From btiwari at ceh.ac.uk Fri Jun 4 10:49:56 2004 From: btiwari at ceh.ac.uk (Bela Tiwari) Date: Fri, 04 Jun 2004 15:49:56 +0100 Subject: [Bio-Linux] How to get full functionality from R and BioConductor on a multi-user Bio-Linux system Message-ID: Hello As some of you, or your users may have noticed, fucntions in the R and BioConductor packages sometimes need to download other packages and files and by default will try to put these in /usr/lib/R/library, which is only writable by root. There is an easy way to make these functions download to directories writable by the user and we have written information about this on the following web page: http://envgen.nox.ac.uk/envgen/software/archives/000455.html The basic information on this topic is copied below, but there is more detail on the web page, including further details about how and why this works. Please note that the line referred to as being in /usr/software/bioenvrc in the instructions below is only present in the copies of that file obtained from us from now on - it won't be in any of your files at this point! cheers Bela The recommended way to set up Bio-Linux to allow users to get the full functionality of R and BioConductor involves two steps: 1) Create a directory the user can write to. For example, the user could type: mkdir ~/myRlib 2) Set the R_LIBS environmental variable to this directory: This can be done by a user manually each time they want to use R, by typing: setenv R_LIBS ~/myRlib or it can be set automatically each time the user logs in; there are a variety of ways to do this. On Bio-Linux, you could set this variable in your /usr/software/bioenvrc file. The most recent images of Bio-Linux include this variable in that file - if you remove the hash in front of the line, the variable will be set when any user logs in. Please note - make sure that if you unhash the line in /usr/software/bioenvrc that the directory name listed there is the same as the name you have chosen for the directory R will be writing to. ************************* Dr. Bela Tiwari Lead Bioinformatician CEH Oxford Mansfield Road Oxford, OX1 3SR 01865 281975 ************************* From dswan at ceh.ac.uk Wed Jun 30 11:37:56 2004 From: dswan at ceh.ac.uk (Dan Swan) Date: Wed, 30 Jun 2004 16:37:56 +0100 Subject: [Bio-Linux] Request for information regarding updateclient upgrades from 2.X to 3.0 Message-ID: <40E2DE54.2080208@ceh.ac.uk> Dear All, We have received a bug report from Martin Ostrowski (Warwick Uni) of a situation which may affect people who have upgrades from Bio-Linux 2.X to 3.0 using "updateclient". If you have upgraded your system using this method (as opposed to a clean install of version 3.0 over a 2.X installation) can you please take 2 minutes to check the output of the following command: yum check-update | grep Server What I need is the output of this command, specifically whether this reports it is using a Red Hat 9.0 base or Red Hat 7.3 base. Please contact me off list with your results to avoid clogging peoples inboxes! This would be extremely helpful for me to determine whether this is an isolated problem or something that I need to distribute a workaround for. Many thanks in advance, Dan -- Dr Dan Swan - Bio-Linux Developer | RHCE EGTDC, CEH, Mansfield Road, Oxford, OX1 3SR Tel: 01865 281 658 Fax: 01865 281 696 http://envgen.nox.ac.uk/ | dswan at ceh.ac.uk