Re: [OFFTOPIC] Removing attribute from all HttpSessions

2004-08-19 Thread Michiel Toneman
Bad caching strategy.
Try using a singleton cache backed by a HashMap. Have an invalidation 
method that takes a pagekey as argument. Create a simple get(pagekey) 
that checks for invalidation and refills the cache entry when necessary.

No need to go iterating over sessions...
Just my $0.02 :-)
Michiel
Allistair Crossley wrote:
Hey,
I am putting a page caching stategy into place per user session. However, there are 
times when I need to refresh all the caches for all users. For example, let's say 5 
users have a cached version of page 5. An external event causes page 5 to invalidate. 
I need an effective strategy to flush all sessions with page 5 cached.
Thoughts appreciated, Allistair.
FONT SIZE=1 FACE=VERDANA,ARIAL COLOR=BLUE 
---
QAS Ltd.
Developers of QuickAddress Software
a href=http://www.qas.com;www.qas.com/a
Registered in England: No 2582055
Registered in Australia: No 082 851 474
---
/FONT

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
 

--
Michiel Toneman  Software Engineer   Bibit Global Payment Services
Regulierenring 10  3981 LB  Bunnik   [EMAIL PROTECTED]
Tel. +31-30-6595168  Fax +31-30-6564464  http://www.bibit.com/
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


RE: [OFFTOPIC] Removing attribute from all HttpSessions

2004-08-19 Thread Allistair Crossley
I should have explained some more I guess. I am currently using OSCACHE at the UI 
layer to cached search results strings per page ID in application scope. It is very 
effective indeed. The search results come from a content management system. When users 
interact with the CMS, events are fired to a listener and I can determine which page 
in the cache needs updating.

However I realised this morning that search results per row are security contolled. 
Therefore user A's search results for page 5 are not necessarily the same as user B's. 

Hence the need for user-level caching. I used my own caching Hashes before but OSCACHE 
is very powerful. OSCACHE's cache JSP tag also provides only application and session 
scope, therefore I was thinking session caching must be the only other solution for my 
problem. But I need to be able to programmatically flush many user sessions. Perhaps I 
need to store the caches created per session somewhere so I can grab them.

Allistair Crossley

 -Original Message-
 From: Michiel Toneman [mailto:[EMAIL PROTECTED]
 Sent: 19 August 2004 09:57
 To: Tomcat Users List
 Subject: Re: [OFFTOPIC] Removing attribute from all HttpSessions
 
 
 Bad caching strategy.
 
 Try using a singleton cache backed by a HashMap. Have an invalidation 
 method that takes a pagekey as argument. Create a simple get(pagekey) 
 that checks for invalidation and refills the cache entry when 
 necessary.
 
 No need to go iterating over sessions...
 
 Just my $0.02 :-)
 
 Michiel
 
 Allistair Crossley wrote:
 
 Hey,
 
 I am putting a page caching stategy into place per user 
 session. However, there are times when I need to refresh all 
 the caches for all users. For example, let's say 5 users have 
 a cached version of page 5. An external event causes page 5 
 to invalidate. I need an effective strategy to flush all 
 sessions with page 5 cached.
 
 Thoughts appreciated, Allistair.
 
 
 FONT SIZE=1 FACE=VERDANA,ARIAL COLOR=BLUE 
 ---
 QAS Ltd.
 Developers of QuickAddress Software
 a href=http://www.qas.com;www.qas.com/a
 Registered in England: No 2582055
 Registered in Australia: No 082 851 474
 ---
 /FONT
 
 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 
   
 
 -- 
 
 Michiel Toneman  Software Engineer   Bibit Global Payment Services
 Regulierenring 10  3981 LB  Bunnik   [EMAIL PROTECTED]
 Tel. +31-30-6595168  Fax +31-30-6564464  http://www.bibit.com/
 
 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 
 


FONT SIZE=1 FACE=VERDANA,ARIAL COLOR=BLUE 
---
QAS Ltd.
Developers of QuickAddress Software
a href=http://www.qas.com;www.qas.com/a
Registered in England: No 2582055
Registered in Australia: No 082 851 474
---
/FONT


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: [OFFTOPIC] Removing attribute from all HttpSessions

2004-08-19 Thread Tim Funk
Look at the SessionListeners to register all the sessions that match your 
criteria into a map. Then whatever is listening on your invalidation strategy 
can update the map. (Actually - the sessions in the map)

-Tim
Allistair Crossley wrote:
Hey,
I am putting a page caching stategy into place per user session. However, there are times when I need to refresh all the caches for all users. For example, let's say 5 users have a cached version of page 5. An external event causes page 5 to invalidate. I need an effective strategy to flush all sessions with page 5 cached.
 
