I am submitting two jobs as below on Linux (R 3.6.0). See below sample code.
library("future")
plan(multicore)
job1 <- future({
print("job1")
})
job2 <- future({
print("job2")
})
while (!resolved(job1)) {
Sys.sleep(0.2)
}
while (!resolved(job2)) {
Sys.sleep(0.2)
}
If a job takes more than 3 min due to the size of data then I am getting below error. If data is small then I am not getting error. Am I missing anything?
Error in sendMaster(try(eval(expr, env), silent = TRUE)) : ignoring SIGPIPE signal Calls: source ... run.MulticoreFuture -> do.call -> -> sendMaster
Additional References:
I found that similar issues have been discussed on different blogs. It sounds like bug was fixed in R 3.6.0 but I am still getting it even using R 3.6.0
https://r.789695.n4.nabble.com/error-in-parallel-sendMaster-td4760382.html
https://github.com/wch/r-source/commit/e08cffac1c5b9015a1625938d568b648eb1d8aee
My job1 and job2 are independent and writes output to file so I don't need to send anything back to main job. Is there any way I can define in job to say that there is no need to send anything back to main job?
Update
I don't get an error if I use plan(multisession)
Session and Core Info
sessionInfo()
R version 3.6.0 (2019-04-26)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Amazon Linux AMI 2018.03
Matrix products: default
BLAS: /usr/local/R/R-3.6.0/lib/libRblas.so
LAPACK: /usr/local/R/R-3.6.0/lib/libRlapack.so
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] future_1.17.0
loaded via a namespace (and not attached):
[1] compiler_3.6.0 tools_3.6.0 parallel_3.6.0 listenv_0.8.0
[5] codetools_0.2-16 digest_0.6.24 globals_0.12.5
future::availableCores(constraint = 'multicore')
system
2
