Experimental Physics and
Industrial Control System

[email protected] (Rozelle Wright) · Thu, 3 Jul 97 14:44:15 MDT

> From [email protected] Wed Jul  2 18:39:01 1997
> Return-Path: <[email protected]>
> Date: Wed, 02 Jul 1997 17:38:12 -0700 (PDT)
> Subject: Bad CALC record kills IOC
> To: [email protected]
> X-Envelope-To: [email protected]
> X-Vms-To: smtp%"[email protected]"
> Mime-Version: 1.0
> Content-Type> : > TEXT/PLAIN> ; > CHARSET=US-ASCII> 
> Content-Length: 780
> X-Lines: 19
> 
> Yes, we all know that, or knew it at one time. And when it happens to an IOC 
> far away, it becomes painful to identify generally as the problem, as well as 
> finding the specific culprit if many changes were made.
> 
> Could the code which does the check in IOCINIT be extracted and snuggled into 
> a little utility for the developer which would check the CALCs in a given 
> database and ease this tragedy??
> 
> Pointers to code and other advice gleefully accepted.
> 
> /Ron Chestnut
> 
Hi Ron,
Tatsuro Nakamura created  a tool called dbst (db static tool) 
which can be used to check fields
in data bases before they are loaded onto the ioc.
It can also be used to produce .db files that have only non-default
fields and to produce .edf files for capfast.
It is available using anonymous ftp at thoth.atdiv.lanl.gov In the file
/Epics/DBST_1_0.tar
In the same directory there are also statically built files
dbst.static.solaris.tar.gz
and 
dbst.static.sun4.tar.gz.
There is documentation on the epics lanl web pages
at
http://www.atdiv.lanl.gov/aot8/epics/dbase/dbst_1.htm.

I tried running it this afternoon with a database containing
the same CALC field that gave you problems and got the following results.

Beowolf% dbst -acombine.db -w,ai,testresults/ai.defaults.list
Bad Field value 
Error   Last token ")"
input line:             field(CALC,"( (A&C) >> 3 ) - B")
   in: path "."  file link_test.db line 9
   in: path ""  file combine.db line 4

db_parse returned -1
dbReadDatabase error; return code=-1, errno=2

I agree that putting spaces in a calculation shouldn't cause
the ioc to crash, but until that problem is fixed, perhaps this
tool could help you.
Rozelle

---------------------------------------------------------
|                                                       |
|Rozelle Wright              Phone (505) 667-4804       |
|Los Alamos Natl Labs LANSCE-8  FAX (505) 665-5107      |
|PO Box 1663 MS-H820         Group Office (505) 667-6087|
|Los Alamos, NM 87545        email : [email protected]   |
|                                                       |
---------------------------------------------------------

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System