memcached failover

[Marvin Addison]

On Mon, Aug 1, 2016 at 2:02 PM Niva Agmon <nagmon at temple.edu> wrote:

> We setup our two Shib nodes with memcached per the Storage Configuration
> <https://wiki.shibboleth.net/confluence/display/IDP30/StorageConfiguration#StorageConfiguration-MemcachedStorageService>
> article on the wiki but we still have to re-login to the idp when one of
> the nodes is down. Is this the expected behavior?
>
Yes, that is expected behavior and it's mentioned in the page you cited:

The high-level effect of this behavior on the IdP session management
service is that a node failure will cause loss of IdP session, which would
impact users as an unexpected authentication.

If you have somehow enabled memcached replication, then I can't speak to
the behavior since I have no experience with that mode of operation.

M
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://shibboleth.net/pipermail/users/attachments/20160803/669054c2/attachment.html>