Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Hello Rob!

Just wanted to summarize again (for my own clarity, sanity) using the 2-pass code you helped me with Rob works.  The generated code properly generates all the 'exclude' statements where we found the RACF ID 'UserID' matches the '/u/home/userid' and is not case-sensitive. For example:

 User         Name                                              Uid  HomeDir                  SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235  /u/home/#cia0a             

RACF ID User "#CIA0A" is correct in our scheme for the proper name for the HomeDir "/u/home/#cia0a" so the code generates an 'exclude' statement as:

exclude key=#CIA0A and subdir=#CIA0A

However, once the 2-Pass code generated is submitted and run it does NOT 'exclude' those in the report.   I think that's what I'm aiming for with this report, trying to find all 'HomeDir' values that don't represent the RACF ID, and also to find any 'uppercase' values in the 'HomeDir'.   

Many thanks Rob for reading thru my ramblings.  

Lynn

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Rob,

Last minute update...  I removed the parameter 'and subdir=#CIA0A'  and the generated 'exclude' statements were in fact excluded.  I guess either the 'subdir=' value is not right or does not contain data, or the 'and' statement does not work on the 'exclude' statement?  I simply removed the 'and subdir=#CIA0A' from a group of statements and they were successfully excluded from the report.  You probably already knew that (asking for 'SubDir' to print out (see code and output I pasted herein). 

Having the 'and subdir=' in the exclude statement gives me more clarity though that these records are the ones we want to keep, only trying to find the odd ones out. 

select c=user s=omvs                                             
  exclude key=#CIA0A 

Hello Rob!

Just wanted to summarize again (for my own clarity, sanity) using the 2-pass code you helped me with Rob works.  The generated code properly generates all the 'exclude' statements where we found the RACF ID 'UserID' matches the '/u/home/userid' and is not case-sensitive. For example:

 User         Name                                              Uid  HomeDir                  SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235  /u/home/#cia0a             

RACF ID User "#CIA0A" is correct in our scheme for the proper name for the HomeDir "/u/home/#cia0a" so the code generates an 'exclude' statement as:

exclude key=#CIA0A and subdir=#CIA0A

However, once the 2-Pass code generated is submitted and run it does NOT 'exclude' those in the report.   I think that's what I'm aiming for with this report, trying to find all 'HomeDir' values that don't represent the RACF ID, and also to find any 'uppercase' values in the 'HomeDir'.   

Many thanks Rob for reading thru my ramblings.  

Lynn

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Hi Lynn

I'm entering my reply a second time, the first reply was held up because I wrote a syntax summary word(home,zzz,'/') except I used x instead of z.  Guess what, 3 time x is a prohibited word in the forum.

Anyway, when I used word(home,3,'/') in the code, I had overlooked the way that word deals with the first / in the directory.  I should have used word(home,4,'/') to catch the 3rd node of the directory!  The code to generate your report should look like:

n type=racf nopage dd=ckr2pass
 define once(nd) boolean where key=x /* any field test will do */
 define each(nd) count
 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,4,"/") where home=:"/u/home/" and 
           missing(word(home,5,"/"))',
/ ' select c=user s=omvs'

 summary each,
'  exclude key=' | key(0) 'subdir=' | key(0)

 summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

------------------------------
Rob van Hoboken
------------------------------

Hi Lynn

I'm entering my reply a second time, the first reply was held up because I wrote a syntax summary word(home,zzz,'/') except I used x instead of z.  Guess what, 3 time x is a prohibited word in the forum.

