All, Dates:Thursday-Apri 7, 2005 Time:1:00PM- 3:00PM Number of Participants:(12) Number to Call: 617-324-7520 (*** CHECK NEW EMAIL TO CONFIRM NUMBER ***) Agenda =================================== 1.) Status of Access to QCDOC 1K Partitions for James and Dru. (Chulwoo scripts to run CPS Test Codes) 2.) Status of MILC Asqtad HMC Production code. (Dru and Carleton) 3.) Status of Chroma DW HMC Production code. (James and Robert) 4.) Et al. Cheers, Rich ---- Prof. Richard C. Brower Physics Department --Boston University 590 Commonwealth Ave. Boston, MA 02215 Tele: 617 353 6052 (BU) 617 253 2599 (MIT) 617 497 8296 (HOME) Fax: 617 353 9393 Email: brower@bu.edu brower@lns.mit.edu From brower@lns.mit.edu Thu Mar 24 13:02:35 2005 Date: Thu, 24 Mar 2005 13:02:24 -0500 From: Richard Brower To: detar@physics.utah.edu, djholm@fnal.gov, edwards@jlab.org, watson@jlab.org, rdm@phys.columbia.edu, brower@lns.mit.edu, avp@lns.mit.edu, simone@fnal.gov, petrov@physics.columbia.edu, mdp@fnal.gov, dgr@jlab.org, hneff@bu.edu, jhetrick@uop.edu, gregory@physics.arizona.edu, orginos@lns.mit.edu, chulwoo@phys.columbia.edu, kilcup@physics.ohio-state.edu, schlagel@bnl.gov, josborn@buphy.bu.edu, dru@physics.arizona.edu, zhang8@cs.uiuc.edu, maresca@physics.utah.edu, sg@indiana.edu, blum@bnl.gov, stratos@bnl.gov Subject: REMINDER SOFTWARE CALL NOW!!!! OOPS! This Email didn't get out! All, Short call tomorrow (from KITP), Dates:Thursday-March 17, 2005 Time:1:00PM- 3:00PM Number of Participants:(12) Number to Call: 617-324-7520 Kerberos: Eric Blum has asked me to get input from the Software Committee on the "immediate needs" for file transfer into and out of BNL. In addition I asked him to report on the situation at BNL. I have attached a note from Eric on overall issues at BNL. I would like Eric, Chip and Don to work on a common solution if at all possible and a game plan of what can be done when. Rich Agenda =================================== 1.) Road Map to Kerberos: (Eric, Chip and Don) 2.) Progress on Asqtad Level 3 tests/interface to MILC? (Dru, Chulwoo and Carleton) 3.) Progress for Chroma on QCDOC (James and Robert) 4.) What are the actual disk storage requirement for LHPC Structure function at the USQCD? (Future discussion) Cheers, Rich - -- Prof. Richard C. Brower Physics Department --Boston University 590 Commonwealth Ave. Boston, MA 02215 Tele: 617 353 6052 (BU) 617 253 2599 (MIT) 617 497 8296 (HOME) Fax: 617 353 9393 Email: brower@bu.edu brower@lns.mit.edu ================= Message From Eric Blum On Kerberos ========================= There may not be a problem installing client software at BNL as long as it does not permit incoming connections and will not require us to install a Kerberos server. We are in the process of configuring the ssh gateways that will be used by qcdoc and may be able to add the Kerberos client to them. As of today, we are using the regular BNL ssh gateways and we will not be able to change their configuration easily. To go further we need to do the following: 1. The software committee needs to specify what protocols are required for data transfer. The transfer mechanism should describe a process that is initiated by someone already logged into BNL. Incoming data transfer to BNL from someone who is not logged in will require sftp or scp. 2. BNL cyber security will need to evaluate the proposed mechanism. 3. BNL needs to complete the configuration of the dedicated QCDOC SSH gateway. The new gateways, by the way, will be load balanced using some specialized Cisco hardware and will share about a TB of network attached storage. They will be very fast and should easily handle large data transfers. All this new hardware needs to be tested before we start using it. 4. The QCDOC hosts will be moved from their current IP addresses to a new subnet using the SSH gateways. The QCDOC users will then have access to QCDOC but not the rest of the BNL campus. 5. With all this completed, we may be able to install a Kerberos client. 6. Then, and only then, can we begin to consider some kind of unified login mechanism with the other sites, whether Fermilab Kerberos, or something else. ------- End of forwarded message -------