[mpich-discuss] Can't boot mpd anymore after cluster reboot

Thomas Ruedas ruedas at dtm.ciw.edu
Wed Jan 27 12:12:04 CST 2010


Hi,
I wonder why I can't boot mpd's on the cluster anymore. If I try to do 
it the way (I think) I used to do it, by executing on the head node:
mpdboot --totalnum=43 --file=machines2

I get this error:
mpdboot_head.gl.ciw.edu (handle_mpd_output 415): failed to connect to 
mpd on head.gl.ciw.edu

but I can execute a single mpd just as
mpd &
on the head node. When I do that and try afterwards:
mpdboot --totalnum=43 --file=machines2

I get this error:
mpdboot_head.gl.ciw.edu (handle_mpd_output 406): failed to handshake 
with mpd on head.gl.ciw.edu; recvd output={}

and all mpds die. What is wrong here?
Thomas
-- 
-----------------------------------
Thomas Ruedas
Department of Terrestrial Magnetism
Carnegie Institution of Washington
http://www.dtm.ciw.edu/users/ruedas/


More information about the mpich-discuss mailing list