[ceph-users] rados client connection to cluster timeout and debugging.

Hi all, I continue to work on my rados client, implementing error handling and trying to reconnect to cluster if it's not available. Here is my tests. Client is not connected to network, so cluster is not available, i configured client_mount_timeout = 0.5 (which is 5 seconds) and getting reasonable error - 110, Connection timed out. Then client is doing second and third attempt and in these cases error is a bit strange - 115, Operation now in progress. So failed rados_connect didn't restore state of connection - bug ? I was trying to enable debug:   "log file" "./client.log"   "log to file" "true"   "debug ms" "10/10"   "debug rados" "10/10"   "debug objecter" "10/10"   "debug monc" "10/10"   "debug client" "10/10" But no luck to get anything in the log file. Running  ceph version 14.2.8 (2d095e947a02261ce61424021bb43bd3022d35cb) nautilus (stable) under archlinux. Help is more then welcome. Daniel Mezentsev, founder (+1) 604 313 8592. Soleks Data Group. Shaping the clouds.