes be ignored by JCS

[Thomas Vandahl]

On 10.03.15 23:12, arnout cator wrote:
 Hi, I have the situation that I have 10+ web apps using JCS. I have this in
 cache.ccf:
 
  ###
 # Stylesheet settings
 ###
 #disable disk caching if WAS doesn't provide serialized dom
 #jcs.region.stylesheet=
 jcs.region.stylesheet=DC
 jcs.region.stylesheet.cacheattributes.MaxObjects=128
 ###
 
 I have no issues with all other regions defined, but I have a problem that
 sometimes amongst the 10 web apps, not consistently the same web apps, the
 stylesheet cache is not invalidated when I do a code release to the 10
 apps. I get over it by removing all disk cache files and restarting the app
 server.
 
 I wondered if it is possible that JCS sometimes reads the # line. Can that
 be causing this?

This is very unlikely because JCS uses the standard JDK methods to read
its configuration files.

 If not, why would just the stylesheet disk cache file not be invalidated
 sometimes? Did anyone in JCS user world see a similar event?

I'm not sure I understand your environment correctly. Do you generally
expect the cache contents to survive an application restart or not?

The disk cache will persist keys and data during an orderly shutdown so
that they can be reloaded on restart. This normally requires
JCSServletContextListener to be present in the web application. However
JCS also registers a shutdown hook to the JVM which may or may not work
in an app-server environment. This would explain the inconsistent
behavior you are experiencing.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JSC 1.3 thread cleanup

[Thomas Vandahl]

On 23.01.15 15:16, Tim Cronin wrote:
 ok, thanks
 
 how different is 2 vs. 1.3?

First of all it is unreleased as of yet. The main change is the use of
the class CacheAccess instead of JCS when using caches. Due to the move
to Apache Commons, the package has changed from org.apache.jcs to
org.apache.commons.jcs. This will affect existing configuration files.
Other than that, your code should continue to work.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JSC 1.3 thread cleanup

[Thomas Vandahl]

On 22.01.15 17:42, Tim Cronin wrote:
 I'm getting tomcat errors on shutdown:
 
 2015-01-22 16:41:51,186 [localhost-startStop-2] ERROR
 org.apache.catalina.loader.WebappClassLoader- The web application
 [/api] appears to have started a thread named
 [ElementEventQueue.QProcessor-1] but has failed to stop it. This is very l
 ikely to create a memory leak.
 
 i tracked it down to
 
 org.apache.jcs.engine.control.CompositeCache:
 
 /**
  * EventQueue for handling element events. 1 should be enough for all
 the regions. Else should
  * create as needed per region.
  */
 public static IElementEventQueue elementEventQ = new ElementEventQueue(
 AllRegionQueue );
 
 then calling CompositeCache.elementEventQ.destroy();
 
 but it can be flaky and IMO should be handled via
 CompositeCacheManager.shutdown().
 
 is there a better way to handle this and do i need to file a bug on this?
 

This has already been fixed in the current trunk. You may check out the
code and see if it works.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Dynamic Cache Expiration on get

[Thomas Vandahl]

Hi Jan,

On 16.01.15 17:12, Jan Swaelens wrote:
 Hello,
 
 I have a region set up to expire its data lets say after 60 seconds.
 There's also a shrink thread running every 100 seconds.
 
 We add an event handler when storing certain items, in this handler we
 implement some logic to re-insert the expired item on the same key.
 
 The problem is, when an item expires 'on get' the expiry handler fires
 async and we can't get the updated value in the requesting thread.
 
 Is it possible to act before the item is actually removed to take the
 keep/toss away call instead of re-inserting it? Any other suggestions for
 solving this pattern?

I understand you want to update your items when they expire, right? This
is normally done in the requesting thread, when the cache returns null
for the requested key. In my experience, this absolutely makes sense
because you will have connections, access rights and stuff available
there. This is usually not the case in a background thread.

That said, I'd like to add that you may add your own implementation of a
memory cache with the desired expiry behavior. See the way the key lists
are handled in LRUMemoryCache and MRUMemoryCache.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] disk persistence - same conf file, different results

[Thomas Vandahl]

On 17.11.14 14:49, Tim Cronin wrote:
 Any ETA on when JCS 2.0 will be released?

