[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] Fw: Problems about condor slots



Hi Dan,
 
Here is the output of the commands you want me to check out:
 
[root@node9 ~]# condor_config_val -v NUM_CPUS
Not defined: NUM_CPUS
[root@node9 ~]# condor_config_val DETECTED_CORES
4
[root@node9 ~]# condor_config_val -v COUNT_HYPERTHREAD_CPUS
Not defined: COUNT_HYPERTHREAD_CPUS
 
It seems everything is correct.
 
2009-12-30

***********************************************
* Hailong Yang, PhD. Candidate
* Sino-German Joint Software Institute,
* School of Computer Science&Engineering, Beihang University
* Phone: (86-010)82315908
* Email: hailong.yang1115@xxxxxxxxx
* Address: G413, New Main Building in Beihang University,
*              No.37 XueYuan Road,HaiDian District,
*              Beijing,P.R.China,100191
***********************************************

发件人: Dan Bradley
发送时间: 2009-12-29  23:52:20
收件人: Condor-Users Mail List
抄送:
主题: Re: [Condor-users] Fw: Problems about condor slots
hailong.yang1115 wrote:
> 1. The slot number of some nodes in the condor pool mismatched the
> number of logic cpu cores, which could be seen from /proc/cpuinfo. The
> slot number of node9 we noticed from condor_status was 6, while the
> logic cpu cores we found from /proc/cpuinfo is 4.
Check the output of the following commands on the machine where you see
this problem:
condor_config_val -v NUM_CPUS
condor_config_val DETECTED_CORES
condor_config_val -v COUNT_HYPERTHREAD_CPUS
> 2. After installed condor on some nodes, we started condor_master but
> nothing happened. We checked the MasterLog file, it gave the following
> error:
> 12/27 10:48:41 ERROR "can't
> safe_open_wrapper(/tmp/condor-lock.ddgrid0.745993478763015/InstanceLock,O_WRONLY|O_CREAT|O_APPEND
> ,S_IRUSR|S_IWUSR) - errno 2" at line 946 in file master.cpp
I'm guessing that your LOCK directory
/tmp/condor-lock.ddgrid0.745993478763015 has been deleted. Running
condor_init should recreate it. However, I would recommend reconfiguring
LOCK to be somewhere else--not in /tmp--so it doesn't get accidentally
deleted again in the future.
--Dan
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/