Hello,
Now I understand the problem. However, we have a bit complex
configuration in Krakow and it will take me some time to fix this
problem. The main reason for this is shared PBS server and Worker Nodes
farm between LCG-1 and CrossGrid Project Testbed...
Piotr
Daniels, T (Trevor) wrote:
>Emanuele, Piotr
>
>I think I see the problem with this now, although I am not very familiar
>with the details of the RB processing. Someone please correct me if this is
>wrong.
>
>The lcgeast MDS (adc0026) at CERN knows of three CEs at Krakow, all hosted
>on zeus02. The same MDS has GlueInformationServiceURL for these three CEs
>set to the site GIIS at zeus24.cyf-kr.edu.pl. However, this site GIIS has
>information about three CEs on the zeus24 host and nothing about zeus02.
>
>When a test job is submitted for a CE on zeus02 (the advertised CEs) the RB
>contacts the site GIIS during rank evaluation for up-to-date information on
>the attributes involved, but as the site GIIS knows nothing about the CEs on
>zeus02 the information it requires is not available, hence the "problems
>during rank evaluation" message.
>
>Piotr, I guess it is down to you to correct this.
>
>Trevor
>.lf n25
>
>Dr Trevor Daniels
>c/o CCLRC eSC Department Phone: (+44)|(0) 1235 778093
>Rutherford Appleton Laboratory Fax: (+44)|(0) 1235 446626
>Chilton, DIDCOT, Oxon, OX11 0QX, UK Email: [log in to unmask]
>The contents of this email are sent in confidence for the use of the
>intended recipient only. If you are not one of the intended recipients do
>not take action on it or show it to anyone else, but return this email to
>the sender and delete your copy of it.
>
>
>-----Original Message-----
>From: Daniels, T (Trevor)
>Sent: Wednesday, December 17, 2003 1:44 PM
>To: [log in to unmask]
>Subject: FW: Job failures at Krakow
>
>
>
>Emanuele, Piotr
>
>The GOC monitor jobs sent to Krakow all fail in the RB with the error
>
>Current Status: Aborted
>Status Reason: Cannot plan: BrokerHelper: All compatible resources are
>unavailable (problems during rank evaluation)
>
>(except when I use the -r option on edg-job-submit to bypass rank
>evaluation)
>
>I see that the MDS does not contain any GlueSARoot entries for Krakow, which
>I suspect is the reason for this failure, since this appears from the RB log
>to be the next attempted action.
>
>Whose responsibility is it to fix this and how?
>
>Trevor
>.lf n25
>
>Dr Trevor Daniels
>c/o CCLRC eSC Department Phone: (+44)|(0) 1235 778093
>Rutherford Appleton Laboratory Fax: (+44)|(0) 1235 446626
>Chilton, DIDCOT, Oxon, OX11 0QX, UK Email: [log in to unmask]
>The contents of this email are sent in confidence for the use of the
>intended recipient only. If you are not one of the intended recipients do
>not take action on it or show it to anyone else, but return this email to
>the sender and delete your copy of it.
>
>
>-----Original Message-----
>From: Daniels, T (Trevor)
>Sent: Monday, December 15, 2003 2:32 PM
>To: [log in to unmask]
>Subject: Job failures at Krakow
>
>
>
>Emanuele (et al)
>
>I've looked more closely at the monitor job failures at Krakow (which I
>failed to mention in the GOC report this morning) and which have now
>persisted for some weeks. The failure appears to be due to a failure of
>edg-job-list-match to find any matching CEs. Here's the details:
>
>With this jdl:
>
>Executable = "/usr/bin/wget";
>Arguments =
>"http://esc.dl.ac.uk/gppmonWorld/gppmon-ack.cgi?fH5pvUWvqL7n62EV5WuaBVFBH1aJ
>41St.CERNRB.Krakow";
>Requirements = other.GlueCEInfoHostName == "zeus02.cyf-kr.edu.pl";
>StdOutput = "monitor.out";
>StdError = "monitor.err";
>OutputSandbox = {"monitor.out","monitor.err"};
>
>edg-job-list-match returns:
>
> Connecting to host lxshare0380.cern.ch, port 7772
>
>===================== edg-job-list-match failure ======================
> No Computing Element matching your job requirements has been found!
> Try again modifying your job description file.
>======================================================================
>
>Yet I can see nothing wrong with the published info in MDS.
>
>Emanuele - any ideas?
>
>Trevor
>.lf n25
>
>Dr Trevor Daniels
>c/o CCLRC eSC Department Phone: (+44)|(0) 1235 778093
>Rutherford Appleton Laboratory Fax: (+44)|(0) 1235 446626
>Chilton, DIDCOT, Oxon, OX11 0QX, UK Email: [log in to unmask]
>The contents of this email are sent in confidence for the use of the
>intended recipient only. If you are not one of the intended recipients do
>not take action on it or show it to anyone else, but return this email to
>the sender and delete your copy of it.
>
>
>
|