There is a vote running currently for 2.0 alpha 1.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] disk persistence - same conf file, different results

[Thomas Vandahl]

On 14.11.14 13:47, joao tiago a. m. viegas wrote:
 I'll probably reformat this question in a simpler way.
 
 The goal is to persist an element and retrieve it after restart of a new vm.

The disk caches persist the keys when the cache is shut down. So you
should call CompositeCacheManager.getInstance().shutDown(); before your
application ends.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] disk persistence - same conf file, different results

[Thomas Vandahl]

On 14.11.14 18:03, Thomas Vandahl wrote:
 On 14.11.14 13:47, joao tiago a. m. viegas wrote:
 I'll probably reformat this question in a simpler way.

 The goal is to persist an element and retrieve it after restart of a new vm.
 
 The disk caches persist the keys when the cache is shut down. So you
 should call CompositeCacheManager.getInstance().shutDown(); before your
 application ends.
 

To be precise: You cannot shut down the cache in the middle of a process
because with jcs 1.3 the cache will not re-initialize correctly. This
should have been fixed to a large extent in jcs 2.0 but still needs
further work.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS Cache behaviour question

[Thomas Vandahl]

On 24.10.14 00:48, arnout cator wrote:
 Can someone tell me how JCS works to show Auxiliary hits and no data
 file size nor key size, considering my setup? 

JCS keeps the disk cache files open during operation. So depending on
your operating system, you may not see updates of the directory entries
of these files. If you shut down JCS properly, the files will be closed
and the size should reflect the actual content.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Remote Cache causes Tomcat Windows Service to hang on shutdown.

[Thomas Vandahl]

On 06.10.14 15:55, Art Thursland wrote:
 Here is the part of the log written during shutdown:
 
 09:42:18,899 INFO  [RemoteCacheStartupServlet] Shutting down remote cache
 09:42:18,903 INFO  [CompositeCache] In DISPOSE, [tokenCache] fromRemote 
 [false]
 Region Name = tokenCache
 HitCountRam = 0
 HitCountAux = 0
 ---LRU Memory Cache
 List Size = 0
 Map Size = 0
 Put Count = 0
 Hit Count = 0
 Miss Count = 0
 09:42:18,903 INFO  [CompositeCache] In DISPOSE, [tokenCache] disposing of 
 memory cache.
 09:42:18,903 INFO  [AbstractMemoryCache] Memory Cache dispose called.  
 Shutting down shrinker thread if it is running.
 

Could you please switch the log level to DEBUG and post the log section?

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: correct estimate of the disk cache size

[Thomas Vandahl]

On 09.10.14 17:43, arnout cator wrote:
 jcs.auxiliary.DC.attributes.OptimizeAtRemoveCount=250

That's the key here.

In general, you would expect the cache size that you calculated.
However, considering the way the disk cache works, this is only true if
your data objects are of exactly the same size (which is very
improbable). So if a new data object is somewhat bigger than the
previously expired one, the cache will append to the cache file and the
file will grow.

If you want to have better control over your disk usage, I'd recommend
to test the BlockDiskCache. It may fit your use case better.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS cache benchmarking for performance

[Thomas Vandahl]

On 09.10.14 16:02, arnout cator wrote:
 What would be the best practice in monitoring JCS caching performance for
 production web applications? I am looking for a way to collect data and
 then give advice on sizing of web applications that heavily use JCS for
 caching of java objects.

Which version do you use?

If 1.3, then your best bet would be JCSAdminBean or JCSAdmin.jsp,
respectively. It can tell you about the filling level as well as hits,
fails, auxiliary hits and fails and much more. It will not give you
timing information, however.

In 2.0, you may use JMX to access this information.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [math] Curve fitting ...

[Thomas Vandahl]

I take it, you ask for help for doing your homework?

Bye, Thomas 

 Am 14.08.2014 um 15:56 schrieb South Light southligh...@gmail.com:
 
 Hi Ted,
 
 Thanks a lot for your suggestion but I need to add it using java.
 
