Hi
We are trying to download data from Adabas files using Natural Program.
READ <table_name> BY ISN
The extraction process fails with the error code NAT3072.
Can you please suggest on how to rectify this?
Thanks
Selva
Selva,
The NAT3072 is related to user queue (or the user file list pool) overflow. You may want to seek help from your DBA regarding ADARUN NU parameter set up.
However, I am interested to know what amount of data you are after to download and why preferring READ BY ISN?
To add,
Adabas assigns a user queue element (UQE) to each active user. There may be a possibility that number of users active at the same time, is greater than the number specified with NU parameter (when you ran your program).
You may want to run your program in a off-peak time and check if it works.
Hi Ats,
Table data volume (no of records)
PROMISE-T1 = 393
PROMISE-T2 = 59625
PROMISE-T3 = 902123
PROMISE-t4 = 28
Have ran the program initially on Sunday which is an off peak time and also tired on Monday and Thursday.
READ (10) <table_name> by ISN.
WRITE <column_name>
END-READ
But still the problem exists.
Thanks
Selva
Selva,
Sounds very strange to me :o
I guess if Natural abends or some such and restarts then user-queue elements may be left over. Discuss this with your DBAs/Admin and check what NU value is set up to? Let us know what they say on this.
The way could either restart the database (painful 
) or change the NU setting to be higher (and restart the database)
Experts do you have any additional point to add?
Everything said here is correct. If you are getting a 3072 (Adabas response code 72), it means the user queue is full.
Besides having the DBA increase the number of user queue elements allowed (NU parm), you can have him/her display the user queue entries to see if there are old entries in need of being purged.
The DBA can do this through SYSAOS or ADADBS OPERCOM DUQA, and then deleting those dead UQE’s.
I assure you that if you have to have the DBA do this often enough, he/she will increase NU. 