png() segfaults inside a parallel::mclapply loop?

Hi,

I can't figure this out -- I have to (slightly different) OSX boxes
running the same version of R that I grabbed from r.research.att.com a
little while ago:

R version 2.15.0 Patched (2012-04-18 r59079)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)

Simply calling png() w/in a parallel loop always crashed the
subprocesses on one machine. This code, for example:

library(parallel)
mclapply(1:10, function(i) {
 png(paste("~/tmp/p", i, "png", sep="."))
 plot(1:10)
 dev.off()
}, mc.cores=8)

The error (from each subprocess) is pasted below, however swapping out
the png() call to pdf() works on both.

The machine that png() fails on is a ~ 2008 2x4-core mac pro. The one
that doesn't have a problem is a late 2010 core2duo macbook pro 13"

Both are running OSX.6.8

Anybody else have this problem?

-steve

Error from one process:

*** caught segfault ***
address 0x108, cause 'memory not mapped'

Traceback:
1: png(paste("~/tmp/p", i, "png", sep = "."))
2: FUN(7L[[1L]], ...)
3: lapply(X = S, FUN = FUN, ...)
4: doTryCatch(return(expr), name, parentenv, handler)
5: tryCatchOne(expr, names, parentenv, handlers[[1L]])
6: tryCatchList(expr, classes, parentenv, handlers)
7: tryCatch(expr, error = function(e) {    call <- conditionCall(e)
if (!is.null(call)) {        if (identical(call[[1L]],
quote(doTryCatch)))             call <- sys.call(-4L)        dcall <-
deparse(call)[1L]        prefix <- paste("Error in", dcall, ": ")
 LO
NG <- 75L        msg <- conditionMessage(e)        sm <- strsplit(msg,
"\n")[[1L]]        w <- 14L + nchar(dcall, type = "w") + nchar(sm[1L],
type = "w")        if (is.na(w))             w <- 14L + nchar(dcall,
type = "b") + nchar(sm[1L],                 type = "b")
 if (w > LONG)             prefix <- paste0(prefix, "\n  ")    }
else prefix <- "Error : "    msg <- paste0(prefix,
conditionMessage(e), "\n")    .Internal(seterrmessage(msg[1L]))    if
(!silent && identical(getOption("show.error.messages"),         TRUE))
{        ca
t(msg, file = stderr())        .Internal(printDeferredWarnings())    }
  invisible(structure(msg, class = "try-error", condition = e))})
8: try(lapply(X = S, FUN = FUN, ...), silent = TRUE)
9: sendMaster(try(lapply(X = S, FUN = FUN, ...), silent = TRUE))
10: FUN(1:8[[7L]], ...)
11: lapply(seq_len(cores), inner.do)
12: mclapply(1:10, function(i) {    png(paste("~/tmp/p", i, "png", sep
= "."))    plot(1:10)    dev.off()}, mc.cores = 8)

-- 
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact

Steve,

you can't use certain graphics devices/backends in forked processes (see multicore docs). ?What is known to work is the Cairo package.

Cheers,
Simon



On May 1, 2012, at 12:30 PM, Steve Lianoglou <mailinglist.honeypot at gmail.com> wrote:

Hi,

I can't figure this out -- I have to (slightly different) OSX boxes
running the same version of R that I grabbed from r.research.att.com a
little while ago:

R version 2.15.0 Patched (2012-04-18 r59079)
Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)

Simply calling png() w/in a parallel loop always crashed the
subprocesses on one machine. This code, for example:

library(parallel)
mclapply(1:10, function(i) {
?png(paste("~/tmp/p", i, "png", sep="."))
?plot(1:10)
?dev.off()
}, mc.cores=8)

The error (from each subprocess) is pasted below, however swapping out
the png() call to pdf() works on both.

The machine that png() fails on is a ~ 2008 2x4-core mac pro. The one
that doesn't have a problem is a late 2010 core2duo macbook pro 13"

Both are running OSX.6.8

Anybody else have this problem?

-steve

Error from one process:

*** caught segfault ***
address 0x108, cause 'memory not mapped'

Traceback:
1: png(paste("~/tmp/p", i, "png", sep = "."))
2: FUN(7L[[1L]], ...)
3: lapply(X = S, FUN = FUN, ...)
4: doTryCatch(return(expr), name, parentenv, handler)
5: tryCatchOne(expr, names, parentenv, handlers[[1L]])
6: tryCatchList(expr, classes, parentenv, handlers)
7: tryCatch(expr, error = function(e) { ? ?call <- conditionCall(e)
if (!is.null(call)) { ? ? ? ?if (identical(call[[1L]],
quote(doTryCatch))) ? ? ? ? ? ? call <- sys.call(-4L) ? ? ? ?dcall <-
deparse(call)[1L] ? ? ? ?prefix <- paste("Error in", dcall, ": ")
?LO
NG <- 75L ? ? ? ?msg <- conditionMessage(e) ? ? ? ?sm <- strsplit(msg,
"\n")[[1L]] ? ? ? ?w <- 14L + nchar(dcall, type = "w") + nchar(sm[1L],
type = "w") ? ? ? ?if (is.na(w)) ? ? ? ? ? ? w <- 14L + nchar(dcall,
type = "b") + nchar(sm[1L], ? ? ? ? ? ? ? ? type = "b")
?if (w > LONG) ? ? ? ? ? ? prefix <- paste0(prefix, "\n ?") ? ?}
else prefix <- "Error : " ? ?msg <- paste0(prefix,
conditionMessage(e), "\n") ? ?.Internal(seterrmessage(msg[1L])) ? ?if
(!silent && identical(getOption("show.error.messages"), ? ? ? ? TRUE))
{ ? ? ? ?ca
t(msg, file = stderr()) ? ? ? ?.Internal(printDeferredWarnings()) ? ?}
? invisible(structure(msg, class = "try-error", condition = e))})
8: try(lapply(X = S, FUN = FUN, ...), silent = TRUE)
9: sendMaster(try(lapply(X = S, FUN = FUN, ...), silent = TRUE))
10: FUN(1:8[[7L]], ...)
11: lapply(seq_len(cores), inner.do)
12: mclapply(1:10, function(i) { ? ?png(paste("~/tmp/p", i, "png", sep
= ".")) ? ?plot(1:10) ? ?dev.off()}, mc.cores = 8)

--
Steve Lianoglou
Graduate Student: Computational Systems Biology
?| Memorial Sloan-Kettering Cancer Center
?| Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact