Closed MichaelChirico closed 4 years ago
NOTES: Failing to restore an initial .RData is a fatal error from 0.65.
Audit (from Jitterbug): Mon Aug 24 16:47:57 1998 pd moved from incoming to startup Fri May 07 06:01:48 1999 ripley changed notes Sun Jul 18 21:03:27 1999 ripley changed notes Sun Jul 18 21:03:27 1999 ripley moved from Startup to Startup-fixed
From: "Michelle McPherson" <michelle.mcpherson@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>> PARTS: 2 I am having the same problem with start-up as Martyn Plummer was (ID#21).
I have downloaded rw1070 twice from the University of Wisconsin, and each time, after using the program once, I was unable to restart the program. The message I am getting reads "Fatal error: unable to restore saved data in .RData".
My OS is Windows 2000. I am trying to save small data files and a few glmmPQL models.
If you can tell me how to avoid this problem, I would appreciate it.
thanks very much, Michelle
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
**END
Bug 16399 has been marked as a duplicate of this bug.
From: Martyn Plummer <plummer@<::CENSORED -- SEE ORIGINAL ON BUGZILLA::>> [This is not so much a bug report as a request for idiot-proofing. However, you did ask us to pick nits...]
If a memory image is too large to be restored, because the vector heap or cons heap is too small, it would be better if R exited immediately (i.e. call "R_Suicide" instead of "error"). This would force the user to specify either the "--no-restore" option or the appropriate "-n N -v N" options, depending whether they want the saved image or not.
At the moment, when I get this message, I exit manually. But I always get the "Save workspace image?" prompt. This is a bit pointless as there is nothing to save and there is a risk that I might accidently press the "y" key, overwriting my .RData file. I haven't done this yet, but with my record it's only a matter of time.
Martyn
METADATA