XCache DevOps Meeting Aug 06, 2020

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

Time: 11AM PT/1PM CT/2PM ET/8PM CET
Where: ZOOM.US

Join from PC, Mac, Linux, iOS or Android: https://unl.zoom.us/j/651969661
Meeting is password-protected: ask #xcache slack channel or xcache@opensciencegrid.org

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

XrootD Development
	• XRootD 5.0.1-rc1 with patches is available.
	• Fixed x509 certificates are handled on the HTTP side.  Allows certificates generated outside of VOMS
	• Full release in 2 weeks (week of Aug 17th)
	• Expect another 4.12 with backported fixes.
	• (Derek) Gstream? Changes being performed by Andy et. al are only for the cache stats gstream, and are configurable (can be turned on/off).

ATLAS
	• Nothing much new.
	• OSG base image is failing, Mat will follow up.

CMS
	• Working on high availability of the redirector.
		• If using DNS round robin, random number of packets will go to either.  XRootD will look at all addresses in the DNS entry for the redirector.
		• If redirector goes down, client will go to which one is available.
		• XRootD will change behavior
			• Scale out for redirector is not “fantastic”.  LSST is performing 4 million requests at same time.
			• Client DNS handling will change:
				• takes all addresses registered in DNS for an entry
				• shuffles ipv4 and ipv6 addresses, but separatly.
				• uses random shuffle, which is not random.
				• Use same sequence of addresses for every client, because not setting a seed.
				• Therefore, not distributing load correctly.
				• The fix is to use the shuffle, which is properly random.
	• Testing EL8 in caches.  It works well.
		• xrootd-scitokens was broken, but new release just today may fix it.  Edgar is fixing.

OSG
	• Internet2 nodes:
		• Houston sent back to vendor, hardware issue
		• Amsterdam shipment, 3/4 machines have cleared customs.  Hopefully only weeks until online.
	• Validation document next week.
	• (Derek) Will send link to Andy to the python monitoring collector
There are minutes attached to this event. Show them.
    • 13:00 13:05
      News
      Conveners: Dr Derek Weitzel (University of Nebraska - Lincoln), Mr Marian Zvada (US CMS)
    • 13:05 13:10
      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)
    • 13:10 13:15
      XCache issues

      XCache issues experienced during the deployment, configuration or otherwise causing problems in production or development.

    • 13:15 13:20
      ATLAS XCache deployment
      Conveners: Dr Ilija Vukotic (LAL), Wei Yang (SLAC)
    • 13:20 13:25
      CMS XCache deployment
      Conveners: Diego Ciangottini (Universita e INFN, Perugia (IT)), Mr Edgar Fajardo Hernandez (UCSD), Prof. Frank Wuerthwein (UCSD)
    • 13:25 13:35
      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)
    • 13:35 13:40
      XRootD transfers monitoring
      Conveners: Dr Derek Weitzel (University of Nebraska - Lincoln), Diego Davila (CMS), Mr Edgar Fajardo Hernandez (UCSD)
    • 13:40 13:45
      AOB