Hello Catalin,
There is a problem with this version.
It has been observed here at IN2P3-CC, but also at TRIUMF and SARA.
We have put a cron to restart the daemon automatically.
It is fixed in 1.6.11 release which is in certification.
Cheers,
David
*David BOUVET*
/EGEE Project team/
IN2P3/CNRS Computing Centre - Lyon (FRANCE)
http://grid.in2p3.fr
Tel. : +33 4 72 69 41 62 | Fax. : +33 4 72 69 41 70 | e-mail :
[log in to unmask]
Condurache, C (Catalin) a écrit :
> Hi,
>
> It seems that after an upgrade to LFC-* packages (ver 1.6.10-4sec.slc4)
> on 24 June, the lfcdaemon behaves abnormally and also dies with the
> following error "Jun 29 05:47:03 lfc0448 lfcdaemon: *** glibc detected
> *** double free or corruption (out): 0xaa580760 ***"
>
> I have no problems for a long period of time before the update.
>
> Anyone with similar problems?
>
> Cheers,
> Catalin Condurache
> RAL Tier1
>
>
>
>> -----Original Message-----
>> From: LHC Computer Grid - Rollout [mailto:[log in to unmask]]
>>
> On
>
>> Behalf Of Remi Mollon
>> Sent: 02 July 2008 10:36
>> To: [log in to unmask]
>> Subject: Re: [LCG-ROLLOUT] lcg-cp and file:/
>>
>> Hi Greig,
>>
>> >From GFAL 1.10.9, if you specify a file with file: protocol, and a
>> relative path, then the current directory ($PWD environment variable)
>> will be used.
>> So my guess is that at Glasgow there is an old version of GFAL (<
>> 1.10.9), you can get the version number with "lcg-cp --version".
>>
>> FYI, the latest production version is GFAL 1.10.11 (and lcg-utils
>>
> 1.6.11).
>
>> Moreover, for lcg-cp and lcg-cr, file: is the default protocol, I mean
>> that it is no more mandatory to specify "file:".
>> For example, if your current directory is /home/user/dir,
>>
> "myfile.test"
>
>> as destination file for lcg-cp, is the same as
>> "file:/home/user/dir/myfile.txt".
>>
>> Regards,
>> Remi.
>>
>>
>> On 07/02/2008 10:26 AM, Greig Alan Cowan wrote:
>>
>>> Hi,
>>>
>>> Can anyone explain the following output of the lcg-cp command as run
>>> on two different sites? I understand the first two examples but I
>>>
> then
>
>>> don't understand the last example which is successful even though
>>>
> the
>
>>> destination file is specified as file:filename rather than
>>> file:/path/to/filename. Is there an environment variable not being
>>> specified?
>>>
>>> Thanks,
>>> Greig
>>>
>>> From Glasgow without path:
>>>
>>> lcg-cp -v -t 120 --vo atlas
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
>>> file:data_1mb
>>> Source URL:
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
>>> File size: 1048576
>>> Source URL for copy:
>>>
>>>
> gsiftp://storage004.ppgrid1.rhul.ac.uk/storage004.ppgrid1.rhul.ac.uk:/ra
> id
>
> /dpmfs/atlas/2008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5.384788.
> 0
>
>>> Destination URL: file:data_1mb
>>> # streams: 1
>>> # set timeout to 120 (seconds)
>>> [globus_gass_copy_get_url_mode]: globus_url_parse returned error
>>>
> code:
>
>>> -8 for url: file:data_1mb
>>>
>>> From Glasgow with path:
>>>
>>> lcg-cp -v -t 120 --vo atlas
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
> file:/tmp/WMS_node007_018435_https_3a_2f_2flcgrb02.gridpp.rl.ac.uk_3a900
> 0_
>
>> 2f_5fG21zU-NYbxNTinPH8tL7w/data_1mb
>>
>>> Source URL:
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
>>> File size: 1048576
>>> Source URL for copy:
>>>
>>>
> gsiftp://storage004.ppgrid1.rhul.ac.uk/storage004.ppgrid1.rhul.ac.uk:/ra
> id
>
> /dpmfs/atlas/2008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5.384788.
> 0
>
>>> Destination URL:
>>>
>>>
> file:/tmp/WMS_node007_018435_https_3a_2f_2flcgrb02.gridpp.rl.ac.uk_3a900
> 0_
>
>> 2f_5fG21zU-NYbxNTinPH8tL7w/data_1mb
>>
>>> # streams: 1
>>> # set timeout to 120 (seconds)
>>>
>>> Transfer took 1000 ms
>>>
>>> From Lancs without path:
>>>
>>> lcg-cp -v -t 120 --vo atlas
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
>>> file:data_1mb
>>> Source SE type: SRMv1
>>> Source URL:
>>>
>>>
> srm://se2.ppgrid1.rhul.ac.uk/dpm/ppgrid1.rhul.ac.uk/home/atlas/generated
> /2
>
>> 008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5
>>
>>> File size: 1048576
>>> Source URL for copy:
>>>
>>>
> gsiftp://storage004.ppgrid1.rhul.ac.uk/storage004.ppgrid1.rhul.ac.uk:/ra
> id
>
> /dpmfs/atlas/2008-06-26/file34d9fd07-7857-4f02-82a8-9a25e3fbdfd5.384788.
> 0
>
>>> Destination URL:
>>> file:/pool/1458393.fal-pygrid-
>>>
> 18.lancs.ac.uk/WMS_node198_018071_https_3a_2f_2flcgrb02.gridpp.rl.ac.uk_
> 3a
>
>> 9000_2fSpiJZM0lEhvjt5HSv7jooA/data_1mb
>>
>>> # streams: 1
>>> # set timeout to 120 (seconds)
>>>
>>> Transfer took 2010 ms
>>>
>>>
>
>
|