[R] Help with R and MySQL

Jeffrey Horner jeff.horner at vanderbilt.edu
Mon Feb 23 16:10:02 CET 2009


Aaron Barzilai wrote:
> Hello,
> 
> This forum has been very helpful to me in the past, and I've run out of ideas on how to solve my problem.
> 
> I had been using R and MySQL (and Perl) together for quite some time successfully on my Windows XP machine. However, I recently had some problems with MySQL (the ibdata file had become 35GB on my hard drive, turns out it's a known bug with InnoDB), and ultimately the way I fixed my problem with MySQL was to upgrade it.  It's working fine now, I can use MySQL however I'd like.  I'm sticking to MyISAM tables for now, though.
> 
> However, I had set up my system so I did a linear regression in R. Originally, this was done in R 2.5.0, I would load in the tables from MySQL to R and then conduct the regression in R.  However, after solving my MySQL problem, I ran into a strange error in R (and DBI/RMySQL).  R connected to the database just fine, and I could even show the tables in the database and load two of them into R.  However, the tables I loaded successfully were only a single column.  Every time I tried to load in a recordset that was multiple columns, I got a relatively nondescript Windows error("R for Windows terminal front-end has encountered a problem and needs to close.  We are sorry for the inconvenience.").  To verify that it wasn't a memory issue,  I even tried "rs <- dbSendQuery(con, "select 'a', 'b'")".  This statement causes the error as well.
> 
> I tried upgrading the packages, and upgrading R from 2.5.0 to 2.8.1.  However, I still get the same errors.  Has anyone run into this problem before?  Any suggestions on how to solve it?


Hi Aaron,

Be sure to read the details of the RMySQL web page:

http://biostat.mc.vanderbilt.edu/RMySQL

You need to make sure and match the version of your MySQL client library 
(not the running MySQL server) with the RMySQL binary that you choose 
from the web page above.

Best,

Jeff
-- 
http://biostat.mc.vanderbilt.edu/JeffreyHorner




More information about the R-help mailing list