XCache DevOps Meeting Oct 03, 2019

US/Central
Brian Bockelman, Derek Weitzel (University of Nebraska - Lincoln), Marian Zvada (US CMS)
Description

Time: 11AM PT/1PM CT/2PM ET
Where: ZOOM.US

Join from PC, Mac, Linux, iOS or Android: https://unl.zoom.us/j/651969661

Or iPhone one-tap :
    US: +16699006833,,651969661#  or +14086380968,,651969661#
Or Telephone:
    Dial(for higher quality, dial a number based on your current location):
        US: +1 669 900 6833  or +1 408 638 0968  or +1 646 876 9923
    Meeting ID: 651 969 661
    International numbers available: https://unl.zoom.us/zoomconference?m=wxCNSgMZiA-cVKSNowGYlQ

Attendees: Marian, Derek, Andy, Edgar, Mat, Matevz, Riccardo, JohnH

News
----
* Welcome Riccardo in XCache team! He is new fellow at CERN hired by Simone Campana to stay in program for 2-3 years in collaboration of ATLAS and CMS needs for ESCAPE project, for their use case XCache is a suitable technology

XRootD development for XCache
-----------------------------
* Data integrity: ongoing discussions on resolution of corrupt data in the cache, during IRIS-HEP meeting it's been discussed TLS is the way we do to insure that corrupted data do not go into cache, Andy says it's overkill while Brian said is cheap, Andy and the devs are looking at alternatives and assessment why that would work better than TLS
* TLS is available in v5 so scitokens can be delivered in time, there are currently two competing token technologies - scitokens and macaroons

ATLAS XCache deployment
-----------------------
* nothing to report

CMS XCache deployment
---------------------
* meeting with Caltech for SoCal caches for miniAOD, all is approved by CMS offline and computing management go forward, there are cases of oversubscribing the disk space and they really want to utilize caches
* newest images used from OSG to deploy cache for CMS
* something is fishy, 7.5TB of data cache writes shows 1.6TB according to Diego's analytics of file accesses, need deeper investigation (correct me if I got this wrong, Edgar)
* some users are not going via CMSSW
* discussion basically for reading nano CMSSW is not used, in order to push users do not do direct access but via CMS suggested solution is have a library that reads the TFC of a site and spits out the NANO location

OSG StashCache deployment
-------------------------
* Houston node installed and connected to 10Gig connectivity, for Sunnyvale problem with rack rails, Amsterdam node all is good just waiting for magic key get through customs

AOB
---
* two pods shutdown to run at Kansas City and another at UCSD due security incident reported, Edgar takes ownership of this, accidentally stored certs in the exported area that has been compromised, incident report filed by security team: INCIDENT 2019-10-03_001

There are minutes attached to this event. Show them.
    • 13:00 13:05
      News
      Conveners: Brian Bockelman, Dr Derek Weitzel (University of Nebraska - Lincoln), Mr Marian Zvada (US CMS)
    • 13:05 13:15
      XRootD development for XCache

      General status of development, XRootD releases with new cache features, bug fixes, etc...

      Conveners: Mr Andrew Hanushevsky (SLAC National Accelerator Laboratory), Matevz Tadel (UCSD), Wei Yang (SLAC)
    • 13:15 13:25
      ATLAS XCache deployment
      Conveners: Dr Ilija Vukotic (LAL), Wei Yang (SLAC)
    • 13:25 13:35
      CMS XCache deployment
      Conveners: Diego Ciangottini (Universita e INFN, Perugia (IT)), Prof. Frank Wuerthwein (UCSD)
    • 13:35 13:45
      OSG StashCache deployment

      Status of OSG deployment, origin/caches operations, stashcp, usage, etc...

      Conveners: Brian Bockelman, Brian Lin (University of Wisconsin-Madison), Dr Derek Weitzel (University of Nebraska - Lincoln), Mr Edgar Fajardo Hernandez (UCSD), John Hicks (Internet2), Mr John Thiltges (University of Nebraska - Lincoln), Mr Marian Zvada (US CMS), Mr Matyas Selmeci (University of Wisconsin-Madison)
    • 13:45 14:00
      AOB