Hi Derek,
so if globus-gridftp doesn't start also on the nodes that is not a
problem? :)
cheers
alessandra
On Thu, 12 May 2005, Ross, D (Derek) wrote:
> Hi,
>
> dCache/SRM doesn't need the globus-gridftp service, it has its own gridftp door started by the dcache-opt service if you've configured it to do that. Both will attempt to use port 2811, so you can only use one.
>
> The error message suggests to me that there's something odd with your pool gw03_1. Does a "show pinboard" in the PoolManager directory of the admin interface after a failed transfer provide any information? You can use "dump pinboard <filename>" to output the pinboard to a file as well.
>
> Derek
>
>
>> -----Original Message-----
>> From: GRIDPP2: Deployment and support of SRM and local storage
>> management [mailto:[log in to unmask]]On Behalf Of owen
>> maroney
>> Sent: 12 May 2005 16:25
>> To: [log in to unmask]
>> Subject: Re: SRM endpoints
>>
>>
>> Hi,
>>
>> Up and running and ready for tests now!
>>
>> Although we have tried the simple srmcp command, and guess what! It
>> failed. :-(
>>
>> We've attached the error message produced by the srmcp
>> command and also
>> we've tarred up the logs on the admin node (attached) from
>> the startup.
>>
>> FYI: The admin node was a completely clean installation, from scratch.
>> There were two problems we discovered needing fixing to get the admin
>> node to install properly using lcg yaim:
>>
>> 1 - the user postgres was not being created, so before
>> running the yaim
>> scripts we had to:
>>
>> useradd -u 26 postgres
>>
>> 2 - the portmap service wasn't running by default and the pnfs
>> installation/setup didn't check this (and so had problems)
>>
>> /sbin/service portmap restart
>>
>> We think the reason the srmcp is failing is related to the fact that
>> there is no gridftp service running on the pool node: when we try to
>> restart it, we get the same error message Alessandra was
>> noticing on the
>> admin node, though. There doesn't seem to be any other
>> services running
>> on the pool node listening on port 2811, though.
>>
>> owen maroney wrote:
>>
>>> Ahem, I quote from my original message:
>>>
>>>> (Although it is currently being restarted so
>>>> don't be surprised if tests fail just now!)
>>>
>>>
>>> Perhaps I didn't mention: it is the entire node that is
>> being restarted
>>> (not merely the dCache services), please be patient!
>>>
>>> We'll let you know when all services are running again.
>>>
>>> cheers,
>>> Owen.
>>>
>>> Owen Synge wrote:
>>>
>>>> On Thu, 2005-05-12 at 14:30 +0100, owen maroney wrote:
>>>>
>>>>> gw04.hep.ph.ic.ac.uk
>>>>
>>>>
>>>>
>>>> Jiri did a quick test, he has some nice scripts but we
>> could not get the
>>>> WSDL using srmcp so is the SRM door up and running?
>>>>
>>>> Regards
>>>>
>>>> Owen
>>>
>>>
>>> --
>>> =======================================================
>>> Dr O J E Maroney # London Tier 2 Technical Co-ordinator
>>>
>>> Tel. (+44)20 759 47802
>>>
>>> Imperial College London
>>> High Energy Physics Department
>>> The Blackett Laboratory
>>> Prince Consort Road, London, SW7 2BW
>>> ====================================
>>
>> --
>> =======================================================
>> Dr O J E Maroney # London Tier 2 Technical Co-ordinator
>>
>> Tel. (+44)20 759 47802
>>
>> Imperial College London
>> High Energy Physics Department
>> The Blackett Laboratory
>> Prince Consort Road, London, SW7 2BW
>> ====================================
>>
>
--
********************************************
* Dr Alessandra Forti *
* Technical Coordinator - NorthGrid Tier2 *
* http://www.hep.man.ac.uk/u/aforti *
********************************************
|