[Firebird-odbc-devel] from [Firebird-devel] Bug in driver

[Vladimir T. <pr...@hm...>]

1)==================================
>  Mel wrote:
>
> I have found the following issue with the driver (it appears in the V1.2,
> 1.2.1,1.3), operating in read committed mode.
> If i open a (browse list) thread (A) update a particular record, open
another
> (browse list) thread(B) and update the same record (committing in both
> cases). Now if i go back to thread A, and do a get (select) on the
database,
> I will get the data that it saved back and not =A0B. If I then try and
update
> the record I get the =A0'deadlock conficts with concurrent update' error.
I can
> keep opening new threads and they are all fine, but if they update a
record
> any previous threads will have the same problem.
>
> Even if I A and B are separate apps on separate machines the problem is
the
> same and even if B totally shutdown I still get the error (and A's inital
snapshot of data)
>
> I have tested the same code with the Gemini Evaluation driver which does
not
> suffer the problem.
>
> I am developing in Clarion and running Firebird 1.5.1 on Gentoo linux and
the
> client is Windows XP Pro with SP2 .

2)==================================
> Hi, Mel!
>
> > I will get the data that it saved back and not =A0B. If I then try and
> > upda=
> > te=20
> > the record I get the =A0'deadlock conficts with concurrent update'
error.
>
>     However everyone want just this answer - > 'deadlock conficts with
> concurrent update' :-)
>     If it is not necessary for you,  please, set Off option NOWAIT from
> you'r DSN of ODBC Administrator
>
> --
> Best regards,
> Vladimir Tsvigun
3)==================================
> Sorry, should have pointed the other options out. No wait is off (if i
leave
> it on the whole thing hangs instead of the error message), I am using
dialect
> 3, quoted identifier is off, the character set is left at NONE and I am
> connecting with servername:alias.
>
> I have used this driver  with other simple programs using the same
language /
> development environment and server without this problem, however this is a
> relatively large and complex application. Although theres still the chance
> that it is my code I have exausted all the debugging and testing I can do,
> and as i stated first, I dont have the same problem with the Gemini
driver,
> and I dont believe that it is actually a deadlock issue despite the error
> message becuase I can update the record in new instances of the app and
even
> if I shut the everything but the original instance down, it still gets the
> error and the data that it first knew about(snapshot), not the updated
data.
>
> Mel

    Now I shall try to execute more tests
    I shall inform result later

--
Best regards,
Vladimir Tsvigun