Thanks again, Dustin. On Tue, Jan 5, 2010 at 1:35 PM, Dustin <[email protected]> wrote:
> > Sorry, I finally woke up and started responding to email. > > Summary: The upgrade breaks setTranscoder (client mutation is a bad > idea) and delete with delay (which was removed from the server). > Other than that, upgrade should only make things better. > > On Jan 5, 6:56 am, Boris <[email protected]> wrote: > > Hi all, I posted this to spymemcached group as well, see if I can get > > better luck here: > > > > I am currently running spy memcached 2.1 client and 1.2.5 server. Both > > run on centos > > (2.6.18 kernel). Looking into upgrading to 2.4.4, mainly for > > configurable "failure mode" support, which is non existant in 2.1. So > > far I noticed one braking interface change > > (MemcachedClient::setTranscoder is gone), but other than that I was > > wondering if there are any known issues with: > > > > 1. Running 2.1 and 2.4.4 clients side by side against the same > > memcached cluster (I am using default SerializingTranscoder). > > > > 2. Running 2.4.4 against 1.2.5 server (I am using ascii protocol) > > > > Thanks, > > Boris > -- --Boris
