Print

Print


Hi Doug,

I think I know what is going one. You have an oss.localroot setup, which is a simple form of N2N. In the current release, the xrootd passes the LFN to the checksum program, instead of PFN (=N2N(LFN)). So on the data server side xrootd runs:

/usr/bin/xrdadler32 /atlas/... instead of /local/xrootd/a/atlas ...

 You can certainly work around this issue by telling xrootd.checksum to use a script which will do a simple N2N. It is not pretty. Andy said that this will be addressed in 3.1

regards,
Wei Yang  |  [log in to unmask]  |  650-926-3338(O)


On Aug 18, 2011, at 2:18 PM, Doug Benjamin wrote:

> Hi Wei,
> 
>   Here is the effect of me running the commands:
> 
> [dbenjamin@atlas11 xrootd-test]$ xrd ascnfs.hep.anl.gov locateall 
> /atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> 
> 
> ------------- Location #1
> InfoType: kXrdcLocDataServer
> CanWrite: true
> Location: '130.202.173.42:0'
> 
> [dbenjamin@atlas11 xrootd-test]$ xrd ascnfs.hep.anl.gov getchecksum 
> /atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> Checksum: adler32 Error_accessing 
> /atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> 
> 
> [dbenjamin@atlas11 xrootd-test]$ ./xrdadler32 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> Error_accessing 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> 
> 
> [dbenjamin@atlas11 xrootd-test]$ xrdcp 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root 
> /tmp/NTUP_SUSY_p428_00153.root
> [xrootd] Total 267.37 MB        |====================| 100.00 % [105.8 MB/s]
> 
> 
> [dbenjamin@atlas11 xrootd-test]$ gdb --args ./xrdadler32 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> GNU gdb (GDB) Red Hat Enterprise Linux (7.0.1-37.el5)
> Copyright (C) 2009 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
> and "show warranty" for details.
> This GDB was configured as "x86_64-redhat-linux-gnu".
> For bug reporting instructions, please see:
> <http://www.gnu.org/software/gdb/bugs/>...
> Reading symbols from /users/dbenjamin/xrootd-test/xrdadler32...done.
> (gdb) run
> Starting program: /users/dbenjamin/xrootd-test/xrdadler32 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> warning: no loadable sections found in added symbol-file system-supplied DSO at 0x2aaaaaaab000
> [Thread debugging using libthread_db enabled]
> [New Thread 0x40a00940 (LWP 3155)]
> [New Thread 0x41401940 (LWP 3156)]
> Error_accessing 
> root://ascnfs.hep.anl.gov:1094//atlas/ANLASC_TEST/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281325_00/NTUP_SUSY_p428_00153.root
> [Thread 0x40a00940 (LWP 3155) exited]
> [Thread 0x41401940 (LWP 3156) exited]
> 
> Program exited normally.
> 
> 
> The xrootd config file is  http://ascnfs.hep.anl.gov:2080/ascnfs-xrootd-clustered.cfg
> 
> Thanks,
> 
> Doug
> 
> 
> On 08/17/2011 04:13 PM, Yang, Wei wrote:
>> Hi Doug,
>> 
>> Can you try the one at http://www.slac.stanford.edu/~yangw/xrdadler32. It will print out some message which will help me to understand where the program stops. Please use it only as client, not in your xrootd configuration file.
>> 
>> regards,
>> Wei Yang  |  [log in to unmask]  |  650-926-3338(O)
>> 
>> 
>> On Aug 16, 2011, at 2:43 PM, Doug Benjamin wrote:
>> 
>>> Hi Wei,
>>> 
>>>   Sorry to hear that you got sick. If you can give me a binary with more debug information. I will gladly run it. Or if someone can tell me how
>>> to build such an exe I will do so.
>>> 
>>> Doug
>>> 
>>> On Aug 16, 2011, at 5:29 PM, Yang, Wei wrote:
>>> 
>>>> Hi Doug,
>>>> 
>>>> Sorry, I got germ from my son and couldn't keep up. I am wondering why xrd's getchecksum fail. Also, can step through the main function of Xrdadler32.cc and tell me which line print out this message? (later, I might probably give you a binary that print out more debug info in order to isolate the problem).
>>>> 
>>>> regards,
>>>> Wei Yang  |  [log in to unmask]  |  650-926-3338(O)
>>>> 
>>>> 
>>>> On Aug 15, 2011, at 1:53 PM, Doug Benjamin wrote:
>>>> 
>>>>> Hi Wei,
>>>>> 
>>>>> Not sure if this helps
>>>>> 
>>>>> This GDB was configured as "x86_64-redhat-linux-gnu".
>>>>> For bug reporting instructions, please see:
>>>>> <http://www.gnu.org/software/gdb/bugs/>...
>>>>> Reading symbols from /usr/bin/xrdadler32...Reading symbols from /usr/lib/debug/usr/bin/xrdadler32.debug...
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> 
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> 
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> done.