 Thanks
 ​again​
 .​
 
 
 
 2014-08-14 2:22 GMT-03:00 Ted Dunning ted.dunn...@gmail.com:
 
 Have you considered using an interactive system like R, Matlab or Octave?
 
 You might be happier.
 
 Or even have you considered goal search in Excel?
 
 
 
 
 On Wed, Aug 13, 2014 at 6:08 PM, South Light southligh...@gmail.com
 wrote:
 
 Hi,
 
 May be someone can help me with this problem.
 
 Given the follow function: y = 10 ^ ((x + 82) / (-10 * A))
 
 I would like to found the A value witch curve fit better for a set of x,y
 values, usually the set is about 20 to 25 x,y values.
 
 I use the CurveFitter class and the ParametricUnivariateFunction
 
 
 ParametricUnivariateFunction function = new
 ParametricUnivariateFunction()
 {
 
 ​  ​
 @Override
 ​  ​
 public double[] gradient(double x, double[] params) {
 ​
 (? comment)
 ​  ​
 }
 
 ​  ​
 @Override
 ​  ​
 public double value(double x, double[] params) {
 
 ​​
 double a = params[0];
 
 ​​
 return Math.pow(10, ((x + 82) /
 ​(​
 -10 *
 ​a
 ​)​
 ));
 
 ​  ​
 }
 
 };
 
 LevenbergMarquardtOptimizer optimizer = new
 LevenbergMarquardtOptimizer();
 
 CurveFitterParametricUnivariateFunction fitter = new
 CurveFitterParametricUnivariateFunction(optimizer);
 
 double[] x = {
 -82
 ,
 ​-85
 ,
 ​-89
 };
 
 double[] y = {
 ​1
 ,
 ​1.4
 ,
 ​2
 };
 
 for (int i = 0; i  x.length; i++)
 ​  ​
 fitter.addObservedPoint(x[i], y[i]);
 
 double[] result = fitter.fit(function, new double[] { 1, 10 });
 
 
 
 ​A. ​Is this the best way to solve the problem or there's another better
 way?
 
 B. What do we need to write on the gradient area (? comment) ?​
 
 Any help will be more then welcome.
 
 Many thanks !!
 

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] event handler callback not getting called

[Thomas Vandahl]

On 19.03.14 18:25, sebb wrote:
 But when an element is evicted from JCS cache im not getting any callback
 though i know cache eviction has happened for element for sure.

 Is there anything else that i need to do apart from what is mentioned in
 the example.

Would you please open an issue in JIRA with this sample code and the
related configuration attached?

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Version 2 release

[Thomas Vandahl]

On 04.02.14 16:52, Gary Gregory wrote:
 I think the biggest hurdle to releasing [JCS] 2.0 is finding a [Commons]
 committer willing to put in the time to go over the component and act as
 the release manager.

Well, I'm actually doing what I can. My hesitation partly comes from the
fact that the test results within the Continuum build are not in line
with mine. Any help will be appreciated as this is going to be my first
release within Apache Commons.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Version 2 release

[Thomas Vandahl]

Gary,

On 08.02.14 17:34, Gary Gregory wrote:
 Mat,
 
 I would suggest getting familiar with the code base in trunk. Can the code
 be released as is or are there problems? Should some issues in Jira be
 fixed for this release?
 
 Major releases like a 2.0 from 1.x are the only time APIs can break binary
 compatibility which means that the package name must change and the Maven
 coords as well.
 
 Does the current code even warrant the 2.0 label or would a 1.x dump be a
 better representation of what is in trunk?

Rest assured that all these things have been taken care of. The API
*has* changed and the modifications *are* substantial. A look into the
commits of the last two years should show that.

Mat,

if you want to help, please check that the build and all tests are
working fine for you. Note that some tests are disabled in the POM as
they seem to be timing sensitive. I did my best to fix these problems,
however. The documentation does not yet reflect the recent API changes.
I'll fix this as soon as I can.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [jcs] Re: JCS disk only and expiration config question

[Thomas Vandahl]

On 25.09.13 19:25, JavaWebDev wrote:
 There doesn't seem to be a a DiskShrinker like there is a memory
 shrinker so it looks like the objects stored on disk will only be
 removed passively when retrieved and identified as expired? I'll look
 into the Recycle bin more.

