Unfortunately, I can't tell you how to solve this problem. However, I can confirm that this crash happens also on my machines (R 2.0.1 on Debian Sarge, AMD Athlon, 750 MHz, 256 MB RAM and R 2.0.0 on SuSE 9.0, Intel P4, 2.6 GHz, 512 MB RAM): R> aaa(1000) alloc of 80000 bytes failed alloc of 40000 bytes failed Speicherzugriffsfehler
We had a model, which consists of 1000+ linear programs with each about 550 rows and 1100 columns (http://www.unipr.it/arpa/dipseq/EAAE/PR/Parallel/ 42b4_henningsen.pdf). On MS-Windows 2000/XP this model always crashed after some hundreds linear programs, but using Linux on the same machines it _never_ crashed. I really can't see the reason why our much larger programs worked, but these small programs let R crash. I hope that this information may help to solve this problem. Arne On Wednesday 30 March 2005 12:16, TEMPL Matthias wrote: > Hello! > > I have a curious problem, which I cannot solve. > With my code I solve thousands of small linear programs with the package > lpSolve automatically. But R crashes sometimes (~always, but always on > different linear programs) in a strange way. For illustration, I tried to > prepare a simple example, which shows the nature of the problem. The > function aaa (see below) declares some constants (only in this special > example) and in the end it solves the linear program. > > aaa(1) > [,1] [,2] > [1,] 0 0 > [2,] 0 0 > [3,] 0 0 > [4,] 0 0 > [5,] 52 0 > [6,] 2 2 > [7,] 0 0 > [8,] 2 0 > [9,] 0 0 > [10,] 0 0 > [11,] 0 0 > [12,] 0 0 > [13,] 54 0 > > Works fine. > Now I make the *same* calculation, say 1000 times: > aaa(1000) > R (I have tried it with R2.0.1, 2.0.0, 1.9.1, 2.1.0dev) crashes completly - > without warning and error message under Windows XP, Intel Pentium 3 with > 256 MB RAM Under Linux SuSe 8.2 R (2.0.1) it crashes again, but in this > case I get the following message: Calloc of 40004 bytes failed on line 114 > of file lpkit.c > ... > Calloc of 80008 bytes failed on line 113 of file lpkit.c > Error: cannot allocate vector of size 3 Kb > > Now I´m completly lost. Solving the linear program one time makes no > problem. Solving it twice in the same way makes no problem either. Running > the same calculation, say 1000 times, causes a crash. Why should there be a > problem with memory? > > For any hint, I would be really happy. > Thank you, > Matthias > > > ### ---------- function aaa ------------ > > aaa <- function(amount=1){ > f.obj <- rep(0,33) > > w <- c(3,4,5,6,11,13,17,22,25,26,27,28,33) > > m <- > matrix(c(0,0,1,1,1,1,0,0,0,0,-1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 >, > > > 0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,-1,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,-1,-1,-1,0,0,0,0,1, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0,0, > > > 0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0, > > > 0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0, > > > 0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, > > > 0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0,0,0,0,-1,0,0,0,0,0,0,0,0,0,0,1 > > ),ncol=33,byrow=TRUE) > > f.dir <- c("=","=","=","=","=","=","=","=","=","=","=", > "=","=","=","=","=","=","=","=","=","=","=","=", > "=","=","=","=","=","=","=","=","=","=") > > f.rhs <- c(-52,0,54,0,2,0,0,0,0,0,0,0,0,0) > > lp.out <- matrix( ncol = 2, nrow = 13 ) > > for(ii in 1:amount){ # - simple iterate the same > > for( i in 1:13 ){ > > f.obj[ w[i] ] <- 1 > > lp.out[ i, 1 ] <- lp("min", f.obj, m, f.dir, f.rhs)$objval > > lp.out[ i, 2 ] <- lp("max", f.obj, m, f.dir, f.rhs)$objval > > f.obj <- rep( 0, 33 ) > > } > > } > lp.out > > } > > ______________________________________________ > R-help@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html -- Arne Henningsen Department of Agricultural Economics University of Kiel Olshausenstr. 40 D-24098 Kiel (Germany) Tel: +49-431-880 4445 Fax: +49-431-880 1397 [EMAIL PROTECTED] http://www.uni-kiel.de/agrarpol/ahenningsen/ ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html