Conference iosg::all-in-1_v30

	Have you checked for a bad file?  (E.g., a non-template file -- such as
a journal file left over from the editor -- in a BLP directory)

	F

    Yes, I have checked that.
    
   $ dir oa$blp:/exclu=*.blp
     %DIRECT-W-NOFILES, no files found
    
    Thanks,
    
    Lourdes

Check the MRE$BUILD_NOGRS_FILE.BLP is it actually ASCII? Or check the file
after this one.	

	Ben

    Ben,
    
    I have checked these files. They are ASCII. they only have one line
    each one, they are:
    
    MRE$BUILD_NOGRS_FILE.BLP:
    
    <mre$e_message.message["%MRE-E-NGR"]>
    
    MRE$BUILD_NOPRS_FILE.BLP:
    
    <mre$e_message.message["%MRE-E-NPR"]>
    
    They mustn't be the problem, because they existed before and never they
    had had problems.
    
    Any idea more?
    
    Thaks and regards,
    
    Lourdes
    
                                                              

    I have some further information on this call.  Anybody have some ideas
    as to what Quotas might be low given the following; 

Checking everything in OA$SITE_BLP_xxx is a BLP file       
there's no corruption in the file where the compilation      
some other test shows the following results:              
    
A)	If entering ALL-IN-1 using ALLIN1/NOCUSTOM, option CCM        
	the same behaviour. This implies the original script         
	is being used.                                 
    
B)	Entering ALL-IN-1 using ALLIN1/NOINIT and doing DO CM_CDS     
	the same result.                                             
    
C)	When entering ALL-IN-1 using ALLIN1/NOINIT and doing          
	CMTXL, no error is generated.                     
    
D)	Same as c) doing <DO OA$TXL_COMPILE CMTXL from any            
	screen.                                                 
    
E)	If you invoke the following script instead of the             
	CM_CDS.SCP, the compilation goes OK (as you can         
	this script contains the same instructions as the           
	CM_CDS.SCP, but the .IF, GOTO and .LABEL ones:          

! OUR_CM_CDS.SCP                                                 
!+                                                               
GET #CM_TXL = "CDS"                                              
OA$FAO OA$_SCM_TXL_COM,"#CM_TEMP",#CM_TXL                        
YESNO_PROMPT #CM_TEMP\OA$FLD_STAY                                
.IF OA$PROMPT_DISPOSE NE 2 THEN .EXIT                            
CLEAR\OA$TXL_COMPILE CMTXL\OA$TXL_REMOVE CMTXL\OA$MSG_PURGE      
GET OA$DCL = "SET PROTECTION = W:RE OA$LIB:CMTXL.TXL"            
OA$TXL_INSTALL CMTXL                                             
PURGE_FILE OA$LIB:CMTXL.TXL,2                                    
OA$FAO OA$_SCM_TXL_COP,"OA$DISPLAY",#CM_TXL                      
.EXIT                                                            
    
F)	If you move the BLP's from OA$SITE_BLP_SHARE to               
	OA$SITE_BLP_LLV and the ones from _LLV to _SHARE, no error       
	is obtained.                                                     
    
G)	If you move every BLP to OA$SITE_SHARE or to                  
	OA$SITE_ENGLISH, no error.                                       
    
H)	If you generate in OA$SITE_BLP_ENGLISH a BLP similar in       
	name to those in OA$SITE_BLP_SHARE (all of them in the form      
	MRE$*.BLP), an access violation occurs, but in a different       
	file than previous tests.                                        
    
As we are thinking in something similar to insufficient          
memory to do the compilation, we've tested increasing the        
process quotas of the ALLIN1 account (WSDEFAULT, WSQUOTA,        
BYTLM, ENQLM, PGFQUOTA, DIOLM, BIOLM) but no result.             
    
    Cheers,
    Andy

	Bit too late for you Andy, but just in case anyone else comes
	across this problem (like I just did!)....

	You will need to execute the OA$TXL_CLOSE function before the
	OA$TXL_REMOVE and OA$TXL_COMPILE functions.

		p1