See the method IndexedDiskCache.optimizeFile(). This will be called if
OptimizeAtRemoveCount  0 and removeCount  OptimizeAtRemoveCount.

Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[jcs] Re: JCS disk only and expiration config question

[Thomas Vandahl]

Hi JavaWebDev,

first of all, the usage of clear names of real persons would be very
much appreciated.

On 24.09.13 21:27, JavaWebDev wrote:
 First. The user and dev mailing lists haven't been active lately and the
 wiki seemed outdated in some sections. (ex: no longer a .commons
 package). Is JCS still being worked on? I chose it because it seems
 faster than ehcache and I need the disk cache to persist across servlet
 container restarts.

The latest commits were just four weeks ago. Currently, I'm maintaining
this alone. All contributions would be welcomed.

 Ideally I'd like to write cache to disk immediately in case the servlet
 crashes without being able to call dispose(). Should I set MaxObjects to
 0 in the region definition of cache.ccf?

See the documentation of the purgatory at
http://commons.apache.org/proper/commons-jcs/IndexedDiskAuxCache.html.
If you set the size of the purgatory to 0 your performance will suffer
seriously. Mind you, JCS is a cache, no data store.

 I'd like the objects in disk cache to expire after 7 days. Which
 configuration parameter(s) do I need to change?

Expiry is handled by the memory cache. Empty spots in the disk cache
will be reused. If you want to clean up every now and then, you may want
to play with the configuration options OptimizeAtRemoveCount and
MaxReycleBinSize.

HTH.
Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Remote Cache

[Thomas Vandahl]

On 25.06.2013 14:03, Sunil Kumar P wrote:

Hi , i'm new to JCS ! i was trying out Remote Cache ! In Failover Server i have 
confiured two Server . When primary Server is down . Data in cache is not empty 
secondary Server . Please Find Below Configuration


Primary Server is 10.44.71.15 9000 and Secondary Server is 10.44.71.14 9000

When i get Connection refused for 71.15 ! Data in Cache is not received in 
71.14 Server ! Please Help me OUt


I cannot see anything wrong in this configuration snippet. Please 
configure a logger for org.apache.jcs at level DEBUG and check the 
initialization sequence of the caches in the logs. JCS is quite chatty 
in DEBUG mode. You should be able to see what goes wrong.


Bye, Thomas.



-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS Lateral Cache

[Thomas Vandahl]

Hi Jean-Marc,

On 25.02.13 04:18, Jean-Marc Spaggiari wrote:
 Regarding DC + LTCP vs LTCP, is that an issue? Becaue if I stop all the
 nodes, I don't want them to have to re-query everything.

No, actually, it should not be an issue. However, if you want to clarify
the behavior of the LTCP cache I'd suggest to first make sure it does
what it should do. Only after you are satisfied with the results,
re-introduce the disk cache.

 If I stop a node, then I restart it, other servers will see that and will
 display the message above. Are they really disabling totally the lateral
 cache because of that? If yes, then it's an issue for my needs.

Normally, JCS constantly tries to fix broken caches. So I'd expect the
connection to be re-established. LateralCacheMonitor is responsible for
this. You should see the log outputs of that class later in the log.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS Lateral Cache

[Thomas Vandahl]

On 08.02.13 14:06, Jean-Marc Spaggiari wrote:
 Thanks for jumping in.

JCS became my playground by accident, after all. :-)

   props.put(jcs.region.robotCache, DC, LTCP);

I'm not sure if it makes sense to have two auxiliaries for the region.
I'd try to get the lateral thing working first.

   props.put(jcs.auxiliary.LTCP.attributes.AllowGet, 
 true);

Try to set this to false. The idea is that if some member of the lateral
cache puts an item, it will be distributed to all the others. So there
is actually no need to allow gets throughout the group.

 2013-02-08 07:41:27,243 [CacheEventQueue.QProcessor-robotCache] ERROR
 org.apache.jcs.auxiliary.lateral.socket.tcp.LateralTCPSender -
...


