[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: List volume

Marc Wallman (Marc Wallman ndsu nodak edu) wrote:
> Justin Crabtree wrote:
> > It is my understanding that RHEL4 is based on FC3 and that any
> > bugs/patches/updates/whatever fixed in FC3 will make their way into
> > RHEL4.  Assuming this assumption is correct, would it not be a good idea
> > to test with FC3 as well as RHEL4?  Also, will it work the other way?
> > Will fixes to RHEL4 make their way into FC3?
> >
> > We are currently using RHEL3 on most of our production boxes and for
> > certain applications (LDAP for one) the packages in RHEL3 are very
> > outdated.  We will probably wait till FC3 becomes final and make a
> > decision then whether to upgrade or wait for RHEL4.  Any opinions?
> I don't see FC3 and RHEL4 as comparable solutions, mainly due to the
> huge difference in lifecycle between the 2 products. FC3 will go out of
> support in about a year and you will have to upgrade/reinstall your
> servers. This short lifecycle is exactly why we are using RHEL which has
> a MUCH longer lifecycle.
> On a note relating to the original post, I have been running RHEL 4beta1
> since about the 1st of October and have had very few problems with it.
> --
> Marc Wallman
> NDSU Information Technology Services

I also enjoy the long lifecycle of RHEL because it means once we decide on
something as standard, I can rely on the fact that it WILL be standard for the
length of the lifecycle. Since we moved to 99% RHEL for our Linux boxes,
management has been much simpler and solutions to common problems last longer.
I'm no longer having to track down multiple issues in multiple operating
systems. I think the real reason I'm sticking with RHEL4 for testing is
because I'm biased, Fedora just isn't my thing :)

We are also using LDAP on RHEL3 and the reason I'm interested in RHEL4 is for
a total authentication, directory and file system solution. Since RHEL3 has no
support for NFSv4, I can't integrate Kerberos and File Systems across the
network right now. Also, the older versions of OpenLDAP and MIT Kerberos in
RHEL3 as are bit of a hold back.

I'm already working on OpenLDAP + MIT Kerberos + Cyrus SASL + NFSv4 as the
solution I'd like to move to (from OpenLDAP + NFSv3) so the RHEL4 beta makes
sense as the testing platform of choice since that is the platform I'll be
using in production if this idea ever goes live. Also, like I said above, I'm
biased :)

BTW: If anyone else in interested in the network auth + directory + file
system service thing I mentioned above, I'm writing a paper to cover my
experiences which I plan to announce on this list some time in the near
future. Anyone interested?

Just some thoughts.

Ryan Thomson, Systems Administrator
University Of Calgary, Biocomputing
Phone: (403) 220-2264
Email: thomsonr ucalgary ca

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]