f



connect() failed: Connection refused (111)

Hello,

i try to build an MPI-Cluster with Ubuntu-9-04.
The helle exampel works fine but when i use MPI_Bcast or MPI_Reduce i
get these error [1].

How can i solve this?
SSH works fine in both directions.

Regards,
Stefan Kuhne

-- 
[1]:
[Head][[4950,1],0][../../../../../../ompi/mca/btl/tcp/btl_tcp_endpoint.c:631:mca_btl_tcp_endpoint_complete_connect]
connect() failed: Connection refused (111)
0
SK-Privat (6)
9/18/2009 3:09:04 PM
comp.parallel.mpi 1534 articles. 0 followers. kisitanggang (69) is leader. Post Follow

3 Replies
6126 Views

Similar Articles

[PageSpeed] 25

Stefan Kuhne schrieb:
> Hello,
> 
> i try to build an MPI-Cluster with Ubuntu-9-04.
> The helle exampel works fine but when i use MPI_Bcast or MPI_Reduce i
> get these error [1].
> 
> How can i solve this?
> SSH works fine in both directions.
> 
My Nodes are in virtualBox under Windows Vista, i think Vista is my problem.
Anyone an idea?

Regards,
Stefan Kuhne
2
Stefan
9/21/2009 8:29:07 AM
Stefan Kuhne schrieb:
> Stefan Kuhne schrieb:
>> Hello,
>>
>> i try to build an MPI-Cluster with Ubuntu-9-04.
>> The helle exampel works fine but when i use MPI_Bcast or MPI_Reduce i
>> get these error [1].
>>
>> How can i solve this?
>
It is solved.
Multi Network is my Problem.

"--mca btl_tcp_if_exclude lo,eth0" has solved it.

Stefan Kuhne
0
Stefan
9/21/2009 8:38:56 AM
I know there will be some one who can help me in solving this problem.

I can compile my helloworld.c program using mpicc and I have confirmed that the script runs correctly on another working cluster, so the local paths are set up correctly I think and the script definitely works.

If I execute mpirun from my master node, and using only the master node, helloworld executes correctly:

mpirun -n 1 -host master --mca btl sm,openib,self ./helloworldmpi
hello world from process 0 of 1
If I execute mpirun from my master node, using only the worker node, helloworld executes correctly:

mpirun -n 1 -host node001 --mca btl sm,openib,self./helloworldmpi
hello world from process 0 of 1
Now, my problem is that if I try to run helloworld on both nodes, I get an error:

mpirun -n 2 -host master,node001 --mca btl openib,self ./helloworldmpi
--------------------------------------------------------------------------
At least one pair of MPI processes are unable to reach each other for
MPI communications.  This means that no Open MPI device has indicated
that it can be used to communicate between these processes.  This is
an error; Open MPI requires that all MPI processes be able to reach
each other.  This error can sometimes be the result of forgetting to
specify the "self" BTL.

  Process 1 ([[5228,1],0]) is on host: hsaeed
  Process 2 ([[5228,1],1]) is on host: node001
  BTLs attempted: self

Your MPI job is now going to abort; sorry.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
It looks like MPI_INIT failed for some reason; your parallel process is
likely to abort.  There are many reasons that a parallel process can
fail during MPI_INIT; some of which are due to configuration or environment
problems.  This failure appears to be an internal failure; here's some
additional information (which may only be relevant to an Open MPI
developer):

  PML add procs failed
  --> Returned "Unreachable" (-12) instead of "Success" (0)
--------------------------------------------------------------------------
*** The MPI_Init() function was called before MPI_INIT was invoked.
*** This is disallowed by the MPI standard.
*** Your MPI job will now abort.
Abort before MPI_INIT completed successfully; not able to guarantee that all other processes were killed!
--------------------------------------------------------------------------
mpirun has exited due to process rank 0 with PID 7037 on
node xxxx exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
*** The MPI_Init() function was called before MPI_INIT was invoked.
*** This is disallowed by the MPI standard.
*** Your MPI job will now abort.
Abort before MPI_INIT completed successfully; not able to guarantee that all other processes were killed!
1 more process has sent help message help-mca-bml-r2.txt / unreachable proc
Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
1 more process has sent help message help-mpi-runtime

i tried using
mpirun -n 2 -host master,node001 --mca btl tcp,sm,self ./helloworldmpi
mpirun -n 2 -host master,node001 --mca btl o
penib,tcp,self ./helloworldmpi
etc..