There is not much we can do about this in 1.3 Would you please open a
JIRA issue talking about excessive logging and I'll try to consider
this for 2.0. You may try to configure this specific class logger to be
silent.

 As you can see, the map sizes are very different on the 2 servers when
 I will have expected hem to be pretty identicals... and it's the same
 on all the servers. all have different map size.

Your stats look a bit strange to me. However I wouldn't care too much
for them. Remember that the cache is not guaranteed to be consistent.
JCS is a cache, not a key-value-store.

 So I'm wondering if the lateral cache is really used. I will say yes
 because I'm getting the exceptions when one server is going down, but
 I'm not seeing the benefits yet.

Well, benefits should be there, even if its difficult to see them from
the statistics. You would be able to see this on an application level,
that is, if an item available in one cache actually saves the expensive
get operation from the original source in another. This would need some
debugging, I'm afraid.

HTH
Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Remote Cache Server Help

[Thomas Vandahl]

On 27.11.12 16:28, Drew Leamon wrote:
 I'm embarrassed!  I just figured out the issue.  My server was using
 2.0.0-SNAPSHOT and my client was using 1.4.0.   Sorry to waste your time.

Not at all. Thanks for telling us.

Bye, Thomas


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Remote Cache Server Help

[Thomas Vandahl]

On 12.11.12 16:28, Drew Leamon wrote:
 registry.host=localhost
 registry.port=1103
 remote.cache.service.port=1103

Shouldn't these ports be different?

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Mystery in Purgatory size

[Thomas Vandahl]

On 26.09.12 17:16, Chopin hu wrote:
 I set the Purgatory size to 1000 in the default region.   I use file disk 
 cache auxiliary (not indexed) so that I have one cache item per data (binary 
 data file).I need to do periodic batch cache using JCS.If the cache 
 items are less than 1000, everything works fine.   if the cache items are 
 more than 1000, to say 1010, then roughly about  10 random cache items 
 would be lost.   

I'd not recommend the use of FileDiskCache. It has some strange file
system dependencies. Always remember that JCS is a cache, not a
key/value store.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] Performance of invalidating a group in LRUMemoryCache

[Thomas Vandahl]

On 20.06.12 23:51, Joseph A Calzaretta wrote:
 Are we doing something wrong?  Is the behavior we're seeing intended or is it 
 a bug?  If it is by design, should it perhaps be documented somewhere that 
 the performance of invalidateGroup is O(n) and not O(1)?

Obviously, the feature was not implemented with your use-case in mind.
Right now, JCS is just a map on steroids with some group support. You
are, however, free to put maps of objects into your cache to implement
what you want. Even better, submit a proposal for a patch to get this
resolved via JIRA and we try to put it in (we prefer unified patches).

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [dbutils] Mapping from JPA annotations

[Thomas Vandahl]

On 19.01.12 21:37, Moandji Ezana wrote:
 Interesting. I've never used MyBatis, but U've read the docs and it didn't
 really seem to help write less SQL.
 
 Do you all only use DbUtils for smaller projects, then?

I prefer Torque (db.apache.org/torque). Nice, fast and simple. And no
SQL to write.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS]--How to get a list of Keys in a region

[Thomas Vandahl]

On 17.01.12 03:35, Chopin hu wrote:
 Thank you so much, Thomas.   Unfortunately, due to the customer requirements, 
 the disk cache instead of indexed disk cache is what they want.  is there any 
 plan JCS will add this support to the disk cache?   I think it is really 
 useful to have it in Disk Cache, particularly for my project at Goddard NASA.

First of all it would be necessary to define what a group means for the
file disk cache (a directory perhaps?). Then implementation should be
easy. You are welcome to contribute. But see below.

 By the way, do you know any timeline for JCS release 1.4?

