Hi Chris et al,
> I just wanted to query the SL*4 has "*less-than-desired patch-level*" SL4.9
> will get security updates until Feb 2012 - glite-Mon and lcg-CE should
> certainly be end of life before then.
It is unfortunate that SL4 is promoting minor versions (4.9?) while it
really only patches security holes; rather than system and packages' bugs.
The xargs (an omnipresent unix utility in scripts) case is provided as a bold
example to defend the "less-than-desired patch-level" statement. Since we're
not claiming here that the OS is "insecure", that statement is still correct.
I just wish that, that was the only annoying bug within the SL4 distribution.
We are a bit under unrest while reading statements like the following:
http://linux.web.cern.ch/linux/scientific4/
"Since the original deadline of having a stable SLC5 distribution in October
2005 can not be met, SLC4 will be used in several key areas at LHC startup."
When it was written, it made sense. 5 years down the road, it's so pointless.
Maybe I am picky!
The rest of this thread, went in a direction that someone could expect:
* grid sites have been asked to install components that have been tested
individually, but not as a logical extension to the existing components.
(ie. what is the probability of having to connect cream@SL5 + lcgCE@SL4?)
* grid sites would like to receive a clear upgrade path for the current
infrastructure which is in production and albeit being archaic, just delivers.
(we are respectful to the lcg-CE code; it is only not maintained lately)
In the end of the day, we are sysadmins trying to deliver services to our
clients (users, physicists or else). If we receive garbage, we give garbage.
cheers,
Fotis
--
CSCS Swiss National Supercomputing Centre | Drs/Eng. Fotis Georgatos
Systems Engineer | HPC Co-Location Services
Via Cantonale, Galleria 2 | CH-6928 Manno
[log in to unmask] | www.cscs.ch | Phone +41 91 610 82 80
|