Personally, building Amber on ubuntu has regularly produced the least amount of build problems and fewest problems with tests compared to other OSs.
Just based on the error reported, this is an indication that there might be a problem with the "read-frame”, though what could have caused it is not as easy guess.
Maybe a simple place to start is to check the file integrity? Have a look at the files in question:
/home/Kai/software/amber20/AmberTools/src/cpptraj/test/Test_Esander_Short/*.save
Maybe do a quick compare:
md5sum /home/Kai/software/amber20/AmberTools/src/cpptraj/test/Test_Esander_Short/*.save
MD5 (gb.dat.save) = a4436aed03bca71e9bc88e89b011f734
MD5 (pme.dat.save) = 5660d9f15eecef44fd0d9baefd163a14
If this matches up, maybe there is a problem with these files? I’m not savvy enough in fortran code to judge if it is correct:
READ(AA(IPT3+1:IPT4-1),'(F8.3)') RPVER
Where I’m not sure what F8.3 specifies, floating point with 8 characters and something (.3). Though if there was a problem I would guess that more people might have reported it?
Best regards
// Gustaf
> On 20 Aug 2020, at 04:00, David A Case <david.case.rutgers.edu> wrote:
>
> On Thu, Aug 20, 2020, 东庚 wrote:
>
>> I am Geng Dong from Shantou University Medical College.
>> Thank you for you reply in amber.org mail list about my problem. Since I
>> can not attach the test.log file to the mail of amber, I send it to you
>> directly.
>
> I am at a loss here. You have many, many errors -- lots are all in the same
> piece of code, but many others are not.
>>
>> We installed amber on Ubuntu 18 LTS with gnu complier.
>
> I personally, and many others, have used Ubuntu18. Can you report the
> result of typing "which gfortran" and "gfortran --version"? Can you try an
> installation on some other machine?
>
> Is there any chance that the encoding of files is odd? Might there have been
> some conversion related to Chinese characters? Or perhaps an encrypted file
> system? I'm fishing for any information you might have about why your
> system might differ from what almost everyone else has.
>
> I'm cc-ing this to the amber list with a truncated version of the log file.
>
> The failure to parse the VERSION string in prmtop files happens several
> hundred times; but other errors pop up as well.
>
> ...regards...dac
>
> <test.log>_______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Aug 20 2020 - 01:00:02 PDT
