ARSLOAD failing to ingest AFP from USS(UNIX System Services ISPF Shell) in zOS

[azkhan]

Hi All,

I am facing a issue where I have the AFP documents place in the USS share path on zOS environment.
I am using OS390 indexer to perform the indexing, passing the details via PARM parameters in my job.
I am using BPXBATCH to execute ARSLOAD-
//ARSLOAD  EXEC PGM=BPXBATCH,REGION=0M                               
//STDPARM  DD DSN=xxx.xxx.xxxx,DISP=SHR    -->PARM input dataset                     
//SYSPRINT DD SYSOUT=*                                               
//SYSABEND DD SYSOUT=*                                               
//STDERR   DD PATH='/tmp/art.stderr',PATHMODE=SIRWXU,               
//             PATHOPTS=(OWRONLY,OCREAT,OAPPEND),PATHDISP=(KEEP,KEEP)
//STDOUT   DD PATH='/tmp/art.stdout',PATHMODE=SIRWXU,               
//            PATHOPTS=(OWRONLY,OCREAT,OAPPEND),PATHDISP=(KEEP,KEEP)
//STDENV   DD *
                                                     
PARM PARAMTERS:
SH /usr/lpp/ars/V9R5M0/bin/arsload   
-h                                   
ARCHIVE                             
-f                                   
-u       
xxxx8 
-p       
xxxx
-g       
LXXXXXXX
-a       
LYYYYYYY
-v                                   
/arstemp/xxx/xx/test.afp

Error:
ARS1134E Invalid structured field header in the file /u/s3074bs/l390.17105953.27
ARS5476E The OS/390 indexer failed with the specified return code. (Arsl390 return code is 6).
ARS4311E Loading failed       
ARS4318E Processing failed for file >/arstemp/xxx/xx/test.afp     

But when I am copy this same AFP into dataset and execute ARSLOAD stand alone job it is successfully completed and documents are loaded successfully.
ARSLOAD PARM paramters:
//PARM DD
-h       
ARCHIVE
-f       
-v       
-u       
xxxx8 
-p       
xxxx
-g       
LXXXXXXX
-a       
LYYYYYYY
-s       
INPUT   

has anyone come across such scenario, is it not possible to load data using OS390 indexer from USS path directly?

[Ed_Arnold]

Some notes:

The safest way (and possibly the only way) to move AFP from an HFS to a regular MVS dataset is a two step process.

1.  OCOPY BIN to a temp dataset

2.  AFRREBLK

Ref:  https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.1.0/com.ibm.zos.v2r1.apsu000/reblk.htm


On V10.1 the OS/390 Indexer functionality was expanded to allow an HFS as an input dataset.  This is a result of the OS/390 Indexer now running on AIX as well as z/OS.

For example:

//   EXEC PGM=ARSLOAD,PARM='-h [other parms] /tmp/whatever'

or in a shell:

arsload -h [other parms] /tmp/whatever

Note:  no -s INPUT

The FILEFORMAT (for linedata documents) is used to read the data and is also how the data is output.

Ed