Anyway, when I used word(home,3,'/') in the code, I had overlooked the way that word deals with the first / in the directory.  I should have used word(home,4,'/') to catch the 3rd node of the directory!  The code to generate your report should look like:

n type=racf nopage dd=ckr2pass
 define once(nd) boolean where key=x /* any field test will do */
 define each(nd) count
 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,4,"/") where home=:"/u/home/" and 
           missing(word(home,5,"/"))',
/ ' select c=user s=omvs'

 summary each,
'  exclude key=' | key(0) 'subdir=' | key(0)

 summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

Hello Rob,

I think the problem was in the 'exclude'.  I altered it to this syntax:

summary each,                                                       

  '  exclude ( key=' | key(0) ' and home=' | home(0) ')'             

and with this formatted exclude statement generated by the 2-pass code:

select c=user s=omvs

 exclude ( key=#CIA0A    and home=/u/home/#cia0a ) 

It worked.  the 'and' needed to be in parentheses.  

I'll continue to work this to see if it's what these users want.  

Many thanks for your 'bounce back' on my ramblings. 

Lynn

Rob,

Last minute update...  I removed the parameter 'and subdir=#CIA0A'  and the generated 'exclude' statements were in fact excluded.  I guess either the 'subdir=' value is not right or does not contain data, or the 'and' statement does not work on the 'exclude' statement?  I simply removed the 'and subdir=#CIA0A' from a group of statements and they were successfully excluded from the report.  You probably already knew that (asking for 'SubDir' to print out (see code and output I pasted herein). 

Having the 'and subdir=' in the exclude statement gives me more clarity though that these records are the ones we want to keep, only trying to find the odd ones out. 

select c=user s=omvs                                             
  exclude key=#CIA0A 

Hello Rob!

Just wanted to summarize again (for my own clarity, sanity) using the 2-pass code you helped me with Rob works.  The generated code properly generates all the 'exclude' statements where we found the RACF ID 'UserID' matches the '/u/home/userid' and is not case-sensitive. For example:

 User         Name                                              Uid  HomeDir                  SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235  /u/home/#cia0a             

RACF ID User "#CIA0A" is correct in our scheme for the proper name for the HomeDir "/u/home/#cia0a" so the code generates an 'exclude' statement as:

exclude key=#CIA0A and subdir=#CIA0A

However, once the 2-Pass code generated is submitted and run it does NOT 'exclude' those in the report.   I think that's what I'm aiming for with this report, trying to find all 'HomeDir' values that don't represent the RACF ID, and also to find any 'uppercase' values in the 'HomeDir'.   

Many thanks Rob for reading thru my ramblings.  

Lynn

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Hi Lynn

I'm entering my reply a second time, the first reply was held up because I wrote a syntax summary word(home,zzz,'/') except I used x instead of z.  Guess what, 3 time x is a prohibited word in the forum.

Anyway, when I used word(home,3,'/') in the code, I had overlooked the way that word deals with the first / in the directory.  I should have used word(home,4,'/') to catch the 3rd node of the directory!  The code to generate your report should look like:

n type=racf nopage dd=ckr2pass
 define once(nd) boolean where key=x /* any field test will do */
 define each(nd) count
 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,4,"/") where home=:"/u/home/" and 
           missing(word(home,5,"/"))',
/ ' select c=user s=omvs'

 summary each,
'  exclude key=' | key(0) 'subdir=' | key(0)

 summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

Hello Rob,

I think the problem was in the 'exclude'.  I altered it to this syntax:

summary each,                                                       

  '  exclude ( key=' | key(0) ' and home=' | home(0) ')'             

and with this formatted exclude statement generated by the 2-pass code:

select c=user s=omvs

 exclude ( key=#CIA0A    and home=/u/home/#cia0a ) 

It worked.  the 'and' needed to be in parentheses.  

I'll continue to work this to see if it's what these users want.  

Many thanks for your 'bounce back' on my ramblings. 

Lynn

Rob,

Last minute update...  I removed the parameter 'and subdir=#CIA0A'  and the generated 'exclude' statements were in fact excluded.  I guess either the 'subdir=' value is not right or does not contain data, or the 'and' statement does not work on the 'exclude' statement?  I simply removed the 'and subdir=#CIA0A' from a group of statements and they were successfully excluded from the report.  You probably already knew that (asking for 'SubDir' to print out (see code and output I pasted herein). 