Not a time line, actually. However this will probably be version 2.0
supporting generics and using the org.apache.commons.jcs package name
for all classes. IOW it *will* break the API, at least partly.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS]--How to get a list of Keys in a region

[Thomas Vandahl]

On 16.01.12 18:06, Chopin hu wrote:
 java.lang.UnsupportedOperationException
 at 
 org.apache.jcs.auxiliary.disk.file.FileDiskCache.getGroupKeys(FileDiskCache.java:143)
 at 
 org.apache.jcs.engine.control.CompositeCache.getGroupKeys(CompositeCache.java:1110)
 at 
 org.apache.jcs.access.GroupCacheAccess.getGroupKeys(GroupCacheAccess.java:192)
 at gov.nasa.gsfc.giovanni.TestCache.getKeyList(TestCache.java:167) 
 at gov.nasa.gsfc.giovanni.TestCache.main(TestCache.java:608)
 
 What could be wrong?

Well, the error message basically says it: Groups are not supported in
the FileDiskCache. I'd recommend the IndexedDiskCache instead. It works
far better.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS]--How to get a list of Keys in a region

[Thomas Vandahl]

On 13.01.12 15:59, Chopin hu wrote:
 Vijayakumar,
 
 Thanks for your response.   The method cache.putInGroup() does not seem to 
 really work in my case.   I am working a simple commandline utility, which 
 will extract all keys from a cache region.   No key/value can be known in 
 advance.   Besides, I don't understand the concept of groupName yet in a 
 cache region.   It looks I don't have a group name either.
 
 But is this some method that I can trick it so that it can return all cache 
 keys from a region?
 
 
 
 --- On Fri, 1/13/12, NM, Vijayakumar vijayakumar...@capgemini.com wrote:
 
 From: NM, Vijayakumar vijayakumar...@capgemini.com
 Subject: RE: [JCS]--How to get a list of Keys in a region
 To: Commons Users List user@commons.apache.org
 Date: Friday, January 13, 2012, 12:41 AM
 
 Hi,
 
 This may be of some help.
 
 JCS cache = JCS.getInstance(regionName);
 cache.putInGroup(key, groupName, value);
 
 Set keys = cache.getGroupKeys(groupName)

This is the actual trick. Just invent some group name and put all your
keys and values into that group. Then you can get a list of keys by
calling getGroupKeys(). It would be more or less an additional wrapper
in your region.

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[JCS] Re: JCS Key Length Issue -- limited to 256 characters long?

[Thomas Vandahl]

First of all: Please use user@commons.apache.org for questions regarding
JCS.

On 04.01.12 18:12, Chopin hu wrote:
 I have a program to download files from remote sites.   Once downloaded a 
 data file, I use the URL as a key and put it to the cache so that I don't 
 have to download it again if the url is found to exist in the cache.   
 However, I found sometimes the caching would fail -- no caching is put to the 
 file system ( I am using Disk File Cache).   the url is working fine for 
 downloading.   After debugging, it turns out that the URL string is too long. 
   When the url string exceeds 256 characters, JCS won't accept it as a proper 
 key.   
 
 So my questions are:
 1.) is a JCS cache key limited to 256 chars long?   is there a way to 
 increase its length?
 2.) If can not get around the cache key length limitation, what could be the 
 possible ways to make a long URL string short enough but still unique enough 
 to serve as a key?  I tried to use the last 256 characters from the url 
 string, but some query string is so long that it could not guarantee a key's 
 uniqueness.

I guess the FileDiskCache is the issue here. The key of the item in
FileDiskCache is directly used as a file name. So what you experience is
a limit of file name length rather than key length.

You should better use another disk cache implementation such as
IndexedDiskCache which does not have such limits.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: Need Help on JCS Remote Cache

[Thomas Vandahl]

On 26.12.11 09:11, Narendra Verma wrote:
 Hi All,
 
 I have made setup of JCS in my Linux environment and trying to make a POC 
 using Remote Cache.

I cannot help you much with this problem. Just a hint that the users
list for JCS changed to user@commons.apache.org. I redirected your
message to there.

Bye, Thomas.


