Re: Pausing RPM installation for user input

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

 



Panu Matilainen wrote:
> On Wed, 2005-06-22 at 13:09 -0600, Bart Whiteley wrote:
> 
>>On Wednesday 22 June 2005 12:21 pm, Dan Trainor wrote:
>>
>>>This software will not be released into the wild.  It will be kept
>>>strictly internal to the company which I work for.
>>
>>Based on this, and the fact that you seem to grasp all the issues, I say 
>>leverage the RPM tool any way you please. 
> 
> 
> Of course people are free to abuse the tools however they please. Abuse,
> because "configure package interactively from rpm %post" goes right
> against rpm design and purpose, even if there are tricks to get around
> it.
> 
> 
>>The only other danger is that someone who doesn't understand the issues, and 
>>is building RPMs that may be "released into the wild", may see your RPMs and 
>>copy the techniques you use. 
>>
>>I've seen cases where someone builds a "bad" RPM, and later other teams use 
>>this bad RPM as their template and soon there are dozens of broken RPMs.  The 
>>original packager may have understood the issues, and consciously made the 
>>RPM "bad" -- like you want to, but others just copy what they assume to be a 
>>good RPM. 
>>
>>So maybe put a big disclaimer in your .spec. ;)
> 
> 
> Yep, rpm's (and practices used in them) have a funny way of finding
> their way to places you never expected them to be. Including things like
> somebody wants to install your package directly during initial system
> installation (eg kickstart) or such. Going around the "rpm filosophy" is
> certainly possible as seen in all sorts of commercial vendor packagings.
> The common thing to them all is that they end up biting you one way or
> the other.
> 
> 	- Panu -

Thanks for your responses, guys -

It's not like I don't understand the implications that my request may
have.  I understand what it could (and maybe would) break.  I really do.
 I appreciate you guys reaffirming to me what is right and what is
wrong, but that's not really what I need ;)

I guess whatever I do, it will involve %post, since I've been told time
and time again that RPM, by design (and otherwise I would agree with
this, if not for this particular project) was designed to be a
completely automated process, requiring no outside input other than the
command issued to actually install the RPM.

Again, I thank you all for your time.  I'm quite sure at this point that
%post is where I'll need to spend the majority of my time here.

Thanks
-dant


[Index of Archives]     [RPM Ecosystem]     [Linux Kernel]     [Red Hat Install]     [PAM]     [Red Hat Watch]     [Red Hat Development]     [Red Hat]     [Gimp]     [Yosemite News]     [IETF Discussion]

  Powered by Linux