Having the 'and subdir=' in the exclude statement gives me more clarity though that these records are the ones we want to keep, only trying to find the odd ones out. 

select c=user s=omvs                                             
  exclude key=#CIA0A 

Hello Rob!

Just wanted to summarize again (for my own clarity, sanity) using the 2-pass code you helped me with Rob works.  The generated code properly generates all the 'exclude' statements where we found the RACF ID 'UserID' matches the '/u/home/userid' and is not case-sensitive. For example:

 User         Name                                              Uid  HomeDir                  SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235  /u/home/#cia0a             

RACF ID User "#CIA0A" is correct in our scheme for the proper name for the HomeDir "/u/home/#cia0a" so the code generates an 'exclude' statement as:

exclude key=#CIA0A and subdir=#CIA0A

However, once the 2-Pass code generated is submitted and run it does NOT 'exclude' those in the report.   I think that's what I'm aiming for with this report, trying to find all 'HomeDir' values that don't represent the RACF ID, and also to find any 'uppercase' values in the 'HomeDir'.   

Many thanks Rob for reading thru my ramblings.  

Lynn

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------

Hello Rob,

I think the problem was in the 'exclude'.  I altered it to this syntax:

summary each,                                                       

  '  exclude ( key=' | key(0) ' and home=' | home(0) ')'             

and with this formatted exclude statement generated by the 2-pass code:

select c=user s=omvs

 exclude ( key=#CIA0A    and home=/u/home/#cia0a ) 

It worked.  the 'and' needed to be in parentheses.  

I'll continue to work this to see if it's what these users want.  

Many thanks for your 'bounce back' on my ramblings. 

Lynn

Rob,

Last minute update...  I removed the parameter 'and subdir=#CIA0A'  and the generated 'exclude' statements were in fact excluded.  I guess either the 'subdir=' value is not right or does not contain data, or the 'and' statement does not work on the 'exclude' statement?  I simply removed the 'and subdir=#CIA0A' from a group of statements and they were successfully excluded from the report.  You probably already knew that (asking for 'SubDir' to print out (see code and output I pasted herein). 

Having the 'and subdir=' in the exclude statement gives me more clarity though that these records are the ones we want to keep, only trying to find the odd ones out. 

select c=user s=omvs                                             
  exclude key=#CIA0A 

Hello Rob!

Just wanted to summarize again (for my own clarity, sanity) using the 2-pass code you helped me with Rob works.  The generated code properly generates all the 'exclude' statements where we found the RACF ID 'UserID' matches the '/u/home/userid' and is not case-sensitive. For example:

 User         Name                                              Uid  HomeDir                  SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235  /u/home/#cia0a             

RACF ID User "#CIA0A" is correct in our scheme for the proper name for the HomeDir "/u/home/#cia0a" so the code generates an 'exclude' statement as:

exclude key=#CIA0A and subdir=#CIA0A

However, once the 2-Pass code generated is submitted and run it does NOT 'exclude' those in the report.   I think that's what I'm aiming for with this report, trying to find all 'HomeDir' values that don't represent the RACF ID, and also to find any 'uppercase' values in the 'HomeDir'.   

Many thanks Rob for reading thru my ramblings.  

Lynn

Hello Rob,

Received your suggestions.  I modified the code to the following:

n type=racf nopage dd=ckr2pass                                        
 define once(nd) boolean where key=true                               
 define each(nd) count                                                
 define subdir(char) as word(home,3,"/") where home=:"/u/home/" and   
        missing(word(home,4,"/"))                                     
 s c=user s=omvs exists(home)                                         
 summary once,                                                        
  'newlist name=OMVSRQ1,',                                            
/ '  tt="zSecure Suite USER OMVS Segment HDir UserID overview",',     
/ '  st="Users HDir=/u/home/ does not match user ID"',                
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and      
           missing(word(home,4,"/"))',                                
/ ' select c=user s=omvs'                                             
 summary each,                                                        
  '  exclude key=' | key(0) 'and subdir=' | key                       
 summary once,                                                        
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',      
/ '  key(8,"User",key,pas) key:pgmrname,',                            
/ '  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")'

