portals-jetspeed-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Kimpton <kimptoc_m...@yahoo.com>
Subject RE: Help! Caching (?) problem in Jetspeed
Date Fri, 10 May 2002 17:50:34 GMT
> --- David Sean Taylor <david@bluesunrise.com> wrote:
> > 
> > 
> > > 
> > > I have no idea what's going on, or why it thinks there's a 
> > > cache miss, but more importantly, why it gives up the ghost 
> > > and stops responding.
> > > 
> > 
> > Try pressing Ctrl-Break (at least this works on Windows) to get
> the
> > state of the active threads
> > That can least give us a starting point.
> > 
> 
> OK - I did this.  These seem to be the key points:
> 
> 
> THE GET PORTLET IS WAITING TO GET A LOCK ON THE DISK CACHE
> INSTANCES
> HASHTABLE
> 
> "Thread-7" prio=5 tid=0x0EFF2C20 nid=0x4d8 waiting for monitor
> entry
> [f5ee000..f5efdb4]
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:490)
>         - waiting to lock <032686E0> (a java.util.Hashtable)
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:478)
>         at
>
org.apache.jetspeed.portal.portlets.JetspeedContent.parse(JetspeedContent.java:176)
>         at
>
org.apache.jetspeed.portal.portlets.JetspeedContent.init(JetspeedContent.java:142)
>         at
>
org.apache.jetspeed.services.portletfactory.JetspeedPortletFactoryService.getPortlet(JetspeedPortletFactoryService.java:310)
> ...
> 
> 
> 
> HAS THE INSTANCES HASHTABLE LOCK, WAITING ON URL MGR INIT
> -BUT THE MESSAGES SEEM TO INDICATE THE URL MGR HAS DONE THE EARLY
> INIT - SO IT SHOULD HAVE FINISHED THE EARLY INIT...
> 
> "DaemonThread:diskcachedaemon" daemon prio=2 tid=0x0AC51668
> nid=0xa30
> waiting on monitor [f35f000..f35fdb4]
>         at java.lang.Thread.sleep(Native Method)
>         at
>
org.apache.jetspeed.services.urlmanager.JetspeedURLManagerService.init(JetspeedURLManagerService.java:107)
>         at
>
org.apache.turbine.services.BaseServiceBroker.getService(BaseServiceBroker.java:306)
>         - locked <06E9EA00> (a java.lang.Class)
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.initEntries(JetspeedDiskCache.java:151)
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:496)
>         - locked <032686E0> (a java.util.Hashtable)
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:478)
>         at
>
org.apache.jetspeed.daemon.impl.DiskCacheDaemon.run(DiskCacheDaemon.java:98)
> ...
> 
> 
> 
> WAITING FOR INSTANCES HASHTABLE
> 
> "DaemonThread:feeddaemon" daemon prio=2 tid=0x0AB73D48 nid=0xc5c
> waiting for monitor entry [f31f000..f31fdb4]
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:490)
>         - waiting to lock <032686E0> (a java.util.Hashtable)
>         at
>
org.apache.jetspeed.cache.disk.JetspeedDiskCache.getInstance(JetspeedDiskCache.java:478)
>         at
>
org.apache.jetspeed.util.SimpleTransform.SAXTransform(SimpleTransform.java:430)
>         at
>
org.apache.jetspeed.daemon.impl.FeedDaemon.getEntries(FeedDaemon.java:256)
>         at
> org.apache.jetspeed.daemon.impl.FeedDaemon.run(FeedDaemon.java:194)
> ...
> 
> 
> So - why is JetspeedURLMgrService is hanging on late init when its
> done the early init...
> 

But anyway, how can 2 threads be inside the synchronized section of
the JetspeedDiskCache.getInstance method - isn't that the point of
the sync...

Any reason why instances needs to be a Hashtable - can we make it a 
non-sync'd Hashmap instead - I'll try it and see...

Chris

=====
------------------------------------------
http://www.soccer2002.org.uk - The Game is On!

__________________________________________________
Do You Yahoo!?
Yahoo! Shopping - Mother's Day is May 12th!
http://shopping.yahoo.com

--
To unsubscribe, e-mail:   <mailto:jetspeed-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:jetspeed-dev-help@jakarta.apache.org>


Mime
View raw message