[GE users] weird qrsh + ssh combo

Jeroen M. Kleijer jeroen.m.kleijer at philips.com
Mon Nov 21 15:11:48 GMT 2005


What ports need to be accessible from the client?
I can telnet to the qmaster on port 536 and I can telnet to the execution 
host on port 537 (my ssh daemon runs on port 538).
Do any other ports need to be accessible?

Met vriendelijke groeten / Kind regards

Jeroen Kleijer
Unix Systeembeheer
Philips Applied Technologies








"Jeroen M. Kleijer" <jeroen.m.kleijer+FromInternet at philips.com> 
2005-11-21 01:06 PM
Please respond to
users at gridengine.sunsource.net


To
users at gridengine.sunsource.net
cc

Subject
Re: [GE users] weird qrsh + ssh combo
Classification








Hi, 

I followed the howto and it worked before I moved a lot of the systems 
frmo site A to site B, however, some workstations still reside on site A 
for about a month or so and after the move the qrsh / ssh combo stopped 
working from site A. 
The combo worked before the move site B but I'm still trying to figure out 
why a regular openssh -Y <machine> <command> still works but the qrsh 
<command> doesn't even though nothing has changed in the cluster 
configuration. 
Is there a way to turn on logging of any kind?

Met vriendelijke groeten / Kind regards

Jeroen Kleijer
Unix Systeembeheer
Philips Applied Technologies 







Reuti <reuti at staff.uni-marburg.de> 
2005-11-21 12:58 PM 

Please respond to
users at gridengine.sunsource.net


To
users at gridengine.sunsource.net 
cc

Subject
Re: [GE users] weird qrsh + ssh combo 
Classification









Hi,

Am 21.11.2005 um 12:47 schrieb Jeroen M. Kleijer:

>
> Hi,
>
> I've got a bit of a strange situation around here.
> I've got a couple of workstations at site A that are perfectly able 
> to do a passwordless "openssh -Y <machine>" to an interactive 
> server on site B. I can remotely start an xterm, uptime etc....
> However, when I add the line "rsh_command /appl/openssh.sge/cur/bin/ 
> ssh -Y" to the clusterconfiguration of these workstations at site 
> A, the qrsh command does absolutely nothing and gets a time-out 
> with the following message in the qmaster-logfile:

there's a Howto to use ssh with SGE:

http://gridengine.sunsource.net/howto/qrsh_qlogin_ssh.html

Did you also change the definition of the daemon process?

Cheers - Reuti

> 11/21/2005 12:36:23|qmaster|nlyehvsaq1lx001|W|job 2366.1 failed on 
> host nlcftcs11 assumedly after job because: job 2366.1 died through 
> signal HUP (1)
>
> I've got a workstation at site B that is exactly the same as the 
> ones at site A but this workstation does seem to work.
> Has anyone encountered this before?
>
> Met vriendelijke groeten / Kind regards
>
> Jeroen Kleijer
> Unix Systeembeheer
> Philips Applied Technologies


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net





More information about the gridengine-users mailing list