But no flag is works.


Can some one reply with the idea.

Thanks in Advance.

Regards--
hamid
0
e
3/21/2014 8:31:12 AM
Reply:

Similar Artilces:

open failed: connect failed: Connection refused
Hi all - I am trying to connect to a computer behind a firewall by being tricky with ssh. I want to start an ssh session from behind the firewall to a computer on the outside, and then run an xterm on the firewalled computer that is displayed on the outside computer, like I show below. Then I can go to the outside computer and have shell access (and thus everything else) to the firewalled computer. {FC = Firewalled computer, OC = Outside computer} [OC shell 1]$ xhost + # for now [FC shell 1]$ ssh -Y -N -L 6007:localhost:6000 OC [FC shell 2]$ export DISPLAY=localhost:7.0 [FC shell 2]$ xterm...

channel 3: open failed: connect failed: Connection refused
I have done a lot of googling and have not been able to figure out what the above error message means. My specific situation is as follows: Case 1: ssh from linux box to linux box Screen1: Client> ssh Server -L 9546:127.0.0.1:9546 Screen2: Client> telnet 127.0.0.1 9546 Then on Screen1 (which is now ssh'd into the Server, I see the following message: channel 3: open failed: connect failed: Connection refused There is nothing actually running on port 9546 of the Server. I just use telnet as a means of sending packets ...

rndc connect failed connection refused
Hello, I'm sorry, I read a lot of documentation and I attempt to configure my=20 rndc.conf and rndc.key but always I have : connection refused ! I use command dnssec-keygen : dnssec-keygen -a hmac-md5 -b 256 -n HOST aforsiori.key It generate to files : Kaforsiori.key.+157+37451.key and Kaforsiori.key.+157+37451.private and what ?? the first file content : aforsiori.key. IN KEY 512 3 157 1XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX=3D the second : Private-key-format: v1.2 Algorithm: 157 (HMAC_MD5) Key: 1XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX=3D so I put in named.conf : algorithm hmac-md5; secret "1XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX=3D"; }; controls { inet 127.0.0.1 allow { 127.0.0.1; } keys { /etc/bind/rndc.key; = }; }; and in rndc.conf : options { default-server localhost; default-key "aforsiori.key"; }; server localhost { key "aforsiori.key"; }; Include "/etc/bind/rndc.key"; and finaly, in rndc.key : key "aforsiori.key" { algorithm hmac-md5; secret "1XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX=3D"; }; What can I do... I don't understand how I use the key files ! help me ! ...

rndc: connect failed: connection refused
Hi all I' am use Bind 9.2.2 I want to have my one Namessever ns1.helpme.ch *********************************************** when I do a command like: root@linux:/etc# rndc reload I become this fault message. rndc: connect failed: connection refused ********************************************** what is wrong hier is my rndc.conf: # Start of rndc.conf key "rndc-key" { algorithm hmac-md5; secret "+vDJ4dRcQ+w1iGP/oLNm2Q=="; }; options { default-key "rndc-key"; default-server 192.168.1.247; default-port 953; ...

Could Not Connect To SMTP Host Connection Failed -- Software caused connection abort: connect
Hi All, Great to have this discussion group which is helping to clear the issues. Thanks in advance to all. I am having strange exception when trying to send email through java. I am using my mail server hostname to connect it. My mail server is SMTP server. When I tried to ping the IP Address it is fine. I tried to execute the same in another machine. The result is same. The exception I am getting is this way. javax.mail.MessagingException: Could not connect to SMTP host: <hostname>, port: 25; nested exception is: java.net.SocketException: Software caused connection abort: conne...

