Thanks Chris.
All
This is a list which will be used to generate discussion at the EGEE
operations workshop next week and to be more effective we need you to
help set the priorities. It is already self-evident that they are all
important... but if we want things addressed in what order?
If I could create a web form quickly for voting I would but today we'll
try to do this by email. Each person on this list is asked to allocate 5
points across the list with more points going to the item THEY consider
of most priority. Thus if you consider something particularly important
put all five against that item. Otherwise select your top choices and
allocate accordingly. You can introduce a new item but it won't get the
top spot now!
The sum of the numbers in () can not exceed 5. All counters are set to
zero in the list below.
Vote against this list by putting a number in the ()
****************************************************
o Local Pool Accounts (0)
o Group quotas on Storage Elements (0)
o Logging: (0)
- Lack of Standard format, interface or location
- Noise
o Sensible passing of Job parameters to LRMS (0)
o Support for heterogeneous clusters (0)
o Poor standard of Error messages (0)
o Lack of fail-over in the user tools, lcg-* commands with
multiple replicas and edg-job-submit with multiple RBs (0)
o Non Portable code or complete lack of source code (0)
o Use of custom code where standard OSS tools are available
with similar functionality (0)
o Lack of remote service status calls (0)
o Lack of remote service management calls (0)
o NEW: Your suggestion (0)
Please send replies directly to Chris (sorry Chris!
[log in to unmask]). We will be checking to see who has (not) given
their feedback but at least one response from each site is expected.
Thanks,
Jeremy
> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:TB-
> [log in to unmask]] On Behalf Of Brew, CAJ (Chris)
> Sent: 15 June 2006 14:31
> To: [log in to unmask]
> Subject: Re: Top 5 Things to change in the LCG Middleware?
>
> Hi All,
>
> Alessandra was looking for a summary to take to the Operations meeting
> next week:
>
> So here's my paraphrasing or the input so far (in no particular order)
> If you disagree with any of it (or I've missed anything off) let me
> know.
>
> Chris.
>
> o Local Pool Accounts
> o Group quotas on Storage Elements
> o Logging:
> - Lack of Standard format, interface or location
> - Noise
> o Sensible passing of Job parameters to LRMS
> o Support for heterogeneous clusters
> o Poor standard of Error messages
> o Lack of fail-over in the user tools, lcg-* commands with
> multiple replicas and edg-job-submit with multiple RBs
> o Non Portable code or complete lack of source code
> o Use of custom code where standard OSS tools are available
> with similar functionality
> o Lack of remote service status calls
> o Lack of remote service management calls
|