Connecting a brand new slave mgm to an existing master


I reinstalled the slave manager on our production EOS cluster. The service eos@mgm refuses to start because it cannot boot the namespace. Is is true that the files (this is the in-memory namespace with files
for persistence) are not here, I was thinking that they would be automatically synchronized from the master MGM…

Log extract:

190404 11:05:29 time=1554368729.468731 func=BootNamespace            level=NOTE  logid=cb1c6e0a-56b8-11e9-afb1-14187764b113 tid=00007f17311948c0 source=Master:1890                    tident=<service> sec=      uid=0 gid=0 name= geo="" eos directory view configure started as slave
190404 11:05:29 time=1554368729.468989 func=BootNamespace            level=CRIT  logid=cb1c6e0a-56b8-11e9-afb1-14187764b113 tid=00007f17311948c0 source=Master:1946                    tident=<service> sec=      uid=0 gid=0 name= geo="" eos view initialization failed after 0 seconds
190404 11:05:29 time=1554368729.469017 func=BootNamespace            level=CRIT  logid=cb1c6e0a-56b8-11e9-afb1-14187764b113 tid=00007f17311948c0 source=Master:1949                    tident=<service> sec=      uid=0 gid=0 name= geo="" initialization returned ec=14 File does not exist and Create flag is absent: /var/eos/md/
190404 11:05:29 time=1554368729.469045 func=Configure                level=CRIT  logid=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx tid=00007f17311948c0 source=XrdMgmOfsConfigure:1475        tident=<single-exec> sec=      uid=0 gid=0 name= geo="" msg="namespace boot failed"
190404 11:05:29 25942 XrootdConfig: Unable to create file system object via
190404 11:05:29 25942 XrootdConfig: Unable to load file system.
------ xrootd protocol initialization failed.
190404 11:05:29 25942 XrdProtocol: Protocol xrootd could not be loaded
------ xrootd initialization failed.

=> What should I do ?

Thank you


I have to apologize: the origin of my difficulties synchronizing files between the master and the new slave was a wrong /etc/eos.keytab.

Sorry, really.