From: Gengbin Zheng (gzheng_at_ks.uiuc.edu)
Date: Thu May 05 2005 - 11:00:22 CDT
Don't use "localhost" in your nodelist file. Localhost can mean any
machine, it confuses both charmrun and the other node "foreignhost".
They both resolve localhost as themselves, and the outgoing messages can
be directed to themselves incorrectly. Use their public name, or put IP
address in nodelist.
Gengbin
Dong Luo wrote:
>Hello,
>I am trying to run namd parallely on two old alpha(dec
>osf1)workstations through network. When I launch namd
>from one machine, everything works fine until:
>Info: Entering startup phase 0 with 13292 kB of memory
>in use.
>I am using ++verbose, so the screen is like this:
>prompt>charmrun $NAMD/namd2 +p2 my.conf > my.log
>++verbose
>Charmrun> charmrun started...
>Charmrun> using ./nodelist as nodesfile
>Charmrun> rsh (localhost:0d) started
>Charmrun> rsh (foreignhost:1d) started
>Charmrun> node programs all started
>Charmrun> node programs all connected
>It then hangs here, by chech the log file, I find it
>hangs at startup phase o.
>
>The two machines can ping themselves, so it's not the
>same problem mentioned in the old post.
>
>Any idea about what's going on? or a way I could get
>more information about what it's trying to do at these
>points?
>
>Thank you
>sweec
>
>Nodelist:
>group main ++pathfix /usr/users/sweec/namd $WORKPATH
>host localhost
>host foreignhost
>
>Here I use ++pathfix cause foreignhost has a different
>path for the current directory /usr/users/sweec/namd
>like /usr_1/sweec/namd. In order to solve the problem,
>I defined WORKPATH in both machines to corresponding
>paths, otherwise there is a error message says No such
>file or directory.
>
>
>__________________________________________________
>Do You Yahoo!?
>Tired of spam? Yahoo! Mail has the best spam protection around
>http://mail.yahoo.com
>
>
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:39:24 CST