How to setup RSHD and start it for use with MPICH

Hi,

If I have posted this message on the wrong newsgroup, please let me know
which one
I should use.

I need to have "rshd" (Remore Shell Daemon) running in order to be able to
use
MPICH (Message Passing Interface) for parallel computing. I checked the file
"/etc/inetd.conf" and the lines for the BSD protocols are there i.e.

shell stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rshd
login stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rlogind

and I also checked to make that the rsh (Remote Shell) software was
installed.

So the question is: how to start the "rshd" ?

On my Linux system the following daemon NOT activated:

BSD remote user info daemon
BSD remote who info daemon

Network Information Service (client part)
Network Information Service (server part)

Should any of these be activated ?

I admit that the above shows my lack of knowledge and understanding of the
role of "rshd"
and any help in becoming more knowledgeable about this would be much
appreciated.

Additional information:

OS: Caldera OpenLinux eDesktop 2.4
CPU: Pentium 166MHz
Memory: 80 MB RAM

(Yes, it is old but besides the "rshd" problem, it does the job )

Thanks in advance !
--
René Girard

[Followup set to COLN]

In comp.os.linux.networking Rene Girard <aa@bb.yy> wrote:
> Hi,
>
> If I have posted this message on the wrong newsgroup, please let me know
> which one
> I should use.
>
> I need to have "rshd" (Remore Shell Daemon) running in order to be able to
> use
> MPICH (Message Passing Interface) for parallel computing. I checked the file
> "/etc/inetd.conf" and the lines for the BSD protocols are there i.e.
>
> shell stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rshd
> login stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rlogind
>
> and I also checked to make that the rsh (Remote Shell) software was
> installed.

Did you HUP inetd? (killall -HUP inetd). After you do that, you can
verify that something is listening on that port with the command
"netstat -at", and looking for a LISTENING entry for the ports shell and
login respectively.

Do you have anything in /etc/hosts.deny and /etc/hosts.allow? You can
use the tcpdchk program to verify that its a given IP can access a given
service. The service name you need to use in hosts.{allow,deny} will be
"in.rshd", or possibly "shell".

You also need appropriate entries in /etc/hosts.equiv for rsh to let
people in. Here is an example of what you need

csatm4.localeth cameron
csatm4atm.localatm cameron
csatm5.localeth cameron
csatm5atm.localatm cameron
csatm6.localeth cameron
csatm6atm.localatm cameron
csatm7.localeth cameron
csatm7atm.localatm cameron
csatm8.localeth cameron
csatm8atm.localatm cameron
localhost cameron
-

You'll need to repeat all the lines above (excppt "-") for every user.
Or at least, this is what I had to do when I was doing that sort of
thing. I'd much rather drop rsh completely though and use SSH instead
(and under Debian Woody at least, if the rsh package isn't installed,
rsh is set up to be ssh)

> I admit that the above shows my lack of knowledge and understanding of
> the role of "rshd" and any help in becoming more knowledgeable about
> this would be much appreciated.

RSH is dangerous and should *ONLY* *EVER* be used in tightly controlled
situations, and *NEVER* accessable via untrusted networks. Consider the
use of the Berkeley R-commands to be highly deprecated.

> Additional information:
>
> OS: Caldera OpenLinux eDesktop 2.4
> (Yes, it is old but besides the "rshd" problem, it does the job )

Extremely old, that was the very first distribution I used when I was
learning Linux all those many years ago, so its over ummm, 6? years old
by now. I hope for your sake, its not (at least directly) accessable via
the internet.

--
Cameron Kerr
cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
Empowered by Perl!

Hi.

90`s called, they want their distro back. :-)
It will be a lot of headache for you to keep such an old distro
functioning. Libraries, etc will be very hard to find and to compile.

Consider installing Debian which is very light weight in default install
and will run on very old hardware. 166Mhz is plenty to run Debian sans X.

I would not use rsh - someone will hack your box really quick. Use ssh
instead. Setup public/private key authentication, share home directory
over NFS, and you are ready to go.

Kerr,

Many thanks for your informative reply. I have tried what you have
proposed but
unfortunately I still have something wrong in my settings and also in my
understanding.

