• Post Reply Bookmark Topic Watch Topic
  • New Topic
programming forums Java Mobile Certification Databases Caching Books Engineering Micro Controllers OS Languages Paradigms IDEs Build Tools Frameworks Application Servers Open Source This Site Careers Other Pie Elite all forums
this forum made possible by our volunteer staff, including ...
Marshals:
  • Campbell Ritchie
  • Jeanne Boyarsky
  • Ron McLeod
  • Paul Clapham
  • Liutauras Vilda
Sheriffs:
  • paul wheaton
  • Rob Spoor
  • Devaka Cooray
Saloon Keepers:
  • Stephan van Hulst
  • Tim Holloway
  • Carey Brown
  • Frits Walraven
  • Tim Moores
Bartenders:
  • Mikalai Zaikin

Internal Replication Domain not working in WAS ND 5.0.2

 
Greenhorn
Posts: 5
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
The cross memory replication is not working. In a failover, the session timestamps ar correct, but the session data is missing - sort of.
The IHS server is on a AIX host, the Deployment Manager and Application server instances are on a separate AIX host.
A cluster (POC) with 3 application servers (POC, POC1, POC2) is defined to use internal replication (domain POC) with each application server defined to be client and server with Tuning parameter set to Medium (FYI: same result when we tried Low).
A POC.ear is deployed to the cluster that has 2 servlets, one is the standard snoop and the other is a servlet that sets session values and echos the values to the browser.
All works as expected until the app server is stopped to force failover. When that happens, snoop is executed with only session information displayed, no session values. If more session values are "submitted", the input parameters that are set in the session are not echoed back, but the last values set in the previous app server are. This behavior is noted consistently accross each of the app servers when they participate in a failover.
We have even tried the client/server setup with the same exact results.
Any ideas?
Ray
 
Greenhorn
Posts: 16
  • Mark post as helpful
  • send pies
    Number of slices to send:
    Optional 'thank-you' note:
  • Quote
  • Report post to moderator
Our solaris 8 environment (WAS 5.0.2 too) works fine - failover using mem-to-mem replication works flawless (client-client topology worked ever since, client-server topology works since WAS 5.0.2), although I used 2 JSPs for storing and accessing (printing) session data.
I'm not sure if this is the cause for your problem, but I've encountered some kind of a "missfunktion" (bug???) in session-management when using mem-to-mem replication. Invalidation of a session and use of getSession to create a new one results in the return of the old session instead of a new one. Also (and that's the important part) session data stored after invalidation gets lost! in subsequent calls. After redirection of the request from the data-saving servlet to a data accessing and printing JSP the data is no longer present.
I've opened a request with IBM support on this one...
Another cause might be a misconfiguration of your replicators (replicator entries). For each replicator inside your replication domain the appropriate AppServer must be specified (I think via a dropdown box when creating the replicator). A wrong selection renders the mem-to-mem replication useless

Hope this helps.
 
Don't get me started about those stupid light bulbs.
reply
    Bookmark Topic Watch Topic
  • New Topic