Membase proxy write to downstream error on 1.6.5, out of usable disk space
My membase cluster is down with the errror SERVER_ERROR proxy_write_downstream. The logs report " Service memcached exited " every few seconds. Moreover, there is no usable disk space left, even though my keys are set to expire after 15 minutes. Are there known issues with membase purging data on disk?
I see the long thread that appears to end in resolution for an earlier version in January.
Are there any known causes of this error in 1.6.5? My whole app is down now, I recovered by launching a new cluster, but will presumably hit this error again on my new cluster as my disk usage is growing and will hit max in a few hours again.