I would like to take few lines to explain better the problem I am trying to
solve.
I am trying to run MPICH on my Linux machine which has only one processor.
The process
that installs MPICH creates at least 5 virtual machines. Therefore as far as
I understand
when I run a MPI program (for example t_mpi1.c) with the command:

mpirun -np 1 ./t_mpi1 (-np is for the number of processors)

I should get the following answer:

First MPI program
[t_mpi1]: I am 0 of 1
Success !: First MPI program

meaning I am processor no. "0" of one processor

however, I get

First MPI program
[t_mpi1]: I am 0 of 0 <--- Note the "0" instead of the one.
Success !: First MPI program.

The MPICH indicates that the daemon of rsh must be running such that when
one does :

mpirun -np 4 ./t_mpi1

the virtual machines come into play and I should get:

First MPI program
[t_mpi1]: I am 0 of 4
[t_mpi1]: I am 1 of 4
[t_mpi1]: I am 2 of 4
[t_mpi1]: I am 3 of 4
Success !: First MPI program.

Below you will see the content of several files that you indicated in you
reply.
With the settings in these files, I was able to do a "rsh localhost" but
even with that
the

mpirun -np 4 ./t_mpi1

command gave me the same answer i.e.

First MPI program
[t_mpi1]: I am 0 of 0
Success !: First MPI program.

Also I did as you indicated :

"kill -HUP inetd"

followed by a

"netstat -at"

and a

"tcpdchk -v"

The results of the "netstat -at" gave the following


Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 *:6000 *:* LISTEN
tcp 0 0 *:1024 *:* LISTEN
tcp 0 0 *:920 *:* LISTEN
tcp 0 0 *:919 *:* LISTEN
tcp 0 0 *:smtp *:* LISTEN
tcp 0 0 *:printer *:* LISTEN
tcp 0 0 *:swat *:* LISTEN
tcp 0 0 *:auth *:* LISTEN
tcp 0 0 *:finger *:* LISTEN
tcp 0 0 *:uucp *:* LISTEN
tcp 0 0 *:imap2 *:* LISTEN
tcp 0 0 *:pop3 *:* LISTEN
tcp 0 0 *:pop2 *:* LISTEN
tcp 0 0 *:exec *:* LISTEN
tcp 0 0 *:login *:* LISTEN
tcp 0 0 *:shell *:* LISTEN
tcp 0 0 *:telnet *:* LISTEN
tcp 0 0 *:ftp *:* LISTEN
tcp 0 0 *:time *:* LISTEN
tcp 0 0 *:daytime *:* LISTEN
tcp 0 0 *:discard *:* LISTEN
tcp 0 0 *:sunrpc *:* LISTEN

As you can see "login,shell, telnet and ftp" are listening

The results for "tcpdchk -v" are:

Using network configuration file: /etc/inetd.conf

>>> Rule /etc/hosts.allow line 6:
daemons: in.rshd
clients: localhost rene.renelinux.net
option: allow
access: granted

>>> Rule /etc/hosts.deny line 9:
daemons: swat
clients: ALL EXCEPT 127.0.0.1
access: denied


At this point I do not know what I need to do next. Any help would
be much appreciated.

Regards


Rene


Here is the content of files "/etc/inetd.conf", "/etc/hosts",
"/etc/hosts.allow", "/etc/hosts.deny" and "/etc/hosts.equiv".
I am sorry if this makes the message long.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Content of file "/etc/inetd.conf":


#
# inetd.conf This file describes the services that will be available
# through the INETD TCP/IP super server. To re-configure
# the running INETD process, edit this file, then send the
# INETD process a SIGHUP signal.
#
# Version: @(#)/etc/inetd.conf 3.10 05/27/93
#
# Authors: Original taken from BSD UNIX 4.3/TAHOE.
# Fred N. van Kempen, <waltje@uwalt.nl.mugnet.org>
#
# Modified for Debian Linux by Ian A. Murdock <imurdock@shell.portal.com>
#
# Modified for RHS Linux by Marc Ewing <marc@redhat.com>
#
# Further modified by Olaf Kirch <okir@caldera.com> for Caldera Open Linux
#
# <service_name> <sock_type> <proto> <flags> <user> <server_path> <args>
#
# Echo, discard, daytime, and chargen are used primarily for testing.
#
# To re-read this file after changes, just do a 'killall -HUP inetd'
#
# Note: builtin UDP services now silently drop packets from ports < 512.
#echo stream tcp nowait root internal
#echo dgram udp wait root internal
discard stream tcp nowait root internal
discard dgram udp wait root internal
daytime stream tcp nowait root internal
daytime dgram udp wait root internal
#chargen stream tcp nowait root internal
#chargen dgram udp wait root internal
time stream tcp nowait root internal
time dgram udp wait root internal
#
# These are standard services.
#
ftp stream tcp nowait root /usr/sbin/tcpd in.ftpd -l -a
telnet stream tcp nowait root /usr/sbin/tcpd in.telnetd

