Last modified 11 years ago Last modified on 03/24/10 16:40:41

[part of a series of meetings]


Timur, Gene, Vejay, Dmitry; Tanya, Jan, Paul, Owen, Patrick, Tigran


[see box to the side]


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 :)

Tanja: managing tickets, pool work

Jan: getting stuff on webadmin; "release watch" work.

Paul: RT tickets, reviewing code, fixing bugs

Owen: ETICS process, dcap release process; (+Jan) SRM client ... waiting on UI/VO-box/WN-box release

Patrick: Vacation for three days; Amsterdam LHCb Tier-1 meeting, convincing them not to use xrootd .. successful so far. Also been in Asia, giving a talk "what a modern storage-element should look like."

Tigran: lot of wine in Africa; removing emails.

Timur: mostly time working on propagating complete VOMS info to space manager. People have membership in VO group but cannot use that membership to authorise their space-manager related operations. Hope to have this in 1.9.5. Another patch to propagate this from doors to space-manager (for implicit space res.). This can only go to trunk, as it modifies ProtocolInfo?, requiring a complete upgrade.

Vijay: updating the useFileSystem patch for cleaner; updated the script that checks whether files are being flushed to tape, supports Chimera and PNFS now. Pushing work to enstore

Dmitry: working on script to recover problems with dCache. Looking at Gerd's patch for space-manager; look at modifying SQL query to; looking at patch from Timur. Rest of time spent on CDF dCache instance.

Irina: nothing special .. changed OS, so still working on beating it into submission.

Status of work for 1.9.5

A (quick?) review of activity needed for the 1.9.5 release

Fix for the checksum: three tickets addressed.

Decision to make a release asap.

  • Owen : need to look at info system and

Work needed on the dCacheConfigure scripts ... an issue where it's not working properly for configuring the info-provider. This needs to be fixed or documented. Will affect 1.9.6 and 1.9.7.

This is not blocking future releases of 1.9.5

Space manager authorisation

The problem affects people who make or release reservations. The patch would affect only the SRM domain. Timur can test this.

What is the difference

Currently we're sending only the single pair of VO-name, VO-role; the patch is to send the whole auth. record. The space-manager can then check all items in the authorisation record.

Second, the authorisation decision is currently made in space manager; patch adds a policy interface. The site-admin can configure the authorisation by choosing which policy is to be used.

This is to solve RT 4552 and other tickets.

There was some concern about whether this patch is refactoring the code.

Timur to update the patch to remove the invocation via reflection.

Status of work for 1.9.6

A (quick?) review of activity needed for the 1.9.6 release

Do a release "soon", after 1.9.5

Status of work on 1.9.7

A (quick?) review of activity needed for the 1.9.7 release

Same changes in 1.9.7 as 1.9.6.

Status of work for Trunk (a.k.a future 1.9.8)

Tigran: committed a patch. Not required poolManager.conf file to start dCache. If you don't have this file then dCache will start in a sane defaults: all pools accessible and treated equally.

If you update the RPM first time then poolManager.conf then this file will be renamed to poolManager.conf.rpmsave. People felt this was rather too disruptive; Tigran couldn't find a way to configure RPM to prevent this behaviour.

We should discussion this over the mailing list.


Paul explained stuff ... Timur happy.

Owen: suggest listing options that are declared but not used; e.g., warning site-admins.

Jan: do we need the third annotation?

Consensus is to go with the third annotation; Paul to submit patch to RB.

Info system

Would like to see information about pools to come from the pools themselves not PoolManager.

Issues from yesterday's Tier-1 meeting


SRM ls availability: should we change the criteria for a file being considered UNAVAILABLE?

Fix this in trunk.


PIC are still waiting for a recovery procedure for their SpaceManager database.

Roughly 60% done: have script that extracts the information. Need to add support for injecting information into space-manager database.


Meeting up tomorrow (Thursday) at the same time 15:30 CET.

The meeting may take more than one hour, depending on how much discussion is needed.

Outstanding RT Tickets

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

Review of RB requests


Same time, next week.