>>>>> done.
>>>>> (gdb) run
>>>>> Starting program: /usr/bin/xrdadler32 root://nsfprv//atlas/dq2/mc10_7TeV/NTUP_SUSY/e577_s933_s946_r1831_r2040_p428/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281327_00/NTUP_SUSY.281327._006235.root.1
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> [Thread debugging using libthread_db enabled]
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> [New Thread 0x40a00940 (LWP 18537)]
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.liblist not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .gnu.conflict not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> warning: section .dynbss not found in /usr/lib/debug/usr/bin/xrdadler32.debug
>>>>> Error_accessing: root://nsfprv//atlas/dq2/mc10_7TeV/NTUP_SUSY/e577_s933_s946_r1831_r2040_p428/mc10_7TeV.105802.JF17_pythia_jet_filter.merge.NTUP_SUSY.e577_s933_s946_r1831_r2040_p428_tid281327_00/NTUP_SUSY.281327._006235.root.1
>>>>> [Thread 0x40a00940 (LWP 18537) exited]
>>>>> 
>>>>> 
>>>>> Also xrd command fails.
>>>>> 
>>>>> Doug
>>>>> 
>>>>> On Aug 13, 2011, at 3:00 AM, Yang, Wei wrote:
>>>>> 
>>>>>> Ok, I probably need you to run debugger and tell me which line print out that Error_accessing. Also, can you use xrd's getchecksum sub-command to retrieve the checksum?
>>>>>> 
>>>>>> regards,
>>>>>> Wei Yang  |  [log in to unmask]  |  650-926-3338(O)
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> On Aug 11, 2011, at 1:01 PM, Doug Benjamin wrote:
>>>>>> 
>>>>>>> Hi,
>>>>>>> 
>>>>>>> the security is unix security and I did xrdcp by hand.
>>>>>>> Thanks,
>>>>>>> Doug
>>>>>>> 
>>>>>>> On Aug 11, 2011, at 3:24 PM, Yang, Wei wrote:
>>>>>>> 
>>>>>>>> another think I forget to ask, how about the security module? and can you do an xrdcp?
>>>>>>>> 
>>>>>>>> regards,
>>>>>>>> Wei Yang  |  [log in to unmask]  |  650-926-3338(O)
>>>>>>>> 
>>>>>>>> 
>>>>>>>> On Aug 11, 2011, at 10:57 AM, Doug Benjamin wrote:
>>>>>>>> 
>>>>>>>>> Hi Wei,
>>>>>>>>> 
>>>>>>>>> The some machines that was using for these tests all are dual nic machines on a public and private network
>>>>>>>>> (yes -  I know a very troublesome idea).   I can use xrdcp to copy the files by hand using the private addresses.
>>>>>>>>> Somehow the xrdalder32 command is failing.
>>>>>>>>> 
>>>>>>>>> Cheers,
>>>>>>>>> 
>>>>>>>>> Doug
>>>>>>>>> 
>>>>>>>>> On Aug 11, 2011, at 1:49 PM, Yang, Wei wrote:
>>>>>>>>> 
>>>>>>>>>> Hi Doug,
>>>>>>>>>> 
>>>>>>>>>> I noticed that the data server is on private IP. Just want to make sure that doesn't play a role.
>>>>>>>>>> 
>>>>>>>>>> regards,
>>>>>>>>>> Wei Yang  |  [log in to unmask]  |  650-926-3338(O)
>>>>>>>>>> 
>>>>>>>>>> 
>>>>>>>>>> 
>>>>>>>>>> 
>>>>>>>>>> On Aug 11, 2011, at 5:18 AM, Doug Benjamin wrote:
>>>>>>>>>> 
>>>>>>>>>>> Hi,
>>>>>>>>>>> 
>>>>>>>>>>> I am trying to understand why my use of xrdalder32 command is failing.
>>>>>>>>>>> 
>>>>>>>>>>> I have a file in xrootd storage -
>>>>>>>>>>> I can see it from the redirector
>>>>>>>>>>> 
>>>>>>>>>>> [root@wnode01 ~]# xrd hnode.atlaspitt:1094 locateall /atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>> ------------- Location #1
>>>>>>>>>>> InfoType: kXrdcLocDataServer
>>>>>>>>>>> CanWrite: true
>>>>>>>>>>> Location: '192.168.100.101:1094'
>>>>>>>>>>> 
>>>>>>>>>>> Yet I get this result from xrdalder32
>>>>>>>>>>> 
>>>>>>>>>>> [root@wnode01 ~]# /usr/bin/xrdadler32  root://hnode.atlaspitt:1094//atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> Error_accessing root://hnode.atlaspitt:1094//atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> 
>>>>>>>>>>> I get a similar error when I query the data server with the file :
>>>>>>>>>>> 
>>>>>>>>>>> [root@wnode01 ~]# /usr/bin/xrdadler32  root://wnode01.atlaspitt:1094//atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> Error_accessing root://wnode01.atlaspitt:1094//atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> 
>>>>>>>>>>> Yet when I query the actual file on disk:   the xrdalder32 succeeds.
>>>>>>>>>>> 
>>>>>>>>>>> [root@wnode01 ~]# /usr/bin/xrdadler32  /local/xrootd/a/atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> d26e385b /local/xrootd/a/atlas/local/bdouglas/data11_7TeV.00180224.physics_Egamma.merge.NTUP_EGAMMA.f368_m806_p536_tid327358_00/NTUP_EGAMMA.327358._000001.root.1
>>>>>>>>>>> 
>>>>>>>>>>> Am I doing something wrong or did I find an undocumented feature.
>>>>>>>>>>> 
>>>>>>>>>>> Thanks,
>>>>>>>>>>> 
>>>>>>>>>>> Doug Benjamin
>>>>>>>>>>> 
>>>>>>>>>>> 
>>>>>>>>>>> 
>