[Rd] Where does install.R go when R gets compiled? Or, how to experiment with changes to install.R?

Uwe Ligges ligges at statistik.tu-dortmund.de
Mon Feb 22 17:10:50 CET 2010



On 20.02.2010 00:04, Paul Johnson wrote:
> On Thu, Feb 18, 2010 at 2:20 PM, Jens Elkner<jel+r at cs.uni-magdeburg.de>  wrote:
>> On Thu, Feb 18, 2010 at 11:33:14AM -0600, Paul Johnson wrote:
>>> I'm pursuing an experiment to make RPM files for R packages
>>> on-the-fly. Any time I install an R package successfully, I want to
>>> wrap up those files in an RPM.   Basically, the idea is to "hack" an
>>> option similar to --build for R CMD INSTALL.
>>
>> Hmm, why not take the easy way:
>>
>>     clean_dst $PROTO
>>     cd $TMPBUILD
>>     mkdir -p $PROTO/R/library
>>     $R_HOME/bin/R CMD INSTALL -l $PROTO/R/library $TMPBUILD
>>
> Yes, I've been there, done that.
>
> I have to administer this on 60 servers in a cluster.  I don't want to
> rebuild all packages on all systems. If I can figure a way to create
> RPM for them, I can script the RPM installs and then I'm sure all the
> systems are identical.
>
> In the worst case scenario, I just have to copy the library tree from
> one machine to another.  But the RPM approach has a bit more built-in
> error checking.
>
> pj


Paul,

beside the already answered parts: in general I'd try to read from some 
network space so that I'd had to make only 1 installation which is also 
useful for easier upgrades and parallel execution where you need 
identical doftware on all nodes.

Best,
Uwe



More information about the R-devel mailing list