ADABCK Restore fails with ADABAS_077 Error


Please excuse my lack of db understanding or if I do not use the correct Adabas terminology.

I am a VMS systems administrator with very limited db experience assigned to do D/R testing for VMS servers with an Adabas db. We have several sites running OpenVMS v6.2-1H3 on Alphaserver 800s and Adabas v3.2.1. The dbs were setup and configured long ago by others who have left. We have a staff of DBAs but they only support “Oracle” and “SQL” so I am on my own for this.

I have tried to do several restores on test machines with limited success; sometimes the restore works, sometimes not. It seems that if I restore all files expect for file 1, the restore works. But if I try to restore file 1, the restore fails.

The steps to restore have been:

  1. Restore disks from backup tape
  2. Delete/rename old db data files
  3. Run DBGEN.COM to create the db template and load files 1-2
  4. Restore Adabas db from BCK001:
    $ ADABCK DBID=086, RESTORE=(5-218) ! exclude files 1-2

This process works but I don’t know if the restore is complete or if it is missing anything.

If I do not load files 1-2 with DBGEN.COM and I try to RESTORE=(1-2) or OVERLAY=(1,2) the restore gives the error:

%ADABCK-E-ADABAS_077, * ASSO/DATA storage exhausted
%ADABCK-I-IOCNT, 6 IOs on dataset BCK001
%ADABCK-I-IOCNT, 20 IOs on dataset ASSO
%ADABCK-I-TERMINATED, 11-NOV-2005 15:58:45.67, used cpu-time : 00:00:00.03

The container created with DBGEN has larger sizes than that of the original db so that should not be the problem, right?

The backup command used to create the backup file is:
$ abck db=086,dump=*

Should I be able to do a similar wildcard command to restore the db from the bck001? i.e.

Should files 1-2 be loaded from DBGEN or restored from BCK001? If BCK001, how do I eliminate the “storage exhausted” error?

Can someone tell me what I am missing or what the correct restore process should be?

Is there a simple method of creating a db from a BCK001 file?

I can provide db summary and contents for db and bck001 if it would help.

Thanks to all.

With Adabas v3.3 the target database must contain at least as many blocks of each blocksize as the files had in the source database.

With Adabas v5.1 the target database is increased automatically, if it is too small.