Issue with Cluster on R4.03 MacOS Big Sur about icamp1 HOT 6 CLOSED

Sorry to hear that. I searched around. This seems a difficult problem without a final solution yet.
rstudio/rstudio#6692

I do not have Mac, thus, cannot test. If you find any solution, I am willing to revise iCAMP.
If you are using Rstudio, did you try the original R?
If R 4.x does not work anyway, you may have to install an old version of R (>=3.5) to run iCAMP, until R or the 'parallel' package developers find the solution.
If your dataset is not very large, you may use our galaxy pipeline http://ieg3.rccc.ou.edu:8080/. please email me ([email protected]) to set up your account.

from icamp1.

I believe I have found a temporary fix:
cl <- parallel::makeCluster(10, setup_timeout = 0.5)

Then specify nworkers= cl and it seemed to work for me. For others having this issue, try some of the other solutions on the above link if this doesn't work.

Thanks!

from icamp1.

Great. I will try to add this to the functions.

from icamp1.

The following code has been added to iCAMP (version >= 1.4.3, can be found in /RPckage/AllVersion).

paste0("RC",taxo.metric," i")
Example output files

c1<-try(parallel::makeCluster(nworker,type="PSOCK"))
if(class(c1)[1]=='try-error'){c1 <- parallel::makeCluster(nworker, setup_timeout = 0.5)}
if(class(c1)[1]=='try-error'){c1 <- parallel::makeCluster(nworker, setup_strategy = "sequential")}

from icamp1.

@bholt20 please let me know if the new version (1.4.3) can solve your problem. I will close this issue if no more question : )

from icamp1.

Yes, this fixed it. Thanks!

from icamp1.