Creating data sets and members from /dsfs

• One minute mvs: data set file system (/dsfs) on z/OS
• Setting up data set file system /dsfs on z/OS
• Creating data sets and members from /dsfs
• Problems I experienced when using /dsfs

You can configure defaults when using dsfs to create data sets and members. For example

Creating a PDS(E)

cd /dsfs/txt/colin
dsadm createparm -path . -pdsmodel "dsntype(library) dsorg(po) lrecl(133) recfm(vb)      
blksize(0)"                                                                                 

For path /dsfs/txt/colin (-path .) it creates a model for pds (-pdsmodel) with the parameters dsntype…blksize(0)

Note:blksize(0) says “pick the best block size”

You can display this using

dsadm fileinfo   -path /dsfs/txt/colin  

and get

path: /dsfs/txt/colin                                                                                
fid                    49,1          anode                  1285,4548                                
length                 16384         format                 BLOCKED                                  
1K blocks              16            dir tree status        VALID                                    
PDS model anode        31,60         PS model anode         0,0                                      
object type            DIR           object linkcount       181                                      
object genvalue        0x00000000    dir version            1                                        
dir name count         431           data set name type     HLQ DIR                                  
recfm                  na            lrecl                  na                                       
data mode              TEXT          data set status        RETRIEVED                                
data set name          COLIN                                                                         
ENQ held               NO                                                                            
direct blocks          0x00001BAF    0x00001BB0    0xFFFFFFFF    0xFFFFFFFF                          
                       0xFFFFFFFF    0xFFFFFFFF    0xFFFFFFFF    0xFFFFFFFF                          
indirect blocks        0xFFFFFFFF    0xFFFFFFFF    0xFFFFFFFF                                        
mtime       Sep  9 10:09:55 2024     atime        Sep  9 10:09:55 2024                               
ctime       Sep  9 10:09:55 2024     create date  Sep  9 2024                                        
not encrypted                        not compressed                                                  
PDS model              dsntype(library) dsorg(po) lrecl(133) recfm(vb) blksize(0)                  
PS model               na                                                                            
vnode,vntok          0x00000050,,0x072003C0      0x0099A7F0,,0x00000000                              
opens                oi=0          rd=0          wr=0                                                
file segments        na         file unscheduled        na                                           
meta buffers         0          dirty meta buffers      2                                            

Creating a sequential file

dsadm createparm -path /dsfs/txt/colin -psmodel "dsorg(ps) lrecl(80) recfm(fb) blksize(3200)"

The command

dsadm fileinfo   -path /dsfs/txt/colin 

now gives

PS model               dsorg(ps) lrecl(80) recfm(fb) blksize(3200)    

The command

echo "hi" > /dsfs/txt/colin/seq2          

creates a data set COLIN.SEQ2 with characteristics

Record format . . . : FB
Record length . . . : 80
Block size  . . . . : 3200

If you specify blksize(0) the system applies a default. In my case Block size . . . . : 27920, which makes good usage of the disk space.

Different data set types

There are three data set types

• bin – binary for example load modules
• rec – records
• txt – text files, with a new line character “at the end of the line”

You can define the data set characteristics for each of these.