#
# Mail and news
#
# Do not uncomment either unless you *really* know what you are doing.
# Both are started as standalone daemons from the /etc/rc.d scripts.
#smtp stream tcp nowait root /usr/bin/smtpd smtpd
#nntp stream tcp nowait root /usr/sbin/tcpd in.nntpd

#
# Shell, login, exec and talk are BSD protocols.
#
shell stream tcp nowait root /usr/sbin/tcpd in.rshd
login stream tcp nowait root /usr/sbin/tcpd in.rlogind
exec stream tcp nowait root /usr/sbin/tcpd in.rexecd
talk dgram udp wait nobody.tty /usr/sbin/tcpd in.talkd
ntalk dgram udp wait nobody.tty /usr/sbin/tcpd in.ntalkd
#dtalk stream tcp wait nobody.tty /usr/sbin/tcpd in.dtalkd

#
# Pop and imap mail services et al
#
pop2 stream tcp nowait root /usr/sbin/tcpd ipop2d
pop3 stream tcp nowait root /usr/sbin/tcpd ipop3d
imap stream tcp nowait root /usr/sbin/tcpd imapd
#
# The Internet UUCP service.
#
uucp stream tcp nowait uucp /usr/sbin/tcpd /usr/sbin/uucico -l
#
# Tftp service is provided primarily for booting. Most sites
# run this only on machines acting as "boot servers." Do not uncomment
# this unless you *need* it.
#
#tftp dgram udp wait root /usr/sbin/tcpd in.tftpd
#bootps dgram udp wait root /usr/sbin/tcpd bootpd
#
# This is for the finger service
#
finger stream tcp nowait nobody /usr/sbin/tcpd in.fingerd -u
/var/run/.ppp_socket stream unix nowait root /usr/sbin/ppp-envoy
ppp-envoy -da
#
# Finger, systat and netstat give out user information which may be
# valuable to potential "system crackers." Many sites choose to disable
# some or all of these services to improve security.
#
#systat stream tcp nowait nobody /usr/sbin/tcpd /bin/ps -auwwx
#netstat stream tcp nowait nobody /usr/sbin/tcpd /bin/netstat --inet
#
# Authentication
#
auth stream tcp nowait root /usr/sbin/in.identd in.identd
swat stream tcp nowait.400 root /usr/sbin/tcpd swat
#
# End of inetd.conf


- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Content of file "/etc/hosts"

# /etc/hosts - hostname_to_address mappings for the resolver library
#
# The format is:
#
# ipaddr fully_qualified_hostname list_of_nicknames
#
# We strongly recommend to use a fully qualified hostname for your machine
# if it is connected to any type of network. You can add short nicknames as
# you like, but the official canonical name is the name that follows the
# IP address and should always be a fully qualified hostname.
#
# If you don't have any network interfaces, you should put your hostname
# in the line of your loopback device, which has the address 127.0.0.1.
#
# LISA or COAS will automatically put the hostname in the right line for
you.
#
#
# !!! Automatically generated by lizard. Do not edit !!!
#
# The loopback device. It should always be available.
#
127.0.0.1 localhost
#
# Your first IP interface (if you are connected to any kind of network)
#
10.162.211.74 rene.renelinux.net .renelinux.net

#
# If you have a nameserver we recommend to use it instead of putting many
# hosts in this file. If you don't have a nameserver though, you can add
# hostnames and their IP addresses below this line.
#
#================================================= ==========================


- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Content of file "/etc/hosts.allow"


#
# hosts.allow This file describes the names of the hosts which are
# allowed to use the local INET services, as decided
# by the '/usr/sbin/tcpd' server.
#
in.rshd:localhost,rene.renelinux.net:allow



- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Content of file "/etc/hosts.deny"


#
# hosts.deny This file describes the names of the hosts which are
# *not* allowed to use the local INET services, as decided
# by the '/usr/sbin/tcpd' server.
#
# The portmap line is redundant, but it is left to remind you that
# the new secure portmap uses hosts.deny and hosts.allow. In particular
# you should know that NFS uses portmap!
swat:ALL EXCEPT 127.0.0.1


- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Content of file "/etc/hosts.equiv"


localhost rene
rene.renelinux.net rene

Note: "rene" here is my user name on my Linux machine




"Cameron Kerr" <cameron.kerr@paradise.net.nz> wrote in message
news:401c8594@news.maxnet.co.nz...
> [Followup set to COLN]
>
> In comp.os.linux.networking Rene Girard <aa@bb.yy> wrote:
> > Hi,
> >
> > If I have posted this message on the wrong newsgroup, please let me know
> > which one
> > I should use.
> >
> > I need to have "rshd" (Remore Shell Daemon) running in order to be able
to
> > use
> > MPICH (Message Passing Interface) for parallel computing. I checked the
file
> > "/etc/inetd.conf" and the lines for the BSD protocols are there i.e.
> >
> > shell stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rshd
> > login stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.rlogind
> >
> > and I also checked to make that the rsh (Remote Shell) software was
> > installed.
>
> Did you HUP inetd? (killall -HUP inetd). After you do that, you can
> verify that something is listening on that port with the command
> "netstat -at", and looking for a LISTENING entry for the ports shell and
> login respectively.
>
> Do you have anything in /etc/hosts.deny and /etc/hosts.allow? You can
> use the tcpdchk program to verify that its a given IP can access a given
> service. The service name you need to use in hosts.{allow,deny} will be
> "in.rshd", or possibly "shell".
>
> You also need appropriate entries in /etc/hosts.equiv for rsh to let
> people in. Here is an example of what you need
>
> csatm4.localeth cameron
> csatm4atm.localatm cameron
> csatm5.localeth cameron
> csatm5atm.localatm cameron
> csatm6.localeth cameron
> csatm6atm.localatm cameron
> csatm7.localeth cameron
> csatm7atm.localatm cameron
> csatm8.localeth cameron
> csatm8atm.localatm cameron
> localhost cameron
> -
>
> You'll need to repeat all the lines above (excppt "-") for every user.
> Or at least, this is what I had to do when I was doing that sort of
> thing. I'd much rather drop rsh completely though and use SSH instead
> (and under Debian Woody at least, if the rsh package isn't installed,
> rsh is set up to be ssh)
>
> > I admit that the above shows my lack of knowledge and understanding of
> > the role of "rshd" and any help in becoming more knowledgeable about
> > this would be much appreciated.
>
> RSH is dangerous and should *ONLY* *EVER* be used in tightly controlled
> situations, and *NEVER* accessable via untrusted networks. Consider the
> use of the Berkeley R-commands to be highly deprecated.
>
> > Additional information:
> >
> > OS: Caldera OpenLinux eDesktop 2.4
> > (Yes, it is old but besides the "rshd" problem, it does the job )
>
> Extremely old, that was the very first distribution I used when I was
> learning Linux all those many years ago, so its over ummm, 6? years old
> by now. I hope for your sake, its not (at least directly) accessable via
> the internet.
>
> --
> Cameron Kerr
> cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
> Empowered by Perl!

[Set Followup-To COLN]
In comp.os.linux.networking Rene Girard <aa@bb.yy> wrote:

> mpirun -np 1 ./t_mpi1 (-np is for the number of processors)
>
> I should get the following answer:
>
> First MPI program
> [t_mpi1]: I am 0 of 1
> Success !: First MPI program
>
> meaning I am processor no. "0" of one processor
>
> however, I get
>
> First MPI program
> [t_mpi1]: I am 0 of 0 <--- Note the "0" instead of the one.
> Success !: First MPI program.

That would indicate to me that it seems to think it's only configured
with 0 processes. I think rsh may not be your problem.

To verify, try the following.

Have a look in /var/log/syslog and /var/log/messages. There should be an
entry in there for every rsh connection that is accepted, and the
command that was run.


Now, what does your /etc/mpich/machines.LINUX file say? (Yours might not
be called exactly that). Mine says something like the following (Note,
I admin a network that some people use for cluster stuff, but I don't
really know much about MPICH.

csatm1
csatm2
csatm3
csatm4
csatm5
csatm6
csatm7
csatm8

For a total of eight processes.

> Content of file "/etc/hosts"
>
> # /etc/hosts - hostname_to_address mappings for the resolver library

> # LISA or COAS will automatically put the hostname in the right line
> # for you.
> # !!! Automatically generated by lizard. Do not edit !!!

> 10.162.211.74 rene.renelinux.net .renelinux.net

That doesn't look good, you should fix that. The last bit should just
say "rene".

