[daip] Strange problem with IMAGR...

[Nissim Kanekar]

Hi Eric,

Yup, I noticed the information on the AIPS page, and re-installed 31DEC17 (and 
also replaced 31DEC15 with 31DEC17 on another machine) and the problems went 
away, and all's well now. The message file issue is more puzzling; I don't seem 
to have a 0-byte MS* file in any of the data areas, as far as I can see.

Thanks!
Nissim

On Mon, 27 Nov 2017, Eric Greisen wrote:

> On 11/23/2017 09:25 AM, Nissim Kanekar wrote:
>> 
>> Hi Eric,
>> 
>> I've come across a strange problem with IMAGR and was wondering whether you 
>> might have seen this before and might know a solution. It's giving me the 
>> following error:
>> 
>> ZMSFI2: REQUEST FOR BYTES 1 THRU = 0
>> ZMSFI2: BEYOND EOF = 0
>>   ZMSGDK: OPER = READ LUN = 12 NREC =        1
>>   ZMSGER: SYSTEM I/O ERROR CODE =         -1
>> MSGWRT ERROR      3 AT OPEN
>> IMAGR1: ZFI2: DELAY 1
>> ZMSFI2: REQUEST FOR BYTES 1 THRU = 0
>> ZMSFI2: BEYOND EOF = 0
>>   ZMSGDK: OPER = READ LUN = 12 NREC =        1
>>   ZMSGER: SYSTEM I/O ERROR CODE =         -1
>> MSGWRT ERROR      3 AT OPEN
>> IMAGR1: ZABORS: signal 11 received
>> ZMSFI2: REQUEST FOR BYTES 1 THRU = 0
>> ZMSFI2: BEYOND EOF = 0
>>   ZMSGDK: OPER = READ LUN = 12 NREC =        1
>>   ZMSGER: SYSTEM I/O ERROR CODE =         -1
>> MSGWRT ERROR      3 AT OPEN
>> IMAGR1: ABORT!
>> AIPS 1: TASK QUIT WITHOUT RESUMING ME
>> AIPS 1: Begin check for any 'standard' scratch files
>> AIPS 1: Scratch files -- destroyed:    2  still active:    0
>> AIPS 1: Resumes
>> 
>> 
>> Also, the message server shows NaN's as the beam minimum and maximum:
>> 
>> QUIXOT> IMAGR1: Field    1 Beam min =    NaN      Jy, max =    NaN      Jy
>> QUIXOT> IMAGR1: Fit Gaussian FWHM =   10.265 x    5.346      arcsec, PA=  
>> -61.5
>> 
>> Do you have any idea what might be going on?
>> 
>> Thanks!
>> Nissim
>> 
>
> You seem to have 2 problems - the first is that your message file is 0 bytes. 
> Go to your first data area and delete the MS* file with 0 bytes (it has your 
> user number in extended hex in the name).
>
> The second seems to be a problem that is creeping around the world. Install 
> or run a MNJ on 31DEC17 to get a whole new set of load modules compiled with 
> gfortran - the Intel compiler has something not compatible with the latest 
> operating systems although we have not been able to determine what that might 
> be.
>
> ERic Greisen
>
>