Provided by: sinfo_0.0.48-2.1_amd64 bug


       sinfod - a monitoring tool for networked computers


       sinfod  [-F]  [-q]  [-V]  [-b #.#.#.#] [-c] [-n card ] [ -N nicevalue ] [ -m marker ] [ -i
       processname ] [-s][-W]


       sinfod distributes system information to other sinfods on your local network.

       These Information include the system load, memory utilization and the `TOP-5' processes on
       every machine.

       It's main advantage is, that these informations are broadcasted using the udp(7) protocol.

       If  each  node  broadcasts  it's  information in a cooperative manner, the network load is
       O(N), where N is the number of nodes in your  network.   Other  systems  to  monitor  your
       cluster  load (e.g.  rup(1) ) are using a polling scheme where every node has to ask every
       other node for the system information: In that case the network load is O(N**2).

       The task of sinfod is to broadcast the relevant information for  the  local  node  and  to
       collect   broadcasts from other nodes.  Other programs ( sinfo(1) ) can get access to this
       information via a simple tcp(7) connection that  normally  is  routed  through  the  local
       loopback interface.


              Unlike  normal  operation, sinfod will not detach from the terminal when given this
              option.  Some debugging informations will be visible.

              Start quietly - for init scripts.

              Print the version number and exit.

       -b #.#.#.#/--bcastaddress=#.#.#.#
              Set the broadcast address of the sinfod.  Default is  In you  want
              to run sinfod without a network card "-b" may be necessary.

              Broadcast the full command line of a process instead of the command only.

       -m marker/--marker=marker
              Mark the machine with a name. e.g. laboratory, pool or cluster.

       -i/--ignore <process>
              Don't broadcast any information on <process> ; --ignore may be set multiple times.

       -t/--top <count>
              Broadcast information on the top <count> processes; default: 5

       -n card/--networkcard=card
              Set  the  name  of  the  network card whose informations will be broadcasted. (e.g.
              eth0).  Default is the network card that saw the  most  traffic  at  the  start  of

       -N nicevalue/--nice=nicevalue
              Set the priority of this demon to nicevalue. Default is 15.

              Deactivate  the  TCP  interface  and the collection of received packets.  The demon
              starts in a broadcast-only mode.

              Start with simple watchdog.


       Juergen Rinas <>