 |
 |
Hi Volodymyr, That would be one way of doing it. Another way is to have dCache still managing the disk cache but allowing xrootd to access that cache directly (i.e., xrootd would actually be running on the disk cache nodes). If you're going to the root workshop at CERN on the 28th, I will be there as well and would be more than willing to discuss various integration options. Andy ----- Original Message ----- From: "Volodymyr Kapchynskyy" <[log in to unmask]> To: "Artem Trunov" <[log in to unmask]>; <[log in to unmask]>; "Patrick Fuhrmann" <[log in to unmask]> Sent: Monday, September 19, 2005 1:51 AM Subject: Re: dCache wants to communicate with XRootd > Hi Artem! > > Poeple are asking if we can make dCache talk > the xrootd protocol. Just as an experiment we > would like to have a small xrootd instance in > front of our tape based dCache instance so > that we can run standard xrootd applications > agains the xrootd/dCache. > > dCache Team, DESY > > > Artem Trunov wrote: > >>Hi, Volodymyr! >> >>I can't say about your problem, but was curiouse about your application - >>why are you trying to organize such a configuration? It would seem that >>either xrootd or dCache is enough to access the data. I am especially >>curious, because we at in2p3 have both dCache and xrootd, but those two >>pools don't communicate with each other. >> >>Thanks, >>Artem. >> >> >> >>On Fri, 16 Sep 2005, Volodymyr Kapchynskyy wrote: >> >> >>>Hi XRootd Community, >>> >>>We here at DESY were working on interfacing XRootd with dCache. >>>Currently we need the stage-in capability from dCache (which in this >>>case is MSS) into XRootd local folder. >>>And also we need a Migration tool from XRootd local folder into MSS >>>(dCache). We use the oss.msgwcmd and >>>oss.stagecmd directives. Also the oss.rcreate directive is set. Well, >>>whenever the StageIn command is called everything works >>>fine and the files are staged ok. But we have some problems with >>>Flushing files to MSS. >>>Calling "msstalk create mssfn mode" in the most straightforward way does >>>not work. >>>It seems to us that >>>1. The "create" command is called BEFORE the actual file was written >>>into the local folder by xrdcp. So there is nothing to flush. >>>2. It is kind of unknown when the file should be start to being flushed. >>>3. "create" should first touch the file on MSS, return "0" saying that >>>evrything is ok and then it can actually flush the file. Otherwise the >>>flushing does not work. >>> >>>So we wonder how actually the create should function and how it should >>>communicate with XRootd? >>>What kind of logic should be behind the implementation of create part of >>>oss.mssgwcmd command? >>> >>>Volodymyr Kapchynskyy, >>>DESY >>> >>> > >