SCIENTIFIC-LINUX-USERS Archives

March 2009

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jan Iven <[log in to unmask]>
Reply To:
Date:
Sun, 8 Mar 2009 12:10:19 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (35 lines)
On 03/04/2009 06:59 PM, Stephan Wiesand wrote:
> On Tue, 3 Mar 2009, Yan Xiaofei wrote:
> 
>> I install slc45 and use the hugemem kernel.
> 
> I'm afraid the client for the hugemem kernel is not very well tested,
> but let's see.
> 
>> But afs client didn't work. I can look the directory and file, But when
>> i open the file, the content of the file is incorrect, and can not copy
>> file out of afs directory.
>>
>> Sorry put the problem here.
> 
> No problem. That's what this list is for. But please provide some more
> information:
> 
> - Are you indeed using SL*C* and not plain SL? That's important to know
> because these have different openafs builds. And what versions are you
> using (output of "rpm -qa kernel\*" ?)?

I confirm that we occasionally see "weird" issues on hugemem kernels
with "our" AFS on a test machine - typically file content looks like
every char is replaced by a 16bit entity (which  occasionally gets
misinterpreted as UTF by the applications). Since nobody seems to be
using that kernel at CERN, we haven't investigated any further so far..
(and we just recompile the modules for hugemem since that is done by a
script).

And yes, you should not be using SLC45 (unless you have applied all
updates, in which case this would be equivalent to SLC47).

Best regards
Jan

ATOM RSS1 RSS2