We got around this issue by grouping memcache servers into logical clusters within our client (2 or 3).
While performing a cache "put", we put into all the clusters (which will put it in just one of the node within the cluster). However while performing a "get", we perform a get from subsequent cluster, only if the previous one fails. With this setup, a memcache server does not act as a single point of failure and if a random memcache server does go down, we can always find the cache from another logical cluster Probably not the best approach to solve this, but if there is any other better approach, please let me know. -- --- You received this message because you are subscribed to the Google Groups "memcached" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