This generated CKR2PASS code as:

newlist name=OMVSRQ1,                                             
  tt="zSecure Suite USER OMVS Segment HDir UserID overview",      
  st="Users HDir=/u/home/ does not match user ID"                 
 define subdir as word(home,3,"/") where home=:"/u/home/" and     
        missing(word(home,4,"/"))                                 
 select c=user s=omvs                                             
  exclude key=#CIA0A and subdir=#CIA0A                            
  exclude key=#CIA0B and subdir=#CIA0B                            
  exclude key=#CIA0C and subdir=#CIA0C                            
  exclude key=#CIA0D and subdir=#CIA0D                            
  exclude key=#CIA0E and subdir=#CIA0E                            
  exclude key=#CIA0F and subdir=#CIA0F                            
  exclude key=#CIA0G and subdir=#CIA0G                            
  exclude key=#CIA0H and subdir=#CIA0H                            
  exclude key=#CIA0I and subdir=#CIA0I                            
  exclude key=#CIA0J and subdir=#CIA0J                            
  exclude key=#CIA00 and subdir=#CIA00                            
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
  key(8,"User",key,pas) key:pgmrname,                             
  uid(10,num,"       Uid") home(20,"HomeDir") subdir(20,"SubDir")

I removed several hundred lines just to shorten it up and test it leaving only those 11 'exclude' statements.  This produced a report as:

zSecure Suite USER OMVS Segment HDir UserID overview - complex TPX      10Dec2024 02:00  
Users HDir=/u/home/ does not match user ID                                               
                                                                                         
User     Name                        Uid HomeDir              SubDir                     
#CIA0A   CICS SCHEDULER ID       1002235 /u/home/#cia0a                                  
#CIA0B   CICS SCHEDULER ID       1002236 /u/home/#cia0b                                  
#CIA0C   CICS SCHEDULER ID               /u/home/#cia0c                                  
#CIA0D   CICS SCHEDULER ID               /u/home/#cia0d                                  
#CIA0E   CICS SCHEDULER ID       1003082 /u/home/#cia0e                                  
#CIA0F   CICS SCHEDULER ID       1003083 /u/home/#cia0f                                  
#CIA0G   CICS SCHEDULER ID       1003084 /u/home/#cia0g                                  
#CIA0H   CICS SCHEDULER ID       1003085 /u/home/#cia0h                                  
#CIA0I   CICS SCHEDULER ID       1003086 /u/home/#cia0i                                  
#CIA0J   CICS SCHEDULER ID       1003087 /u/home/#cia0j                                  
#CIA00   CICS SCHEDULER ID       1001923 /u/home/#cia00                                  
#CIA00QF CICS SCHEDULER ID       1003518 /u/home/#cia00qf                                
#CIA00QV CICS SCHEDULER ID       1003517 /u/home/#cia00qv                                
#CIA00TF CICS SCHEDULER ID       1003537 /u/home/#cia00tf                                
#CIA00W  CICS SCHEDULER ID       1001924 /u/home/#cia00w                                 
#CIA00WF CICS SCHEDULER ID       1003538 /u/home/#cia00wf  

.

.

.

several hundred entries.   

I hope this help with the 'exclude'.  That 'SubDir' I included comes up blank.  

Many thanks!

Lynn

Yes, success after removing the NAME=OMVSGEN parameter.  No error.  Moving on to my 'exclude' processing in the 'gen'd code now.  Many thanks for your help!  