 
 My System Setup:
 
 
 * RMI Server : Running on 10.55.164.215 (using RMI Start up Servlet)
 
 * Client-1: Running on box 10.55.164.215 on same machine where serer 
 is running
 
 * Client-2: Running on box 10.55.164.45
 
 
 My Configuration Files:
 
 * Server Configuration
 
 
 # --
 # Registry used to register and provide the
 # IRemoteCacheService service.
 registry.host=10.55.164.215
 registry.port=1102
 # call back port to local caches.
 remote.cache.service.port=1102
 # rmi socket factory timeout
 remote.cache.rmiSocketFactoryTimeoutMillis=5000
 # cluster setting
 remote.cluster.LocalClusterConsistency=false
 remote.cluster.AllowClusterGet=false
 # --
 
 # DEFAULT CACHE REGION
 jcs.default=
 jcs.default.cacheattributes=org.apache.jcs.engine.CompositeCacheAttributes
 jcs.default.cacheattributes.MaxObjects=1000
 jcs.default.cacheattributes.MemoryCacheName=org.apache.jcs.engine.memory.lru.LRUMemoryCache
 
 # PRE-DEFINED CACHE REGIONS
 jcs.region.userCache=
 jcs.region.userCache.cacheattributes=org.apache.jcs.engine.CompositeCacheAttributes
 jcs.region.userCache.cacheattributes.MaxObjects=1000
 jcs.region.userCache.cacheattributes.MemoryCacheName=org.apache.jcs.engine.memory.lru.LRUMemoryCache
 jcs.region.userCache.cacheattributes.UseMemoryShrinker=false
 jcs.region.userCache.cacheattributes.MaxMemoryIdleTimeSeconds=3600
 jcs.region.userCache.cacheattributes.ShrinkerIntervalSeconds=60
 jcs.region.userCache.cacheattributes.MaxSpoolPerRun=500
 jcs.region.userCache.elementattributes=org.apache.jcs.engine.ElementAttributes
 jcs.region.userCache.elementattributes.IsEternal=false
 
 
 
 
 
 
 * Client-1 and Client-2  Configuration at separate boxes - See the 
 above in System Setup Section
 # DEFAULT CACHE REGION
 # sets the default aux value for any non configured caches
 jcs.default=RC
 jcs.default.cacheattributes=org.apache.jcs.engine.CompositeCacheAttributes
 jcs.default.cacheattributes.MaxObjects=1000
 jcs.default.cacheattributes.MemoryCacheName=org.apache.jcs.engine.memory.lru.LRUMemoryCache
 jcs.default.elementattributes.IsEternal=false
 jcs.default.elementattributes.MaxLifeSeconds=3600
 jcs.default.elementattributes.IdleTime=1800
 jcs.default.elementattributes.IsSpool=true
 jcs.default.elementattributes.IsRemote=true
 jcs.default.elementattributes.IsLateral=false
 
 # CACHE REGIONS AVAILABLE
 
 # Regions preconfigured for caching
 jcs.region.userCache=RC
 jcs.region.userCache.cacheattributes=org.apache.jcs.engine.CompositeCacheAttributes
 jcs.region.userCache.cacheattributes.MaxObjects=1200
 jcs.region.userCache.cacheattributes.MemoryCacheName=org.apache.jcs.engine.memory.lru.LRUMemoryCache
 jcs.region.userCache.elementattributes.IsEternal=false
 jcs.region.userCache.elementattributes.MaxLifeSeconds=7200
 jcs.region.userCache.elementattributes.IdleTime=1800
 jcs.region.userCache.elementattributes.IsSpool=true
 jcs.region.userCache.elementattributes.IsRemote=true
 jcs.region.userCache.elementattributes.IsLateral=false
 
 # Remote RMI Cache set up to failover
 # This remote client does not receive
 jcs.auxiliary.RC=org.apache.jcs.auxiliary.remote.RemoteCacheFactory
 jcs.auxiliary.RC.attributes=org.apache.jcs.auxiliary.remote.RemoteCacheAttributes
 jcs.auxiliary.RC.attributes.FailoverServers=10.55.164.215:1102
 jcs.auxiliary.RC.attributes.LocalPort=1202
 jcs.auxiliary.RC.attributes.RemoveUponRemotePut=true
 jcs.auxiliary.RC.attributes.RmiSocketFactoryTimeoutMillis=5000
 jcs.auxiliary.RC.attributes.GetOnly=false
 jcs.auxiliary.RC.attributes.Receive=true
 
 
 
 
 
 
 
 Client Test Programs:
 
