Would it be better, if the actual connection pool is defer started
until client actually call get or set ?
Since client application is expecting network error, so it is a much
better time for client to handle such situation ?
Also, if I set the pool size to be larger than 1, then current code
will try many times to connect to the remote server in sequence. This
will pose a much longer delay for client application if the remote
memcache server is down.
Looking for your update.
Thanks.
2011/11/3 dennis <killm...@gmail.com>:
> --
> You received this message because you are subscribed to the Google Groups "xmemcached" group.
> To post to this group, send email to xmemc...@googlegroups.com.
> To unsubscribe from this group, send email to xmemcached+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/xmemcached?hl=en.
>
>