[infinispan-dev] getting stopped cache
Ales Justin
ales.justin at gmail.com
Tue Aug 28 04:10:18 EDT 2012
NPE is thrown at diff place; see #125 line in LM class.
No initializer is found, hence NPE on initializer::initialize(Cache).
Sent from my iPad
On Aug 28, 2012, at 9:25, Dan Berindei <dan.berindei at gmail.com> wrote:
> Cache.stop() should not remove the cache from the cache manager.
>
> But AFAIK the ComponentRegistry that holds all the cache-scoped components is destroyed, and recreated by Cache.start(). In the constructor it should register Query's CommandInitializer again, so LifecycleManager should find it and you shouldn't get the NPE...
>
> Dan
>
>
> On Tue, Aug 28, 2012 at 12:15 AM, Ales Justin <ales.justin at gmail.com> wrote:
> Uh, the workaround results in NPE:
>
> 23:10:27,160 ERROR [org.apache.catalina.core.ContainerBase.[jboss.web].[default-host]] (http-/192.168.1.101:8080-2) Exception sending request initialized lifecycle event to listener instance of class org.jboss.capedwarf.appidentity.GAEListener: java.lang.NullPointerException
> at org.infinispan.query.impl.LifecycleManager.cacheStarted(LifecycleManager.java:125)
> at org.infinispan.factories.ComponentRegistry.start(ComponentRegistry.java:178) [infinispan-core-5.2.0.ALPHA2.jar:5.2.0.ALPHA2]
> at org.infinispan.CacheImpl.start(CacheImpl.java:523) [infinispan-core-5.2.0.ALPHA2.jar:5.2.0.ALPHA2]
> at org.infinispan.AbstractDelegatingCache.start(AbstractDelegatingCache.java:343) [infinispan-core-5.2.0.ALPHA2.jar:5.2.0.ALPHA2]
> at org.jboss.capedwarf.common.infinispan.InfinispanUtils.checkCache(InfinispanUtils.java:72) [capedwarf-common-1.0.0-SNAPSHOT.jar:1.0.0-SNAPSHOT]
> at org.jboss.capedwarf.common.infinispan.InfinispanUtils.getCache(InfinispanUtils.java:82) [capedwarf-common-1.0.0-SNAPSHOT.jar:1.0.0-SNAPSHOT]
>
> This is the code I added:
>
> private static <K, V> Cache<K, V> checkCache(String cacheName) {
> final Cache<K, V> cache = cacheManager.getCache(cacheName, false);
> if (cache != null) {
> final ComponentStatus status = cache.getStatus();
> if (status != ComponentStatus.INITIALIZING && status != ComponentStatus.RUNNING) {
> cache.start();
> }
> }
> return cache;
> }
>
> Any other idea?
>
> -Ales
>
> On Aug 27, 2012, at 10:51 PM, Ales Justin <ales.justin at gmail.com> wrote:
>
> > OK, let me try this workaround.
> >
> > But should Cache::stop also remove it or not?
> >
> > If yes, I might dig deeper into finding the culprit,
> > if not, then the workaround should do, I guess.
> >
> > -Ales
> >
> >> As a workaround, couldn't you check what the status of the cache is before using it? If it's stopped, you could call cache.start() again.
> >>
> >> On Aug 25, 2012, at 3:44 PM, Ales Justin wrote:
> >>
> >>> When running CapeDwarf tests these days I get this broken behavior:
> >>> (using: Ispan 5.2.0.Alpha2)
> >>>
> >>> On 1st test all works OK:
> >>> * getting the X cache
> >>> * passing it to SearchManager (from Ispan Query)
> >>> * doing queries, etc
> >>> * stopping the X cache on app (aka Arquillian test) undeploy
> >>>
> >>> But when running 2nd app (more Arquillian tests) it breaks.
> >>>
> >>> I'm again using X cache, via this lookup:
> >>> (where toCacheName == X)
> >>>
> >>> final Cache<K, V> cache = cacheManager.getCache(toCacheName(config, appId), false);
> >>> if (cache != null)
> >>> return cache;
> >>>
> >>> And I actually get some cache, where I would expect none.
> >>>
> >>> Since this cache is not new, it's never run against Ispan Query's LifecycleManager,
> >>> hence there is no QueryInterceptor in its component registry.
> >>>
> >>> Which is where the app breaks: SearchManager requires QueryInterceptor in cache' component registry.
> >>>
> >>> Shouldn't Cache::stop also remove the cache from EmbeddedCacheManager?
> >>> btw: this used to work, dunno when it broke ...
> >>>
> >>> Anyone else seeing something similar?
> >>>
> >>> -Ales
> >>>
> >>> ----------
> >>>
> >>> 15:32:06,810 ERROR [org.apache.catalina.core.ContainerBase.[jboss.web].[default-host]] (http-/192.168.1.101:8080-3) Exception sending request initialized lifecycle event to listener instance of class org.jboss.capedwarf.appidentity.GAEListener: java.lang.IllegalArgumentException: Indexing was not enabled on this cache. class org.infinispan.query.backend.QueryInterceptor not found in registry
> >>> at org.infinispan.query.impl.ComponentRegistryUtils.getComponent(ComponentRegistryUtils.java:40)
> >>> at org.infinispan.query.SearchManagerImpl.<init>(SearchManagerImpl.java:57)
> >>> at org.infinispan.query.Search.getSearchManager(Search.java:39)
> >>> at org.jboss.capedwarf.datastore.AbstractDatastoreService.<init>(AbstractDatastoreService.java:61) [capedwarf-datastore-1.0.0-SNAPSHOT.jar:1.0.0-SNAPSHOT]
> >>> at org.jboss.capedwarf.datastore.JBossDatastoreService.<init>(JBossDatastoreService.java:56) [capedwarf-datastore-1.0.0-SNAPSHOT.jar:1.0.0-SNAPSHOT]
> >>> at com.google.appengine.api.datastore.DatastoreServiceFactory.getDatastoreService(DatastoreServiceFactory.java) [appengine-api-1.0-sdk-1.7.0.jar:]
> >>> at org.jboss.capedwarf.log.JBossLogService.requestStarted(JBossLogService.java:215) [capedwarf-log-1.0.0-SNAPSHOT.jar:1.0.0-SNAPSHOT]
> >
>
>
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev
>
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/infinispan-dev/attachments/20120828/fb976a6c/attachment-0001.html
More information about the infinispan-dev
mailing list