> # hosts.deny This file describes the names of the hosts which are

> swat:ALL EXCEPT 127.0.0.1

You should normally have ALL:ALL in here, but you can fix that later.

--
Cameron Kerr
cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
Empowered by Perl!

Hi Cameron,

Many thanks for your informative reply and your patience with my questions.

I looked at the file /usr/local/mpich-1.2.5/share/machines.LINUX and there
are 5 lines
containing the following:

rene.renelinux.net

This is expected as far as I understand because these are virtual machines.

Now in file /etc/hosts, I corrected the line

"10.162.211.74 rene.renelinux.net .renelinux.net" to simply "10.162.211.74
rene.renelinux.net"

I also corrected the file /etc/hosts.deny and now it has

ALL:ALL as you indicated

I also corrected file /etc/hosts.equiv and now it has the following

rene.renelinux.net
rene.renelinux.net
rene.renelinux.net
rene.renelinux.net
rene.renelinux.net
localhost

I put "rene.renelinux.net" 5 times because the file
/usr/loac/mpich-1.2.5/share/machines.LINUX has
that line 5 times.

in the user account "rene", I tried to start a "rsh" shell with the command

rsh -d rene.renelinux.net

doing a "netstat -at" I got as the first line:

Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 1 192.168.1.102:1023 rene.renelinux.ne:login SYN_SENT

The documentation of netstat indicates that this means that a foreign
connection is being attemped (state = SYN_SENT)

Note here that 192.168.1.102 is the address given by the router to the
computer

So if I am not able to establish a connection with that local address
192.168.1.102 then there is no way
that MPICH will run properly and that my program would give
with the command:

mpirun -np 1 t_mpi1 (name of sample MPI program is t_mpi1.c)

process 0 of 1.

I tried doing "rsh localhost". It does login but doing "mpirun -np 1 t_mpi1"
gives the same answer

process 0 of 0.

Perharps it is the IP address given by the router "192.168.1.102" that is
causing the problem:

IP address of "rene.renelinux.net" is 10.162.211.74 and the Local IP Address
it is trying to login is
192.168.1.102

Beyond this I do not know what to do. Again, there are things I do not
understand about that rshd so
I will have to read to understand better. Any suggestion ?

Regards

Rene






"Cameron Kerr" <cameron.kerr@paradise.net.nz> wrote in message
news:401dcfc9@news.maxnet.co.nz...
> [Set Followup-To COLN]
> In comp.os.linux.networking Rene Girard <aa@bb.yy> wrote:
>
> > mpirun -np 1 ./t_mpi1 (-np is for the number of processors)
> >
> > I should get the following answer:
> >
> > First MPI program
> > [t_mpi1]: I am 0 of 1
> > Success !: First MPI program
> >
> > meaning I am processor no. "0" of one processor
> >
> > however, I get
> >
> > First MPI program
> > [t_mpi1]: I am 0 of 0 <--- Note the "0" instead of the one.
> > Success !: First MPI program.
>
> That would indicate to me that it seems to think it's only configured
> with 0 processes. I think rsh may not be your problem.
>
> To verify, try the following.
>
> Have a look in /var/log/syslog and /var/log/messages. There should be an
> entry in there for every rsh connection that is accepted, and the
> command that was run.
>
>
> Now, what does your /etc/mpich/machines.LINUX file say? (Yours might not
> be called exactly that). Mine says something like the following (Note,
> I admin a network that some people use for cluster stuff, but I don't
> really know much about MPICH.
>
> csatm1
> csatm2
> csatm3
> csatm4
> csatm5
> csatm6
> csatm7
> csatm8
>
> For a total of eight processes.
>
> > Content of file "/etc/hosts"
> >
> > # /etc/hosts - hostname_to_address mappings for the resolver library
>
> > # LISA or COAS will automatically put the hostname in the right line
> > # for you.
> > # !!! Automatically generated by lizard. Do not edit !!!
>
> > 10.162.211.74 rene.renelinux.net .renelinux.net
>
> That doesn't look good, you should fix that. The last bit should just
> say "rene".
>
> > # hosts.deny This file describes the names of the hosts which are
>
> > swat:ALL EXCEPT 127.0.0.1
>
> You should normally have ALL:ALL in here, but you can fix that later.
>
> --
> Cameron Kerr
> cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
> Empowered by Perl!

