hello, I found a very very very serious bug in recovery, I running a testing cluster with 4 nodes, 3 nodes are running as normal mode, 1 node is running as gateway mode, if I shutdown the 3 normal nodes, but keep the gateway running, then I restart the 3 normal nodes, then , the 3 nodes need to recovery, when recovery is completed, all data will be wipe , I check the sheepdog store dir, there are all empty. if I have no gateway node in the cluster, this issue not Occur. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.wpkg.org/pipermail/sheepdog-users/attachments/20140606/015f6c1d/attachment.html> |