Re: rndc: connect failed: connection refused
> Hi all > I' am use Bind 9.2.2 > I want to have my one Namessever ns1.helpme.ch > *********************************************** > when I do a command like: > root@linux:/etc# rndc reload > I become this fault message. > rndc: connect failed: connection refused > ********************************************** > what is wrong hier is my rndc.conf: > > # Start of rndc.conf > key "rndc-key" { > algorithm hmac-md5; > secret "+vDJ4dRcQ+w1iGP/oLNm2Q=="; > }; > > options { > d...

Connect fails: Connection Refuse on HP-UX.
Hi all, I have a server running on HP-UX machine. When I try to connect a HP-UX client to the server using TCP/IP, it fails. The reason is connection refused. What is the possible cause? Fyi, my server has yet to receive any connections. -HS Hon Seng Phuah wrote: > Hi all, > > I have a server running on HP-UX machine. When I try to connect a > HP-UX client to the server using TCP/IP, it fails. The reason is > connection refused. > > What is the possible cause? Fyi, my server has yet to receive any > connections. Wrong port number, wrong IP adress or sg else... fi...

rndc: connect failed: connection refused #4
Hi If this isn't the right list for this I appologise now To restart named etc I've been using the command line /etc/rc.d/init.d/named restart but recently this has stopped working, ie restart will stop named but doesn't restart it again when I attempt to start named using rndc it returns the error 'rndc: connect failed: connection refused' the key in both named.conf and rndc.conf are the same, in my named.conf I have the control section controls { inet 127.0.0.1 allow { 127.0.0.1; } keys { "dnsadmin" ; }; }; and the options section in the rndc section is options { default-key "dnsadmin"; default-server 127.0.0.1; }; Any idea's why rndc is failing to connect? Thanks Greg In article <cktpb4$2q86$1@sf1.isc.org>, Greg Mathews <greg@hostingtheinter.net> wrote: > Hi > > If this isn't the right list for this I appologise now > > To restart named etc I've been using the command line > > /etc/rc.d/init.d/named restart Try: rndc exec > > but recently this has stopped working, ie restart will stop named but > doesn't restart it again > > when I attempt to start named using rndc it returns the error > > 'rndc: connect failed: connection refused' Rndc works by sending a control message to named. If named isn't running, there's nothing for it to connect to in o...

rndc: connect failed: connection refused #3
I have installed 9.2.2 on Mac OS X 10.3, and seem to be unable to start bind, or connect with rndc. I have checked severaltutorials and messages posted here, but seem to be missing something (??). Any help would be grately appreciated. thanks in advane, Ralph top half of named.conf = include "/private/etc/rndc.key"; controls { inet 127.0.0.1 allow { localhost;} keys { rndc-key; }; }; options { directory "/var/named"; version "Bind"; //Forwarders send unresolved requests to my ISP forwarders { 206.12...

rndc: connect failed: connection refused #2
Hi all I' am use Bind 9.2.2 I want to have my one Namessever ns1.helpme.ch *********************************************** when I do a command like: root@linux:/etc# rndc reload I become this fault message. rndc: connect failed: connection refused ********************************************** what is wrong hier is my rndc.conf: # Start of rndc.conf key "rndc-key" { algorithm hmac-md5; secret "+vDJ4dRcQ+w1iGP/oLNm2Q=="; }; options { default-key "rndc-key"; default-server 192.168.1.247; default-port 953; ...