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


RE: [OFFTOPIC] Removing attribute from all HttpSessions

2004-08-19 Thread Allistair Crossley
Thanks Tim. Done that ... any reason why sessionCreated is always called twice?

2004-08-19 12:31:02,920 - DEBUG (com.qas.newmedia.intranet.iq.IQSessionListener:31) - 
new session: 0E3C9F21B145FF21903D8D35E540217F
2004-08-19 12:31:02,920 - DEBUG (com.qas.newmedia.intranet.iq.IQSessionListener:31) - 
new session: 0E3C9F21B145FF21903D8D35E540217F

Also, sessionDestroyed I am guessing is not actually called when a user closes their 
browser but on session timeout after inactivity correct?

Allistair Crossley

 -Original Message-
 From: Tim Funk [mailto:[EMAIL PROTECTED]
 Sent: 19 August 2004 11:48
 To: Tomcat Users List
 Subject: Re: [OFFTOPIC] Removing attribute from all HttpSessions
 
 
 Look at the SessionListeners to register all the sessions 
 that match your 
 criteria into a map. Then whatever is listening on your 
 invalidation strategy 
 can update the map. (Actually - the sessions in the map)
 
 -Tim
 
 Allistair Crossley wrote:
 
  Hey,
  
  I am putting a page caching stategy into place per user 
 session. However, there are times when I need to refresh all 
 the caches for all users. For example, let's say 5 users have 
 a cached version of page 5. An external event causes page 5 
 to invalidate. I need an effective strategy to flush all 
 sessions with page 5 cached.
   
 
 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]
 
 


FONT SIZE=1 FACE=VERDANA,ARIAL COLOR=BLUE 
---
QAS Ltd.
Developers of QuickAddress Software
a href=http://www.qas.com;www.qas.com/a
Registered in England: No 2582055
Registered in Australia: No 082 851 474
---
/FONT


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Re: [OFFTOPIC] Removing attribute from all HttpSessions

2004-08-19 Thread Tim Funk
If its in web.xml twice - it would be called twice.
sessionDestroyed is called when the container kills the session. Which is 
done during session timeout or session.invalidate()

-Tim
Allistair Crossley wrote:
Thanks Tim. Done that ... any reason why sessionCreated is always called twice?
2004-08-19 12:31:02,920 - DEBUG (com.qas.newmedia.intranet.iq.IQSessionListener:31) - 
new session: 0E3C9F21B145FF21903D8D35E540217F
2004-08-19 12:31:02,920 - DEBUG (com.qas.newmedia.intranet.iq.IQSessionListener:31) - 
new session: 0E3C9F21B145FF21903D8D35E540217F
Also, sessionDestroyed I am guessing is not actually called when a user closes their 
browser but on session timeout after inactivity correct?
Allistair Crossley

-Original Message-
From: Tim Funk [mailto:[EMAIL PROTECTED]
Sent: 19 August 2004 11:48
To: Tomcat Users List
Subject: Re: [OFFTOPIC] Removing attribute from all HttpSessions
Look at the SessionListeners to register all the sessions 
that match your 
criteria into a map. Then whatever is listening on your 
invalidation strategy 
can update the map. (Actually - the sessions in the map)

-Tim
Allistair Crossley wrote:

Hey,
I am putting a page caching stategy into place per user 
session. However, there are times when I need to refresh all 
the caches for all users. For example, let's say 5 users have 
a cached version of page 5. An external event causes page 5 
to invalidate. I need an effective strategy to flush all 
sessions with page 5 cached.

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]