Hi Maarten,
It's locally mounted
ickstarted 16:26 22-Mar-2005
[root@cmslcgce root]# df -k /opt/globus/tmp
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda1 10104268 2070852 7520144 22% /
[root@cmslcgce root]#
-Ian
On May 23, 2005, at 4:40 PM, Maarten Litmaath, CERN wrote:
> On Mon, 23 May 2005, Ian Fisk wrote:
>
>> Hi Maarten,
>> On May 23, 2005, at 4:18 PM, [log in to unmask] wrote:
>>
>>> On Mon, 23 May 2005, Ian Fisk wrote:
>>>
>>>
>>>> We are experiencing an interesting problem with the
>>>> grid-monitor-manager. It appears to manifest itself in two ways
>>>>
>>>> 1.) users report problems of jobs never exiting the scheduling state
>>>> 2.) Machine memory usage increases until the system goes unstable
>>>>
>>>> What appears to be happening is under some circumstances the lock
>>>> file
>>>> in /opt/globus/tmp for the grid-monitor-manager is corrupted. The
>>>>
>>>
>>> In what way? Can you send me examples?
>>
>> The log gets nulls appended to the start of the lock file. The
>> normal lock has something like
>>
>> 687 1116880933
>>
>> I think the first is PID and the second is a time stamp. After
>> corruption, the file has a series of nulls in front. The more
>> starts, the more nulls.
>
> Is /opt/globus/tmp on an NFS?
|