R devel - Mar 2008 - crazy behavior after Error (PR#10956)

Full_Name: Jan Somorcik
Version: 2.6.0
OS: Windows XP HOME
Submission from: (NULL) (158.195.31.33)


I used to run a simulation with 5000 iterations. For a long time everything was
OK, but since yesterday R stops after 500-1500 iterations [never at the same
position] with an [random?] Error message [always something else, e.g.:

Error in ans[[i]] <- tmp : incorrect number of arguments to "<-"

And after that R gets crazy:
Most of the commands result in an Error message, just an example:
> d
[1] 3
> d<-7
Error: could not find function "<-"
> objects()
Error in objects() : 
  Value of SET_STRING_ELT() must be a 'CHARSXP' not a
'double'
> sum(c(1,2,3))
Error: could not find function "sum"


It's always different. The most crazy situation was:
the result of an ARBITRARY command was a STRING containing the call of the
previuos command :-)))

What is the problem?
Is it an issue of Windows XP? It's beeing updated regularly and the on-line
scanners [Panda, Symantec, Kaspersky... did not report infection]

Other applications running:
-Total Commander
-notepad
-AVG Free antivirus

The behaviour is not reproducible, it is always different.

somorcik at fmph.uniba.sk wrote:
> Full_Name: Jan Somorcik
> Version: 2.6.0
> OS: Windows XP HOME
> Submission from: (NULL) (158.195.31.33)
>
>
> I used to run a simulation with 5000 iterations. For a long time everything
was
> OK, but since yesterday R stops after 500-1500 iterations [never at the
same
> position] with an [random?] Error message [always something else, e.g.:
>
> Error in ans[[i]] <- tmp : incorrect number of arguments to
"<-"
>
> And after that R gets crazy:
> Most of the commands result in an Error message, just an example:
>
>   
>> d
>>     
> [1] 3
>   
>> d<-7
>>     
> Error: could not find function "<-"
>   
>> objects()
>>     
> Error in objects() : 
>   Value of SET_STRING_ELT() must be a 'CHARSXP' not a
'double'
>   
>> sum(c(1,2,3))
>>     
> Error: could not find function "sum"
>
>
> It's always different. The most crazy situation was:
> the result of an ARBITRARY command was a STRING containing the call of the
> previuos command :-)))
>
> What is the problem?
> Is it an issue of Windows XP? It's beeing updated regularly and the
on-line
> scanners [Panda, Symantec, Kaspersky... did not report infection]
>   
It sure looks as though whatever you are doing in your simulations is buggy.

Duncan Murdoch
> Other applications running:
> -Total Commander
> -notepad
> -AVG Free antivirus
>
> The behaviour is not reproducible, it is always different.
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>