Viewing posts for the category GlusterFSJoe Julian 2 years ago
I filed a bug back in July 2013 asserting that a "feature" in the gluster CLI was a potential security bug. I've recently had that bug rejected as "Not a bug" so here it is.Joe Julian 3 years ago
Split brain. Sounding like something from a B movie about zombies, it's probably more terrifying to data storage people than flesh eating undead would be.Joe Julian 3 years, 4 months ago
GlusterFS communicates over TCP. This allows for stateful handling of file descriptors and locks. If, however, a server fails completely, kernel panic, power loss, some idiot with a reset button... the client will wait for ping-timeout (42 by the defaults) seconds before abandoning that TCP connection. This is important because re-establishing FDs and locks can be a very expensive operation. As glusterbot says in #gluster:Joe Julian 3 years, 6 months ago
I retweeted the other day:Joe Julian 3 years, 7 months ago
A couple times a year someone comes in to #gluster and thinks it would be awesome to ave a puppet module that you could define servers and the pupet module would automatically manage adding those servers as peers and dynamically changing volumes to utilize the new assets. They get inspired and start writing manifests... then they realize they actually have to get something completed and abandon that pie-in-the-sky philosophy and just use the very convenient cli.