Thanks for clarifying that, it finally makes sense.
On 23 November 2011 13:21, Manik Surtani <manik(a)jboss.org> wrote:
putForExternalRead() is only used when reading from a DB and saving
the value in the cache. So this does not affect updates. Updates call put(), not pFER(),
and put() will cause the tx to abort of the put() fails.
On 22 Nov 2011, at 19:15, Sanne Grinovero wrote:
> Why is there a general agreement about not aborting the transaction if
> we fail to update the cache?
>
> I can see why when we fail to write it for the first time, but if we
> fail to update a value which is stored already that's going to affect
> staleness of data read by subsequent transactions.
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/infinispan-dev
--
Manik Surtani
manik(a)jboss.org
twitter.com/maniksurtani
Lead, Infinispan
http://www.infinispan.org
_______________________________________________
infinispan-dev mailing list
infinispan-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/infinispan-dev