At Owen's request I flipped the Steve switch back again:
<use_urlcopy_script> false </use_urlcopy_script>
And the srmcp duly hung trying to copy. So Steve was right
(well of course he was).
-j
> -----Original Message-----
> From: GRIDPP2: Deployment and support of SRM and local storage
> management [mailto:[log in to unmask]]On Behalf Of
> Jensen, J
> (Jens)
> Sent: 11 August 2005 11:03
> To: [log in to unmask]
> Subject: Re: srmcp out of dpm problem
>
>
> > try
> > srm://dpm.epcc.ed.ac.uk:8443/dpm/epcc.ed.ac.uk/home/dteam/test29.txt
> > If that does not generate the same error, then I will email
> > the cern guys
> > and wait on their deliberations.
>
> I had a few problets, having an old config file in
> ~/.srmconfig/config.xml.
>
> This could be related to one of Mona's problems: the hanging srmcp.
> Because that's what happened to me.
>
> I moved the old file out of the way, then let srmcp create a
> new one. Then I edited it to set
> <use_urlcopy_script> true </use_urlcopy_script>
>
> Then I edited the next line, to define the location of the
> copy script:
> <urlcopy> /opt/d-cache/srm/sbin/url-copy.sh </urlcopy>
>
> The default is sbin/urlcopy.sh (no '-').
>
> Without this script, srmcp tries to copy without protocols, and
> will hang. Mona can you please check that this solves one of
> your problems.
>
> Then - success!! I got the file.
>
> --jens
>
|