Rene Girard wrote:
-> Hi,
->
-> If I have posted this message on the wrong newsgroup, please let me know
-> which one
-> I should use.
->
-> I need to have "rshd" (Remore Shell Daemon) running in order to be
able to
-> use
-> MPICH (Message Passing Interface) for parallel computing. I checked
the file
-> "/etc/inetd.conf" and the lines for the BSD protocols are there i.e.
-> <snip>


You would be better off recompiling mpich with the configuration option
-rsh=/usr/bin/ssh and not messing with rsh. You will take a bit of a
performance hit but the cost in configuring an extra insecure service
was too much for me.

--
Dr. Douglas O'Neal
Manager, Bioinformatics Center
Delaware Biotechnology Institute
(302) 831-3456

Hi,

Many thanks for your help, I was able to solve the problem. The problem
was in the /etc/hosts file with the line:

#
# Your first IP interface (if you are connected to any kind of network)
#
10.162.211.74 rene.renelinux.net

The IP address was the permanent IP address the FORMER Internet provider
gave me at that time. With the NEW internet
provider I am using now that IP address is no longer valid. So when I was
trying to login to "rene.renelinux.net" by doing "rsh rene.renelinux.net", I
saw doing "netstat -at" that the foreign address "rene.renelinux.net" was
trying to log on the local IP address that is given by my Lynksys Router.
The fact that in file /etc/hosts.allow I had the line

in.rshd: localhost, rene.renelinux.net

the "rene.renelinux.net" corresponds to the old IP address 10.162.211.74
which did not correspond to the local IP address
given by the Lynksys Router. This mismatch of the IP addresses made it
impossible to logon my own machine "rene.renelinux.net".

I was able to find that out because I tried to get the YAMI message passing
interface library going and it was not working. Using the "netstat -at"
command and also looking at the comment above the line "10.162.211.74
rene.renelinux.net " which say
in fact "put the IP address that corresponds to your network interface"
being in this case the Lynksys Router which connect
the two PCs (P4/ 2.4 GHz and AMD 2500XP Barton) of respectively my son and
daughter and my (old, Pentium 166 Mhz
I am old so I get to use the old stuff!!) PC to the internet. So by doing
that change, MPICH appears to work
fine (I still need to check the results it gave me with a professor at
McGill Univ. in Montreal) and YAMI works just fine.

Thanks again for your help and patience.

Regards

Rene Girard






"Cameron Kerr" <cameron.kerr@paradise.net.nz> wrote in message
news:401dcfc9@news.maxnet.co.nz...
> [Set Followup-To COLN]
> In comp.os.linux.networking Rene Girard <aa@bb.yy> wrote:
>
> > mpirun -np 1 ./t_mpi1 (-np is for the number of processors)
> >
> > I should get the following answer:
> >
> > First MPI program
> > [t_mpi1]: I am 0 of 1
> > Success !: First MPI program
> >
> > meaning I am processor no. "0" of one processor
> >
> > however, I get
> >
> > First MPI program
> > [t_mpi1]: I am 0 of 0 <--- Note the "0" instead of the one.
> > Success !: First MPI program.
>
> That would indicate to me that it seems to think it's only configured
> with 0 processes. I think rsh may not be your problem.
>
> To verify, try the following.
>
> Have a look in /var/log/syslog and /var/log/messages. There should be an
> entry in there for every rsh connection that is accepted, and the
> command that was run.
>
>
> Now, what does your /etc/mpich/machines.LINUX file say? (Yours might not
> be called exactly that). Mine says something like the following (Note,
> I admin a network that some people use for cluster stuff, but I don't
> really know much about MPICH.
>
> csatm1
> csatm2
> csatm3
> csatm4
> csatm5
> csatm6
> csatm7
> csatm8
>
> For a total of eight processes.
>
> > Content of file "/etc/hosts"
> >
> > # /etc/hosts - hostname_to_address mappings for the resolver library
>
> > # LISA or COAS will automatically put the hostname in the right line
> > # for you.
> > # !!! Automatically generated by lizard. Do not edit !!!
>
> > 10.162.211.74 rene.renelinux.net .renelinux.net
>
> That doesn't look good, you should fix that. The last bit should just
> say "rene".
>
> > # hosts.deny This file describes the names of the hosts which are
>
> > swat:ALL EXCEPT 127.0.0.1
>
> You should normally have ALL:ALL in here, but you can fix that later.
>
> --
> Cameron Kerr
> cameron.kerr@paradise.net.nz : http://nzgeeks.org/cameron/
> Empowered by Perl!