 Input parameters  while running client-1 : java com.brightsky.util.JCSTest  
 /cache-client1.ccf
 Input parameters  while running client-2 : java com.brightsky.util.JCSTest  
 /cache.ccf-client1
 
 
 package com.brightsky.util;
 import java.util.Date;
 
 /*
  * This code is property of BrightSky, Inc. Use, duplication and disclosure
  * in any form without permission of copyright holder is prohibited.
  *
  * (C) Copyright BrightSky, Inc. 2011. All rights reserved.
  */
 
 /**
  * @author nverma
  *
  */
 public class JCSTest {
 public static void main(String a[]) throws InterruptedException {
 
 System.out.println(\n 
 ===\n
  );
 System.out.println(\n === QA Node - 
 +a[0]+ \n );
 System.out.println(\n 
 ===\n
  );
 
  

Re: [JCS] - Remote store recovery

[Thomas Vandahl]

On 20.12.11 08:57, NM, Vijayakumar wrote:
 Hi All,
 
 One of the JCS feature mentioned in the website is Remote store recovery, 
 can somebody point me the documentation or example of how this is achieved. 
 Thanks in advance.

Maybe this?

http://commons.apache.org/jcs/apidocs/org/apache/jcs/auxiliary/remote/RemoteCacheRestore.html

Bye, Thomas.


-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: JCS LFU Implementation

[Thomas Vandahl]

Am 09.12.2011 22:06, schrieb Mir Tanvir Hossain:

Hi, from the documentation it seems that JCS does support LFU. However, I
could not find the LFU implementation on the JCS code. I might be missing
something very obvious. Could anyone please point out the class on JCS that
implements the LFU algorithm? I'd really appreciate that.


Maybe this was meant?

http://svn.apache.org/viewvc/commons/proper/jcs/trunk/src/experimental/org/apache/jcs/engine/memory/arc/

Bye, Thomas

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] the data becomes zero

[Thomas Vandahl]

On 15.09.11 07:27, NM, Vijayakumar wrote:
 Thanks for the reply, I have seen this problem many times, sometimes during 
 server restart but most of the time it is during usage of JCS, for example I 
 access data from JCS and next moment the entire cache file is empty and I 
 don't get any data.

Could I ask you for some more investigations? If the reset happens
during normal operation, you should see a log message at warning level
containing The file is corrupt. Do you see this?

I modified the responsible class IndexedDisk recently to use NIO for the
disk operations. Would you be able to test with the current trunk of JCS?

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

Re: [JCS] the data becomes zero

[Thomas Vandahl]

On 14.09.11 15:31, NM, Vijayakumar wrote:
 We are using IndexedDiskCache, the size of the data file in disk is about 
 250MB, the problem is sometimes, the data gets erased from the disk (the data 
 file size becomes zero) without any error message any where?

Does this happen during normal use or only at restart? The restart
mechanism is somewhat picky concerning problems in the file and tends to
reset the data file then.

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org

[jcs] Re: Is there any method to save .key file other than dispose()/shutdown()

[Thomas Vandahl]

On 27.08.11 14:12, gurunath wrote:
 I am trying to use jcs indexeddiskcache in my web application.  Here
 i need to save the cached data in disk whenever i change cached data, data
 file is updating but key dose't update untill we call dispose()/shutdown().
 After calling dispose()/shutdown() i couldn't preform put()/get()
 operations.
 
I tried with BlockDiskCache by specifying
 KeyPersistenceIntervalSeconds attribute but this one degrades performance if
 we keep less time.
 
So I am looking for other methods to save .key file. Is there any
 other method?

Please note that JCS moved to Apache Commons recently, so in the future
please direct your JCS-related questions to user@commons.apache.org and
mention [jcs] in the subject.

On the subject: JCS doesn't provide this feature out of the box. You
are, however, free to derive from IndexedDiskCache to expose the
required functionality. This should be no big change. My question would
be though, why you want to save the cache data on put. Are you sure that
using a *cache* is the right thing to do for your application?

Bye, Thomas.

-
To unsubscribe, e-mail: user-unsubscr...@commons.apache.org
For additional commands, e-mail: user-h...@commons.apache.org