SCIENTIFIC-LINUX-USERS Archives

February 2008

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:
Killian De Volder <[log in to unmask]>
Reply To:
Killian De Volder <[log in to unmask]>
Date:
Sat, 23 Feb 2008 20:51:38 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (60 lines)
Good day,

I now installed SL4 ...
It seems however that perl(SOAP::Lite) seems to be missing from it:
--> Processing Dependency: perl(SOAP::Lite) for package:
glite-data-catalog-api-perl
--> Processing Dependency: perl(SOAP::Lite) for package:
glite-data-transfer-api-perl

(Note:
    I did a minimal installation to safe hard disk space and
    I will install any other package / dependency that is needed.
    Why I tell this? I'm not sure if doing so also omits repos from the
instaltion.

)
Did I do something wrong or should I just really user tarUI instead ?
But personally I prefer not installing not-in-your-software-tree-files
(wel atleast not over / ).

Thanks in advance,
Killian

Jan Iven wrote:
> On 22/02/08 16:56, Killian De Volder wrote:
>> Hello,
>> I'm settings up a glite-UI and following this tutorial:
>> https://twiki.cern.ch/twiki/bin/view/LCG/GenericInstallGuide310.
>> However at a certain point during the installation
>> (https://twiki.cern.ch/twiki/bin/view/LCG/GenericInstallGuide310#The_install)
>>
>> I get the next list of missing dependencies:
>> Error: Missing Dependency: libboost_program_options.so.1 is needed by
>> package glite-wms-common
> ..
>
>> (The tutorial has been written for SL4 and I am using SL51).
>
> So the SL5 "boost" RPM provides
> libboost_thread.so.2 (etc) whereas your application has been linked
> against an older version. Either you find a way to install the SL4
> version on your SL5 box (which may cause other dependency issues,
> unless you repackage it), or you are out of luck.
>
>> What would be the correct method to resolve this issue please?
>
> I would contact the developer team and see whether they have a
> timeline for a SL5-compatible version. Given that CERN has only very
> recently decided to do anything with SL*5, this may not be the case,
> but external requests would help (perhaps) to raise the priority of
> such a release.
>
> You might be able too run a SL4 virtual machine instead on your SL5
> box. Or (if you want production stability instead of sorting out
> packaging troubles) go back to SL4 for now.
>
> just my 0.02 €
> Jan
>

ATOM RSS1 RSS2