OK, so there was a fix removed from CKRCARLA after I left IBM.  I fixed the issue related to

CKR0391 12 Duplicate NEWLIST NAME=OMVSGEN at CKRCMDV line 1, already defined at CKRCMDV line 1 

but that fix is no longer there.  To get the code I gave you going, you have to remove the parameter NAME=OMVSGEN from the first line, so:

n type=racf nopage dd=ckr2pass

Next:

 define once(nd) boolean where true
CKR0128 12 Expecting relational operator or "(" instead of end of line at CKRCMDV line 2       
CKR0987 12 Syntax error: text expected instead of end of line at CKRCMDV line 2                

Can you try if this error still occurs after you removed the NAME parameter?  If it is still there, to change line 2 into 

define once(nd) boolean where complex==complex

------------------------------
Rob van Hoboken

Original Message:
Sent: Tue December 10, 2024 01:10 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello Rob, 

Thanks very much for taking a look at this.  I'm still working it but here are my results so far:

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442

Original Message:
Sent: Tue December 10, 2024 11:42 AM
From: Rob van Hoboken
Subject: CARLa for Reporting on OMVS Segments

Hi Lynn.

Normally, you can use the == operator to compare the contents of one field with another.  You can also use DEFINE to collect a substring or other token from a string.  So if the uppercase/lowercase issue didn't apply, you would write your report like so:

newlist name=OMVSRQ1,
 tt="zSecure Suite USER OMVS Setment HDir UserID overview",       
 st="Users HDir=/u/home/ does not match user ID"
 define masklist as profile
 define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))
 s c=user s=omvs subdir<<>>masklist
 sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),       
   key(8,"User",key,pas) key:pgmrname,                            
   uid(10,num,"       Uid") home(20,"HomeDir")   

However, the profile key (user ID) is in upper case and the 3rd word in the directory is lower case,  and CARLa's == and <<>> operators compare field contents with field contents, without accepting different case values  (I believe).

Luckily, when you compare a field value with a text (literal) value, the SELECT command first converts both values to upper case.  So your report can be constructed as a two pass query, like so:

n type=racf name=OMVSGEN nopage dd=ckr2pass 
 define once(nd) boolean where true
 define each(nd) count

 s c=user s=omvs exists(home)

 summary once,
  'newlist name=OMVSRQ1,',
/ '  tt="zSecure Suite USER OMVS Setment HDir UserID overview",',
/ '  st="Users HDir=/u/home/ does not match user ID"',
/ ' define subdir as word(home,3,"/") where home=:"/u/home/" and missing(word(home,4,"/"))',
/ ' select c=user s=omvs'

summary each,
'  exclude key=' | key(0) 'and subdir=' | key

summary once,
  ' sortlist " - complex"(tt,page) complex(tt,page) stamp(tt),',
/ '  key(8,"User",key,pas) key:pgmrname,',
/ '  uid(10,num,"       Uid") home(20,"HomeDir")'

This code generates a CARLa program that selects the OMVS segment of all users, and excludes the ones where the user ID value is found in the 3rd node of the home field.  Once this CARLa program is shown, you can issue the RUN command in the command line to produce the report.

Note: I have typed this from memory and not tried it.  There are probably quotes of commas missing.  Better some idea than no solution, right?

------------------------------
Rob van Hoboken

Original Message:
Sent: Mon December 09, 2024 07:08 PM
From: Lynn Gilson
Subject: CARLa for Reporting on OMVS Segments

Hello zSecure Community!

I have been asked to produce a report to discover all the OMVS HDir values (/u/home/userid) where the userid (lowercase or uppercase) is compared to the UserID and see the list of ones that DO NOT match.  A second 'task' was to generate the same list but have 'uppercase' values in the HDir home directory value (/u/home/USERID).  "Ours (mine) is not to reason why, ours (mine) is but to do and ..."  

Using the zSecure panels,  I was able to fashion a program as:

--------------------------------------------------------------------------

-------------------------------------------------------------------------- 

... and this produced a report as:

--------------------------------------------------------------------------

------------------------------
Lynn Gilson
SysProg
Ensono Corp.
Downers Grove, IL,USA
+001-630-944-1442
------------------------------