wiki:developers-meeting-20140618
Last modified 5 years ago Last modified on 06/19/14 10:57:16

[part of a series of meetings]

Participants

Paul, Tigran, Karsten, Gerd, Al, Dmitry

Agenda

[see box on the right-hand side]

Postcards

Up to two minutes (uninterrupted) per person where they can answer two questions:

  • What I did last week (since the last meeting),
  • What I plan to do in the next week.

No questions until we get through everyone :)

Karsten: Tickets, Reviews, Helped setting up packing machines at DESY Al: Testing in Replica Manager, put up Design Specs,

Almost done reading through Gerds Patch

Dimitry: Working on tickets: NPE and fixed another one. Fixed SRMClient ArrayOOB Error

Debugged some

Gerd: Attending Board Meeting, Networking, File Corruption Debugging,

Rewrite of ActiveTransferPage?

=> Wicket is not that bad

Paul: Mostly getting demo to work, then doing the demo and now getting dCache release out

Tigran: Bakathon, started working on JGlobus Code, fixed Unit Tests, doing further cleanup

Working on NFS to implement callbacks, trying to avoid multiple writers which causes inconsistencies in the namespace. Merged some stuff into official release.

Plans for patch-releases

Should we make a new patch release?

In progress. Will get those out asap.

Next round of releases will be next Wednesday.

Will move to a new release machine soon.

Should we backport IPv6 changes?

We should clarify if they plan to use it in production or rather for testing for now.

We are expecting people to upgrade to 2.10 which includes IPv6 support.

For the IPv6 support doors and pools have to be updates for the redirecting.

=> We will not backport except if really good reasons turn up.

Trunk activity

LSDMA Demonstrator Meeting

Paul: It mostly worked. Found some bugs or differences between Globus and dCache view. For us this means:

-We need to support md5sum check during transfer -md5 checksum per default

Open Questions:

-What is GlobusOnline? expecting to happen is ABOR is called -What transfer should be aborted if there are multiple in progress.

Will talk to GO and clarify things.

Hopefully we can have a production level IdP and the test IdP running at DESY

Side remark: There were 2 probems: Globus did not realize if a transfer was successful.

There happened a error during transfer, ...

those were server side problems.

NFS Testing

Tigran: We didn't find many bugs. Tested RHEL6/7 and upstream kernel. Worked with MS ActiveDirectory? Mostly polishing work.

JAAS configuration was a pain, improved that. There is another company PrimaryData?) building up a NFS41 product that will come out at fall.

Other

Gerd's working on some simple Wicket improvements

Issues from [FIXME: Add link to yesterday's Tier-1 meeting]

Marc, Ulf, Xavier

Marc: Migrated to 2.6, 2 tickets, problem with + in Uri is fixed

Some more problems remotely related to dCache

Ulf: Hardware problems on pools, exchanging disks caused pools to act weird.

Gerd: It seemed that when copying the files, the checksum matched, but

later on the checksum did not match anymore. Still needs more investigation

Paul: Would be interesting to do a binary diff Gerd: Right, but we are trying again to see if we can reproduce

Upgraded tape pool to 2.9, next will be special driver for

Xavier: Everything fine so far, running 2.6.29. The problem with dcap seems to be

solved. They have one file that lost location info on tape.

Issues from EMI

New or noteworthy

Outstanding RT Tickets

[This is an auto-generated item. Don't add items here directly]

Review of RB requests

DTNM

Proposed: same time, next week.