Re: rndc: connect failed: connection refused #6
Thank You all:) Steve, I have for some reason two named binaries, one /usr/sbin/named and the other /usr/local/sbin/named which named returns the /usr/local path By adding the path of named to the startup script has solved the problem Thanks again Greg At 19:44 17/10/2004, Steve Sandau wrote: ><successful startup sniped> > >> > >>if I then restart using the command > >> > >>/etc/rc.d/init.d/named restart > >> > >>I just get > >> > >>Oct 17 18:21:31 dedi5 named[6183]: starting BIND 9.2.1 -u named > >>Oct 17 18:21:31 dedi5 named[6183]: using 1 CPU > >>Oct 17 18:21:31 dedi5 named: named startup succeeded > >> > >>If I then do > >> > >>ps aux|grep named > >> > >>it returns no running processes and named is not running > >> > >>I note that it doesn't appear to be loading the conf file when using the > >>startup script > > > > > > -d debug-level Set the daemons debug level to debug-level. > > Debugging traces from named become more verbose > as the > > debug level increases. > > -f Run the server in the foreground (i.e. do not daemonize). > > -g Run the server in the foreground and force all logging > > to stder...

Re: rndc: connect failed: connection refused #5
> Date: Sun, 17 Oct 2004 17:31:04 +0100 > To: Barry Margolin <barmar@alum.mit.edu>, comp-protocols-dns-bind@isc.org > From: Greg Mathews <greg@hostingtheinter.net> > Subject: Re: rndc: connect failed: connection refused > > OK thanks for the info Barry and Jim > > It appears that my problem is not with rndc as that does work when named is > running, my problem is actually with the /etc/rc.d/init.d/named which is > not starting named although it tells me it has. I have been restarting What do your system messages tell you? Why is named NOT starting? I believe that you may have a problem with the named config since it has been MY experience that named always starts when properly configured... The start up scripts DO work... This includes the "restart" option to /etc/init.d/named. All THAT does it do /etc/init.d/named stop /etc/init.d/named start (Internally to the script of course.) [... beginning of script and switch statement ...] restart() { stop start } [...end of script...] Regards, Gregory Hicks > named when adding/editing zones but I now realise that I should just be > reloading it. > > This solves my problem except when my box is rebooted, as it was by my host > last night for network maintenance and named isn't starting at boot up but > I can monitor that another way and it doesn't happen that often, last night > being th...

Re: rndc: connect failed: connection refused #4
OK thanks for the info Barry and Jim It appears that my problem is not with rndc as that does work when named is running, my problem is actually with the /etc/rc.d/init.d/named which is not starting named although it tells me it has. I have been restarting named when adding/editing zones but I now realise that I should just be reloading it. This solves my problem except when my box is rebooted, as it was by my host last night for network maintenance and named isn't starting at boot up but I can monitor that another way and it doesn't happen that often, last night being the first reboot for over a year. Thanks again Greg At 14:26 17/10/2004, Barry Margolin wrote: >In article <cktpb4$2q86$1@sf1.isc.org>, > Greg Mathews <greg@hostingtheinter.net> wrote: > > > Hi > > > > If this isn't the right list for this I appologise now > > > > To restart named etc I've been using the command line > > > > /etc/rc.d/init.d/named restart > >Try: rndc exec > > > > > but recently this has stopped working, ie restart will stop named but > > doesn't restart it again > > > > when I attempt to start named using rndc it returns the error > > > > 'rndc: connect failed: connection refused' > >Rndc works by sending a control message to named. If named isn't >running, there's nothing for it to c...

lan connection to JBoss ... connection refused but connectivity is ok...
I am trying to connect from a client on the same network - and have disabled all firewalls. I have my jboss starting with the bind address of the local ip 192.168.1.104 I can telnet to 192.168.1.104 1099 and see http://linux:8083 which is resolvable - and it ends with 192.168.1.104. I cannot (will not?) annotate my beans to hardcode the IP address... But from another client with jndi.properties of java.naming.factory.initial=org.jboss.security.jndi.LoginInitialContextFactory java.naming.provider.url=jnp://192.168.1.104:1099 java.naming.factory.url.pkgs=org.jboss.naming.cl...

