Rev 2.7 Bug (Crash)

[Jan Schenkel]

--- "Brian K. Maher" <plsntbreez at mac.com> wrote:
> Hi Folks,
> 
> I was testing a SQL locking issue and found that the
> revDataFromQuery function will cause Rev to crash if
> the query fails due to a locked record.  My ODBC
> connection specifies "repeatable read" as the
> default isolation level in the ODBC DSN
> configuration.  The code is as follows:
> 
> on mouseUp
>   put revOpenDatabase("ODBC","S2K",,"xyz","xyz")
> into vDB
>   
>   try
>     put revDataFromQuery(,,vDB,"SELECT * FROM
> PUB.Customer") into vResult
>     answer (line 1 of vResult)
>   catch vError
>     answer vError
>   end try
>   
>   revCloseDatabase vDB
>   
>   answer "Done"
> end mouseUp
> 
> Cheers, Brian
> 

Hi Brian,

If I might venture to guess, you're using ODBC to
connect your Revolution stack to a Progress database.
While I haven't seen Rev crash with it, I do know that
using ODBC on Progress can cause lock-table overflow,
affecting your Progress4GL applications.

The company I work for has been using Progress for 12
years but always used Progress4GL, and warn users
against the use of ODBC on their operational database.

Our theory is that the ODBC driver is probably
translating the sql query into a for each using
share-lock or even exclusive-lock instead of no-lock
-- unless you set the isolation level to 'Read
Uncommitted'

Which in turn brings up the fascinating subject of
Dirty, Repeatable, and Phantom Reads -- but let's not
get carried away ;-)

Best regards,

Jan Schenkel.

Quartam Reports for Revolution
<http://www.quartam.com>

=====
"As we grow older, we grow both wiser and more foolish at the same time."  (La Rochefoucauld)

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com