Hi,

Am 05.09.2012 um 21:50 schrieb Tillmann Stieger:

> thx, for your quick reply. Yes we have a little cluster with 4 servers. The 
> problem only exists on this machine and also for all users.
> The only difference is that this particular machine is the master host. The 
> other machines are just execution hosts and one of them is also a shadow 
> master. Otherwise they are pretty much identical, software wise.

is this machine mounting /home in a different way than the other machines? Is 
the home for all users inside the private cluster or all using some fileserver 
available outside the cluster too?

-- Reuti


> till
> 
> On 9/5/12 5:09 PM, Reuti wrote:
>> Hi,
>> 
>> Am 05.09.2012 um 15:06 schrieb Tillmann Stieger:
>> 
>>> I have been searching for a solution to my problem for quite a while. But 
>>> no luck so far. I hope that someone here can help me out.
>>> 
>>> On one specific machine I always get the following error-message:
>> You mean you have a larger cluster and problem exists only on one machine, 
>> and there for one or all users?
>> 
>> Is this node in some wy differently configured than the others?
>> 
>> -- Reuti
>> 
>> 
>>> Job 6941 caused action: Job 6941 set to ERROR
>>> 
>>> User        = xxx
>>> Queue       = all.q@xxx
>>> Start Time  = <unknown>
>>> End Time    = <unknown>
>>> failed opening input/output file:09/05/2012 12:48:00 [1111:12201]: can't 
>>> stat() "/home/xxx" as stdout_path
>>> Shepherd trace:
>>> 09/05/2012 12:48:00 [1111:12200]: shepherd called with uid = 1111, euid = 
>>> 1111
>>> 09/05/2012 12:48:00 [1111:12200]: starting up 2011.11
>>> 09/05/2012 12:48:00 [1111:12200]: warning: starting not as superuser 
>>> (uid=1111)
>>> 09/05/2012 12:48:00 [1111:12200]: setpgid(12200, 12200) returned 0
>>> 09/05/2012 12:48:00 [1111:12200]: do_core_binding: "binding" parameter not 
>>> found in config file
>>> 09/05/2012 12:48:00 [1111:12200]: no prolog script to start
>>> 09/05/2012 12:48:00 [1111:12200]: parent: forked "job" with pid 12201
>>> 09/05/2012 12:48:00 [1111:12200]: parent: job-pid: 12201
>>> 09/05/2012 12:48:00 [1111:12201]: child: starting son(job, 
>>> /opt/sge6_2/default/spool/crab2/job_scripts/6941, 0);
>>> 09/05/2012 12:48:00 [1111:12201]: pid=12201 pgrp=12201 sid=12201 old 
>>> pgrp=12200 getlogin()=<no login set>
>>> 09/05/2012 12:48:00 [1111:12201]: reading passwd information for user 'xxx'
>>> 09/05/2012 12:48:00 [1111:12201]: setosjobid: uid = 1111, euid = 1111
>>> 09/05/2012 12:48:00 [1111:12201]: setting limits
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_CPU setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_FSIZE setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_DATA setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_STACK setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_CORE setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_VMEM/RLIMIT_AS setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: RLIMIT_RSS setting: (soft 
>>> 18446744073709551615(INFINITY), hard 18446744073709551615(INFINITY)) 
>>> resulting: (soft 18446744073709551615(INFINITY), hard 
>>> 18446744073709551615(INFINITY))
>>> 09/05/2012 12:48:00 [1111:12201]: setting environment
>>> 09/05/2012 12:48:00 [1111:12201]: Initializing error file
>>> 09/05/2012 12:48:00 [1111:12201]: switching to intermediate/target user
>>> 09/05/2012 12:48:00 [1111:12201]: tried to change uid/gid without being root
>>> 09/05/2012 12:48:00 [1111:12201]: try running further with uid=1111
>>> 09/05/2012 12:48:00 [1111:12201]: closing all filedescriptors
>>> 09/05/2012 12:48:00 [1111:12201]: further messages are in "error" and 
>>> "trace"
>>> 09/05/2012 12:48:00 [1111:12201]: can't stat() "/home/xxx" as stdout_path: 
>>> Permission denied KRB5CCNAME=none uid=1111 gid=100 100
>>> 09/05/2012 12:48:00 [1111:12200]: wait3 returned 12201 (status: 6656; 
>>> WIFSIGNALED: 0,  WIFEXITED: 1, WEXITSTATUS: 26)
>>> 09/05/2012 12:48:00 [1111:12200]: job exited with exit status 26
>>> 09/05/2012 12:48:00 [1111:12200]: reaped "job" with pid 12201
>>> 09/05/2012 12:48:00 [1111:12200]: job exited not due to signal
>>> 09/05/2012 12:48:00 [1111:12200]: job exited with status 26
>>> 09/05/2012 12:48:00 [1111:12200]: now sending signal KILL to pid -12201
>>> 09/05/2012 12:48:00 [1111:12200]: no tasker to notify
>>> 09/05/2012 12:48:00 [1111:12200]: failed starting job
>>> 09/05/2012 12:48:00 [1111:12200]: no epilog script to start
>>> 
>>> Shepherd error:
>>> 09/05/2012 12:48:00 [1111:12201]: can't stat() "/home/xxx" as stdout_path: 
>>> Permission denied KRB5CCNAME=none uid=1111 gid=100 100
>>> 
>>> Shepherd pe_hostfile:
>>> xxx 1 all.q@xxx UNDEFINED
>>> 
>>> 
>>> 
>>> Has someone experienced similar issues?
>>> 
>>> I would really appreciate any advice.Thanks.
>>> till
>>> 
>>> _______________________________________________
>>> users mailing list
>>> [email protected]
>>> https://gridengine.org/mailman/listinfo/users
>> 
> 


_______________________________________________
users mailing list
[email protected]
https://gridengine.org/mailman/listinfo/users

Reply via email to