Re: rndc: connect failed: connection refused #3
>>>>> "Greg" == Greg Mathews <greg@hostingtheinter.net> writes: Greg> To restart named etc I've been using the command line Greg> /etc/rc.d/init.d/named restart Greg> but recently this has stopped working, ie restart will stop Greg> named but doesn't restart it again Correct. There is no restart facility in BIND9's rndc. rndc even says that. [There wasn't really one in BIND8 either. It was faked. An "ndc restart" would kill and re-start the name server but didn't know about the command-line arguments that had been supplied to the original invocation of named.] The script above is probably invoking rndc. If you insist on restarting the server, you will have to find some other way of doing that: say by fixing this script. OTOH that could be a waste of time. A name server should only ever need to be restarted after the software has been replaced. Greg> when I attempt to start named using rndc it returns the Greg> error Greg> 'rndc: connect failed: connection refused' Greg> Any idea's why rndc is failing to connect? Because you haven't told the name server to create a listener socket for rndc control messages. Read the section on named.conf's controls{} statement in the ARM. ...

Error connecting to SMTP Server. 111 : Connection refused
When I try to send an e-mail from one of the webppliance 3.1.x box am getting the error Error connecting to SMTP Server. 111 : Connection refused However there is no problem in receiving the e-mail. The sendmail service is running on the server soon I recieve the complains from the users about the same error Can anybody Please help me. Thanks ...

Re: rndc: connect failed: connection refused #2
> Thanks Partick > > The key published is an old one, and not currently valid. I was trying > to show the reader that the keys matched, so the keys did not seem > (??) to be the problem :) > > Still working on the problem. > > Ralph Firstly you are down rev. Secondly use 127.0.0.1 (not localhost) in rndc.conf. > Patrick <patrick@armage.nospamthanx.org> wrote in message news:<c0u8dj$unb$1@ > sf1.isc.org>... > > Ralph wrote: > > > I have installed 9.2.2 on Mac OS X 10.3, and seem to be unable to > > >...

The connection was refused when trying to connect to ...
Wow. Really having a hard time getting my apache web server communicatiing with the outside world. I have a new install of Linux 8. My webserver works fine when used locally but get connection refused other wise. When setting up my server initially I set up with medium firewall, thinking I could modify the settings later. I have tried modifying through security settings, using lokkit and trying to modify the iptables script manually. Is it possible that it isn't the firewall settings at all and some other setting? I am fairly new to Linux and would appreciate any suggestions. Thank...

The connection was refused when trying to connect to ...
Wow. Really having a hard time getting my apache web server > communicatiing with the outside world. I have a new install of Linux > 8. > > My webserver works fine when used locally but get connection refused > other wise. > > When setting up my server initially I set up with medium firewall, > thinking I could modify the settings later. > > I have tried modifying through security settings, using lokkit and > trying to modify the iptables script manually. > > Is it possible that it isn't the firewall settings at all and some > other setting? ...

kprop: Connection refused in call to connect while opening connection to may129.chpc.vnu.edu.vn
Hi everyone, I am installing kerberos, I prepare kerberos system with two KDC named: master.ole.abc (master) slave.ole.abc (slave) I followed the installation guide until http://web.mit.edu/Kerberos/krb5-1.6/krb5-1.6.2/doc/krb5-install.html#Propagate-the-Database-to-Each-Slave-KDC I got an error when I did : kprop -f /opt/krb5-1.6.1/var/krb5kdc/slave_datatrans slave.ole.abc and the error : kprop: Connection refused in call to connect while opening connection to slave.ole.abc I have search google and found someone said that let "kpropd" running in slave KDC. But when I enter co...

Connecting using a Socket
This answer is probably a simple one, but I haven't done a ton of socket programming in Java. On a ConnectException, how can I determine whether the exception was caused by a refused connection (as opposed to a host being unreachable or some other problem?) Sure, I could call getMessage() and see if the message is "Connection refused" - but will that work if my app is translated into another language? -- Steve Sobol, Victorville, California, USA Microsoft's new marketing slogan for Windows is "Life Without Walls." But if you have no walls, how can you have wi...

