[R] Trouble building R2.1.0 from source on Linux: package VR

Peter Dalgaard p.dalgaard at biostat.ku.dk
Sun Jun 19 23:22:55 CEST 2005


Frank Gibbons <fgibbons at hms.harvard.edu> writes:

> Hi,
> 
> Following on from suggestions made last week, I decided to install R 2.1.0 
> on my Linux machine. I'm running into a problem there however, as shown:
> 
> make[1]: Entering directory 
> `/d0/home/fgibbons/tmp/R2.1.0/R-2.1.0/src/library/Recommended'
> make[2]: Entering directory 
> `/d0/home/fgibbons/tmp/R2.1.0/R-2.1.0/src/library/Recommended'
> begin installing recommended package VR
> WARNING: ignoring environment value of R_HOME
> tar: Skipping to next header
> tar: Archive contains obsolescent base-64 headers
>   incomplete literal tree
> 
> gzip: VR.tgz: invalid compressed data--format violated
> tar: Error exit delayed from previous errors
> ERROR: cannot extract package from 'VR.tgz'
> make[2]: *** [VR.ts] Error 1
> make[2]: Leaving directory 
> `/d0/home/fgibbons/tmp/R2.1.0/R-2.1.0/src/library/Recommended'
> make[1]: *** [recommended-packages] Error 2
> make[1]: Leaving directory 
> `/d0/home/fgibbons/tmp/R2.1.0/R-2.1.0/src/library/Recommended'
> make: *** [stamp-recommended] Error 2
> 
> This is most unusual - I must have built R from source five or six times 
> going back to 1.5.0 and don't recall any problems like this. Does anyone 
> have any suggestions about where I might look for the source of this problem.
> 
> In particular, I'm interested in using package 'vsn', whose installation is 
> (I believe) blocked by this problem. I downloaded the source from CRAN this 
> afternoon, choosing the version 2.1.0 'stable' code. I don't have sysadmin 
> privileges on the build machine, but this has never been a problem before - 
> I just pass --prefix=$HOME to the configure script. FWIW, I believe the 
> sysadmins use Debian.
> 
> This is a pretty time-critical matter for me (I wouldn't have chosen to 
> upgrade now, were it not for my earlier problem with merge when there are 
> empty labels), any assistance greatly appreciated. Thanks in advance.

Well, at this point, you might as well wait for 2.1.1 to come out
tomorrow around noon (CEST). If you have R_HOME in your environment,
you should get rid of it, but I don't see how that could cause the
subsequent errors. My best guess is that the tarball got botched in
the download somehow.

BTW, "Linux" does not suffice for us to identify your distribution
much less the version. I think /etc/issue will tell you in most cases
if you don't know.

-- 
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark          Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)                  FAX: (+45) 35327907




More information about the R-help mailing list