Hi, Have you seen the document about cache flushing delivered with the sp2 and sp3 ? Have you monitored any UDP traffic on port 2712 ? br,
No to both questions. I don’t have this document, and it isn’t possible for me to trace UDP on a cluster. In fact I cannot investigate this matter further. That’s why I posted this stuff; I hope I can interest other people to continue.
quote:
Hi, Have you seen the document about cache flushing delivered with the sp2 and sp3 ? Have you monitored any UDP traffic on port 2712 ? br,
Originally posted by Esa - 2005 May 31 : 11:23:03
Would you have a link to this document? I’d appreciate it, thanks! Regards, JRA
Hi, A small clarification - Basically MS introduced a new cache mechanism (common flush) with ver 3.0 SP2 and ver 2.5 SP4. You can find this documentation in the product CD of Axapta 3.0 SP2. But I don’t think there was any new information on caching in Axapta 3.0 SP3 product CD. Till some time ago, end users were able to download Axapta documentation from -http://technet.navision.com/usered/Axapta30/Axapta30.htm For reason unknown, this site doesn’t seem to work anymore [:(] Any way, if you need this document please drop me an email. Regards, Harish Mohanbabu
We have experienced a lot of the same issues that have been mentioned here during our data migration phase. The only other piece I can add is that most of the problems were related to “highly static” or pre-loaded tables. Changes to these tables are made directly to the database, but the clustered servers will only be refreshed at night, making it look like the data has not changed through the interface. Our consultant tells us this is “by design”. The static tables do not need to be viewable for these tables because any transactions that rely on the data will still be correct as they are direct from the database. I don’t like it and I have trouble explaining it to the users, but I guess it will work.
Hi all, Interesting Topic! We work with Ax3.0 Sp4 at this moment with 2 company’s live (about 180 and 45 users average logged on) and 3 new company’s planned for this year (45 and 65 uses and one starting with 20 and growing to about 400 next year and a half…). For the first to company’s we are busy installing a cluster (2 servers, 1 instance of each company on each server). In a first small test we got a call that seems to be a caching problem (one user didn’t see the changes of the other…). Max. cache sync. Time is not filled ! Any idea whether this can be the reason for not caching / the differences noticed by the users? Thanks, Johan