wiki:developers-meeting-20160106
Last modified 3 years ago Last modified on 01/06/16 16:00:29

[part of a series of meetings]

Participants

Anupam, Christian, Gerd, Marina, Olufemi, Tigran, Jürgen, Paul

Agenda

[see box on the right-hand side]

Postcards

Anupam: Working on review requests, WebDav? auth handler, planning for further work together with Paul

Christian: back to DCAP, will submit patch soon. EGI testbed: performance benchmarks looking good.

Gerd: several issues on their system, several Jetty threads busy-waiting. Fixed. GSI delegation in SRM client: JGlobus doesn't deal with split certs which got introduced by a security fix in the JVM. Xrootd pools crash because of OOM problems: Logs point to netty buffer leaks. Collecting further info.

Marina: REST API together with Olufemi. Starting to coordinate integration.

Olufemi: Web interface and REST API work.

Tigran: File system work, modifying nearline storage providers. Will result in user-visible change.

Paul: Lingering patches from RB, code review, starting move of release management to Marina and Jürgen Jürgen: Move to release management, triaging a problem in the SmallFiles? script

Special topics

Release management for SRM client:

Paul suggested getting rid of SRM-Common and moving the functionality elsewhere, argues for clean separation between server and client packages. This needs further discussion about modularization, which will be done among the release management team.

Chunked HTTP upload:

Needed for the OwnCloud? sync client. This would require the client to always get to the same pool. At the moment, only a sub-optimal proxying approach works. So we'll need to implement direct support for chunked PUTs (noting that the OwnCloud? client's configuration seems not to work in this case and we can't really influence the upstream). Tigran suggests workarounds like an nginx server in front of dCache, managing the transfers.

Trunk activity

no significant activity, due to the holidays

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

Reports of an LHCb instance suddenly stopping. Interacting with the requests didn't seem to work. Restarting the PoolManager domain worked. Investigation ongoing.

Another issue: Files in dCache whose apparent directories don't exist any more. See ticket #8864.

A cell communication issue affects one user, causing outages under high load. Investigation ongoing. Paul suggests keepalive messages on the channels to check for such issues.

Plans for patch-releases

We'll try too keep the quick weekly release cycle.

Outstanding Documentation

Outstanding RT Tickets

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

Review of RB requests

New, noteworthy and other business

DTNM

Proposed: same time, next week.