​Fermilab will support EDUROAM (starting May 16) to gain access to the Fermilab network and you can use your Fermilab Services account to log in to the EDUROAM network at remote institutions


FIFE Workshop will be held at Fermilab on June 20-21 (Readytalk will be available)


FIFEMON has cool new monitoring (use services uname/pw)


LArSoft Usability Workshop is on June 22-23


Distributed Computing Access with Federated Identities (DCAFI) DUNE transition will be June 27 – July 11 : job submission without using Kerberos

general, certificates, gridFermilab7/25/2016

​Fermilab Facility outages can be found here http://fop.fnal.gov/outages.html


​KCA certificates will be officially decommissioned September 30. Moving to using CILogon certificates. (Digicert is going away too!!)


​FYI: CILogon certificate DNs remove your middle initial.


​How do I open a ticket when I have a problem with Fermilab related items? Go to servicedesk.fnal.gov and the Scientific Computing Services selection on the left hand side.


​AFS – FNAL Computing working on deploying basically empty .bashrc and .bash_profile files


​TotalView debugger being replaced by Allinea Forge debugger. Documentation coming.


​ If you need a CERN account – see Maxine


​Use dCache for storing/reading big files – eventually blue arc will not be available from the worker nodes.


​Persistent (analysis) dCache pool files do NOT end up on tape. These files are experiment controlled and will use pool space until explicitly deleted.  This area does fill up, we may need quotas.


​Scratch dCache pool files do NOT end up on tape. dCache tries to keep scratch files for at least 30 days.  Least recently used file is deleted first. No quotas.


​Read/Write dCache pool files DO end up on tape. Least recently used file is deleted first (usually there is no to very little space available here)


​dCache ‘mv’ gotcha – if you ’mv’ a file from one pool type to another, it will retain the behavior of the origianl pool type. Use ‘cp’ in order to change file types.


​There is more dCache monitoring then you can shake a stick at.


​Use the grid file systems as follows –

  • /dune/app : mounted on the gpgrid nodes, but not on OSG, code development and installation – WILL BE IMPLEMENTING 100 GB/user QUOTA END of MAY 2016
  • /dune/data/users : interactive data files (not mounted on grid worker nodes : use ifdh cp on worker nodes to copy files to here)
  • /dune/data2/users : interactive data files (not mounted on grid worker nodes : use ifdh cp on worker nodes to copy files to here)
  • /pnfs/dune/scratch/users : dCache data files
  • /pnfs/dune/persistent/users : dCache data files

​DELETE your files before you leave DUNE/no longer have access to these areas


​Jobs that run past their resource limits get held, not killed (do jobsub_q to see reason why). Understand why, kill jobs and resubmit.

grid jobsFermilab

​How do priorities affect grid job startup?

  • Lower priority numbers are better.  Assuming resource matching. Lowest priority wins. Exponential relaxation time.
grid jobsFermilab

​How are resources determined between experiments? Jobs running at a given moment – tune priority factors, and relaxation factor)

  • Priority factor of accounting group: per user
    • How many jobs are running, how many running recently (past 1.5 days approx.)
    • Only changed when user asks.  Default to start with is same for all.
  • Relaxation factor : is global
grid jobsFermilab

​Are priority numbers comparable across experiments?

  • If two different exps have same priority (a tie) – there is a tie-breaking mechanism. Otherwise are the same.
grid jobsFermilab

​The LBNE Dropbox is dead!! Long live the DUNE Dropbox!! (used for uploading files to enstore and registering with SAM)


​You are more likely to get a grid slot for your job if you request the lifetime you actually need. This will be enforced in early summer 2016.

grid jobsFermilab

​Remember : DUNE resources are for DUNE work and not other experiments.


​We have a 16-core build node, dunebuild01


​When you submit jobs and use the --memory option you can give units in both MB and GB. jobsub_submit interprets 1 GB as 1024 MB, not 1000 MB. So --memory=2GB is equivalent to --memory=2048MB, not 2000MB.

grid jobsFermilab

​fermiprint.fnal.gov is unavailable to computers on the eduroam
wireless network.  host fermiprint.fnal.gov returns ";; connection timed out; no servers could be reached" on a mac on eduroam, and Firefox emitted "Server not found -- Firefox can't find the server at fermiprint.fnal.gov"  printing may still work, but users cannot add printers.


​Using touch to modifiy the date stamps on files in the dCache
scratch area is not enough to update the
use date.  I think you have to open the files but we can test this,
though a test will take a while (1 month).