On 2 July 2014 11:41, Ewan MacMahon <[log in to unmask]> wrote:
>> -----Original Message-----
>> From: GRIDPP2: Deployment and support of SRM and local storage management
>> [mailto:[log in to unmask]] On Behalf Of John Hill
>> Sent: 02 July 2014 11:32
>> To: [log in to unmask]
>> Subject: Re: http/webdav at UK
>>
>> This is the exchange on the ATLAS support list. In case it is useful or
>> relevant, attempted WebDAV accesses generate entries in
>> /var/log/http/error_log of the form
>>
>> [Wed Jul 02 11:24:39 2014] [error] A subtree cannot specify a different
>> DAV provider than its parent.
>>
>
> On the Cambridge problem, it might be useful to know what files are
> in /etc/httpd/conf.d/ because at first reading, that error sounds
> like it thinks something's being defined twice. Also, is the head
> node just a head node, or is it combined with a pool node?
Indeed: this error usually happens (in httpd/WebDAV in general) when
you have overlapping paths being shared by different virtual hosts
etc.
It'd be nice to know what the config was, and how many daemons are running...
Sam
>
> On the more general points that we deviated into a bit at the end of
> the meeting, the Oxford SE is behaving like the Liverpool and Lancaster
> ones in that its 'front page' isn't accessible, but the 'top of the
> namespace' (at https://t2se01.physics.ox.ac.uk/dpm/physics.ox.ac.uk/ )
> is. Also, a quick look at a couple of pool nodes' logs suggests that
> the httpd falling over all the time problem may indeed have gone away
> with the update to 1.8.8; I'm seeing routine restarts many times a day
> before that, and none after.
>
> Ewan
|