Re: "persistent" RPMs

Yuri Csapo <[email protected]> Tue, 30 Jun 2009 11:02:20 -0600
Newsgroups org.kernel.vger.linux-admin
Message-ID <[email protected]>
Gerardo, thank you for your reply.

I do appreciate the problems of holding something back. While I understand this situation is not 
sustainable in the long term, I don't see why I shouldn't be able to hold back a particular package, 
of course in the process holding back all its required and dependent packages, while letting other 
parts of the system update as usual. Of course eventually we'll reach some shared library that will 
end up locking up upgrades across the whole system but I can make a decision then.

The reason I ask the question is because, although I've been working with Linux since 1996 (and 
other Unices before then), I've managed to stay away from RedHat until about a couple of months ago. 
Now I'm having to learn "the RedHat way" of doing things. I may still be proven wrong but so far my 
feeling is that, frankly, if I have to contend with proprietary poorly documented incantations and 
nonexistent support and still pay for the privilege, I'd rather go to Windows Server. But I digress.

On Ubuntu, you can tell apt to simply 'hold' a package and it will just do that. Does anybody know 
of anything similar on RedHat?

Thanks

Yuri


Gerardo Juarez-Mondragon wrote:
> In my opinion, there are two scenarios:
> 
> (1) the RPM was made by you of an application you wrote yourself or is
> a third party RPM of a very specific application. In this case, being
> an application 'off the distribution tree', it will never be upgraded,
> except when you or the authors write such an upgrade.
> 
> (2) the application is a regular application, registered as part of
> the Linux distribution. In this case the chances of it not being
> upgraded are slim, because most bulk upgrades will find it listed and
> act accordingly. Theoretically, you could manually upgrade parts of
> your distribution omitting this one package. However, it will
> eventually clash with some upgrade affecting a vital library. This
> could happen in case (1) above as well, except for the simplest
> applications, since most use shared libraries; you could get away for
> some time though, by keeping your application using static libraries.
> Again, this last only with scenario (1).
> 
> The only way I have found of making software in this condition work is
> by not upgrading at all. This is the reason why you find some servers
> with an old distribution and they cannot be upgraded, unless they
> break free from the 'untouchable' software.
> 
> Hope this is of use,
> Gerardo
> 
> 
> On Tue, Jun 30, 2009 at 10:57 AM, Yuri Csapo<[email protected]> wrote:
>> Does anybody know how to instal an RPM on a RH-derived system and make it
>> persistent so that future upgrades don't replace them?
>>
>> --
>> Yuri Csapo
>> Academic Computing & Networking
>> Colorado School of Mines
>> CT-256
>> Phone:  (303) 273-3503
>> Fax:      (303) 273-3475
>> Email:   [email protected]
>>
>> Please use the following link to open a service request:
>> http://helpdesk.mines.edu
>> ===========================================
>> With a PC, I always felt limited
>> by the software available.
>> On Unix, I am limited only by my knowledge.
>> --Peter J. Schoenster
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-admin" in
>> the body of a message to [email protected]
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-admin" in
> the body of a message to [email protected]
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Yuri Csapo
Academic Computing & Networking
Colorado School of Mines
CT-256
Phone:  (303) 273-3503
Fax:      (303) 273-3475
Email:   [email protected]

Please use the following link to open a service request:
http://helpdesk.mines.edu
===========================================
With a PC, I always felt limited
by the software available.
On Unix, I am limited only by my knowledge.
--Peter J. Schoenster