Print

Print


Hi Wilko,

  since I have been able to set up my servers, I tried the error on redir.

  In my opinion it works, I have a redirector on noric02 and a data 
server on noric04, and this is what I get:

050621 16:32:18 9113 Xrd: ReadRaw: Disconnection reported 
onnoric04.slac.stanford.edu:1094
050621 16:32:18 9113 Xrd: PhyConnection: Disconnecting socket...
050621 16:32:18 9113 Xrd: XrdClientMessage::ReadRaw: Error reading 8 bytes.
050621 16:32:18 9113 Xrd: CheckAutoTerm: Self-Cancelling reader thread.
050621 16:32:18 9113 Xrd: SocketReaderThread: Reader Thread exiting.
050621 16:32:18 001 Xrd: ClientSock::SendRaw: Disconnection detected 
writing 500000 bytes to socket 4 
(server[noric04.slac.stanford.edu:1094]). Revents=32
050621 16:32:18 001 Xrd: WriteRaw: Disconnection reported 
onnoric04.slac.stanford.edu:1094
050621 16:32:18 001 Xrd: PhyConnection: Disconnecting socket...
050621 16:32:18 001 Xrd: WriteToServer: Error sending 500000 bytes in 
the data part to server [noric04.slac.stanford.edu:1094].
050621 16:32:18 001 Xrd: PhyConnection: Disconnecting socket...
050621 16:32:18 001 Xrd: HandleServerError: Redir count=1
050621 16:32:18 001 Xrd: HandleServerError: Communication error with 
server [noric04.slac.stanford.edu:1094]. Rebouncing here.
050621 16:32:18 001 Xrd: TakeUrl: parsing url: noric04.slac.stanford.edu
050621 16:32:18 001 Xrd: TakeUrl:    HostWPort:   noric04.slac.stanford.edu
050621 16:32:18 001 Xrd: TakeUrl:    File:   /
050621 16:32:18 001 Xrd: TakeUrl:    Host:   noric04.slac.stanford.edu
050621 16:32:18 001 Xrd: TakeUrl:    Port:   0
050621 16:32:18 001 Xrd: HandleServerError: Received redirection to 
[134.79.86.54:1094]. Token=[].
050621 16:32:21 001 Xrd: GetDomainToMatch: GetHostName(134.79.86.54) 
returned name=noric04.slac.stanford.edu


  From here it's clear that the client rebounces to noric04 instead of 
going back. So I think we have to clarify ourselves.

  Instead, due to the new commits to XrdClientUrl, it seems that the 
session stuff is (again!) broken, at least through redirections. I'll 
check better.


  Fabrizio

Wilko Kroeger wrote:
> Hello Pete
> 
> I will not make it to the meeting, so here is the status of xrootd at SLAC:
> 
> - No problems with the kan and bbrprod clusters.
> - Still didn't switch to "writing via the redirector" for skimming.
>   (Will be done when the next production version is available)
> 
> - Testing the latest development version 20050620-0614
> 
>   * able to write via the redirector (didn't work before
>     because of the session id problem).
>   * xrdcp (for writing) goes back to the redirector if a data server goes
>     don. My understanding is that that xrdcp (after being redirected
>     ones) should not go back to the redirector.
>   * Can't get XrdClientAdmin to work (reported in another thread)
>   * xrootd.protocol option causes server not to start if used in config
>     file (reported in previous email)
> 
>   I will do more test in the next few days.
> 
> Cheers,
>    Wilko
> 
> 
> On Mon, 20 Jun 2005, Peter Elmer wrote:
> 
> 
>>  Hi All,
>>
>>  The next meeting on xrootd development, testing, deployment and monitoring
>>will be Tuesday, 21 June, 2005 at 10:15 CA, 19:15 Europe and 18:15 UK. The
>>location at SLAC is SCS conference room B. Others can just congregate
>>virtually.
>>
>>  The connection info for this meeting is:
>>
>>   phone:     +1-510-665-5437
>>   passcode:  4465
>>
>> Agenda:
>>
>>   o problem list
>>     o http://xrootd.slac.stanford.edu/problems.html
>>   o New development/features
>>      o outstanding problems for those using production version 20050328-0656
>>      o new pre-production version 20050620-0614
>>         o testing!
>>         o Final production version???
>>      o 1000 server test
>>      o server side monitoring - Jacek
>>      o ./configure --prefix=/bla/bla
>>      o posix and TXNetFile client merge
>>      o security stuff
>>      o async read-ahead
>>      o rfio mps scripts
>>      o TXNetFile on Windows
>>      o "admin" interface (shutdown, redirecting clients, ...)
>>      o proxy support
>>      o unsolicited responses
>>      o updates needed by Castor people
>>      o SRM
>>   o Sites
>>     o Example configurations
>>     o SLAC
>>        o olb03/olb04/kanolb-a renaming (arrghh!)
>>     o Cornell
>>        o problems over the past week
>>     o Padova - Fulvio
>>     o CNAF - Enrica
>>     o RAL - Chris/Manny
>>        o config rpm/files?
>>     o In2p3 - Jean-Yves
>>     o FZK - Gregory
>>     o BNL
>>     o anybody else...
>>   o xrootd procedures
>>      o restart of servers (deployment of new versions)
>>      o problems with particular data servers
>>      o adding a new server
>>
>>  I have almost certainly forgotten some things, so please feel free to bring
>>anything important up during the meeting.
>>
>>                                    Pete
>>
>>-------------------------------------------------------------------------
>>Peter Elmer     E-mail: [log in to unmask]      Phone: +41 (22) 767-4644
>>Address: CERN Division PPE, Bat. 32 2C-14, CH-1211 Geneva 23, Switzerland
>>-------------------------------------------------------------------------
>>