[R] RPM-style install (SLED 10.1)
Stas Kolenikov
skolenik at gmail.com
Tue May 13 01:15:52 CEST 2008
That's what I ended up doing. I've had complaints about blas, too, but
I rpm-ed them out with --nodeps. I've compiled R from the sources (I
had to add some -devel- packages, too), and then copied over the
stats.*.so library from my fully successful 2.6.2 compilation to
semi-successful 2.7.0 installation. So I cannot actually tell if I
have 2.7.0 or 2.6.2 on this computer... and I don't know if any other
problems may appear later.
And yes I have a ThinkPad although it is 32 bit.
On 5/12/08, Horace Tso <Horace.Tso at pgn.com> wrote:
> Stas, this doesn't solve your problem but may shed some light on what might have gone wrong. Just recently I installed R 2.7.0 under openSuse 10.3 in a brand new 64-bit Lenovo ThinkPad. The first install failed because of a dependency error just like yours. It complained it couldn't find BLAS and libgfortran. I then got on the varous repositories and installed a couple flavours (don't remember how many) of fortran and finally it went through. Right now, in my /usr/lib64, i have libgfortran.so.2.0.0.
>
>
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Stas Kolenikov
> Sent: Monday, May 12, 2008 7:33 AM
> To: r-help at r-project.org
> Subject: [R] RPM-style install (SLED 10.1)
>
> I am trying to install R on a SLED 10.1 machine.
> R-base-2.7.0-7.1-i586.rpm fails with
>
> stas at linux-6b8s:~/RPMs> rpm -Uvh R-base-2.7.0-7.1.i586.rpm
> warning: R-base-2.7.0-7.1.i586.rpm: Header V3 DSA signature: NOKEY,
> key ID 14ec5930
> error: Failed dependencies:
> libgfortran.so.1 is needed by R-base-2.7.0-7.1.i586
>
> I tried to trick it into believing there's the library by setting up the link:
>
> stas at linux-6b8s:~/RPMs> ls -l /usr/lib/libgf*
> lrwxrwxrwx 1 root root 29 2008-05-08 17:45
> /usr/lib/libgfortran.so.1 -> /usr/lib/libgfortran.so.3.0.0
> lrwxrwxrwx 1 root root 20 2008-05-08 17:37
> /usr/lib/libgfortran.so.3 -> libgfortran.so.3.0.0
> -rwxr-xr-x 1 root root 2251139 2008-05-01 09:43 /usr/lib/libgfortran.so.3.0.0
>
> but that did not work, either. I went on and installed GCC's gfortran,
> but it did not provide libgfortran.so.1 either:
>
> stas at linux-6b8s:~/RPMs> ls -l /usr/irun/lib/libgf*
> -rw-r--r-- 1 1005 1011 4330512 2008-03-02 02:29 /usr/irun/lib/libgfortran.a
> -rwxr-xr-x 1 1005 1011 1009 2008-03-02 02:29 /usr/irun/lib/libgfortran.la
> -rwxr-xr-x 1 1005 1011 2509309 2008-03-02 02:29 /usr/irun/lib/libgfortran.so
> -rwxr-xr-x 1 1005 1011 2509309 2008-03-02 02:29 /usr/irun/lib/libgfortran.so.3
> -rwxr-xr-x 1 1005 1011 2509309 2008-03-02 02:29
> /usr/irun/lib/libgfortran.so.3.0.0
>
> Any reason R wants that package instead of the newer one? Of course
> rpm --nodeps was an option, but then it fails to load r-stats with the
> same message about the missing library:
>
> Error in dyn.load(file, DLLpath = DLLpath, ...) :
> unable to load shared library '/usr/lib/R/library/stats/libs/stats.so':
> /usr/lib/R/library/stats/libs/stats.so: undefined symbol: _gfortran_pow_r8_i4
> During startup - Warning message:
> package stats in options("defaultPackages") was not found
>
> I am stuck... I am trying to compile R from 2.6.2 sources in the
> meantime, but that's painful for a Linux newbie like me on a brand-new
> computer that does not necessarily have all the packages. I almost
> want to switch to Ubuntu as R installation back there was a single
> click :))
>
> --
> Stas Kolenikov, also found at http://stas.kolenikov.name
> Small print: Please do not reply to my Gmail address as I don't check
> it regularly.
>
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
--
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: Please do not reply to my Gmail address as I don't check
it regularly.
More information about the R-help
mailing list