Problems Connecting... Connection Refused Error...
hi, i am using Eudora 5.2.1 on WIN 2000 Pro... I have patched with MS Blaster and Symnatec FixBlaster tool has not found any Worm on my PC... Since Aug. 16, I am having difficulty connecting my mail servers to retrieve emails. On the same network, I have no problem accessing from other computers. I have re-installed Eudora but problem still persists. I can ping mail servers and access Internet fine.... Any help would be greatly appreciated. Error: Connection Refused (While Login Process)... Regards, Demi ...

connect() connection refused error in lunix
I am trying to connect to a server with IP addres for that i did the following way main(int argc,char ** argv) { int sock_descriptor; unsigned long port; struct sockaddr_in sock_address; sock_descriptor = socket(PF_INET, SOCK_STREAM, 0); memset(&sock_address, 0, sizeof(sock_address)); sock_address.sin_family = PF_INET; sock_address.sin_addr.s_addr=inet_addr((const char*) argv[1]); port = strtol(argv[2], NULL, 10); sock_address.sin_port = htons((unsigned short) port); result = connect(sock_descriptor, (const struct sockaddr *) &sock_address, sizeof(struct sockaddr_in));...

Apache warning: Connection refused: connect to listener
I just installed Apache/2.0.47 (Unix) mod_ssl/2.0.47 OpenSSL/0.9.7b in my server and everything seems to be working fine. I have apache configured to server both secure and insecure pages. However, I keep getting the following line in my error log file: [Wed Sep 24 12:51:15 2003] [warn] (61)Connection refused: connect to listener This happens over and over, especially when I first start the server (but not always), and also randomly during operation. There are no other warnings or errors in my log file. I don't notice any problems with my web pages, but I don't like seei...

Web resources about - connect() failed: Connection refused (111) - comp.parallel.mpi

Connection - Wikipedia, the free encyclopedia
Text is available under the Creative Commons Attribution-ShareAlike License ;additional terms may apply. By using this site, you agree to the ...

iMedia Connection: Interactive Marketing News, Features, Podcasts and Video - iMediaConnection.com
High-quality data, if not used properly, can still lead marketers to make bad decisions. Consider these common ways that numbers are used to ...

HTTP persistent connection - Wikipedia, the free encyclopedia
... tacked on to an existing protocol. If the browser supports keep-alive, it adds an additional header to the request: Following this, the connection ...

CareerSonar Turns Facebook Friends Into Job Connections
Looking for a job ? Among your Facebook friends lies the potential for employment. CareerSonar , a new service, brings together a person’s connections ...

Alec Baldwin, Tommy Hilfiger and a Designer With a ‘Downton’ Connection
It was something of a British invasion at Michael’s today with plenty of posh folks (Majesty magazine EIC Ingrid Seward , the always dashing ...

Igniting Brand Growth Via Emotional Connections
Nobel Prize winning research in neuroscience and behavioral economics proves that all humans are irrational, emotional creatures. Both marketers ...

Did machete-wielding Ohio suspect have radical connections? - Videos - CBS News
... investigators, including possibly the FBI and Joint Terrorism Task Force are looking into whether a brutal rampage in Ohio had any connection ...

Deadpool Really Did Have A Major Connection To Marvel And The Avengers, Here's How
The Deadpool movie was full of a ton of Marvel related easter eggs, literally more than we could count. Most of them were obvious, a few were ...

Second teenage boy arrested in connection with TalkTalk hack
The 16-year-old was arrested in west London on Thursday, and has since been released on bail.

Zika virus-microcephaly connection hunted in Brazil - CNN.com
Is the Zika virus linked to microcephaly? The CDC and Brazilian health officials are leading the first-ever large scale investigation to find ...

Resources last updated: 2/24/2016 11:44:02 PM