I am using the jperraud branch and just tried to load R on CentOS using the new initialization that passes in strings ( this was to test/avoid the whole --no-save/-vanilla stuff). When I tried this I received the following error:
*** glibc detected *** mono-sgen: double free or corruption (fasttop):
Followed by a very long debug list and signs that this occurred in the constructor/initialization call. Examining the string as argumetns being passed, I found this was being passed:
rdotnet_app --slave --interactive --no-save --no-restore-data --max-ppsize=50000
Changing this to remove the interactive bit as:
rdotnet_app --quiet --slave --no-save --no-restore-data --max-ppsize=50000
Seemed to avoid the bad memory pointer free (whch I gather is what the glibc detected thing was). I have no idea what caused this, but it seems here might be a problem with conflicting arguments or god knows what.
*** glibc detected *** mono-sgen: double free or corruption (fasttop):
Followed by a very long debug list and signs that this occurred in the constructor/initialization call. Examining the string as argumetns being passed, I found this was being passed:
rdotnet_app --slave --interactive --no-save --no-restore-data --max-ppsize=50000
Changing this to remove the interactive bit as:
rdotnet_app --quiet --slave --no-save --no-restore-data --max-ppsize=50000
Seemed to avoid the bad memory pointer free (whch I gather is what the glibc detected thing was). I have no idea what caused this, but it seems here might be a problem with conflicting arguments or god knows what.