Re: DBD::Oracle installation - make test failed - ORA-28000: accout locked

On Jun 15, 8:32 am, c...@xxxxxxxxx (Charles Jardine) wrote:
On 14/06/07 11:38, John Scoles wrote:

seems this is a bug in the code when dealing with Solaris. So I would
just ignore it for now

here is a link to it

http://www.nntp.perl.org/group/perl.dbi.dev/2006/01/msg4401.html

The above URL points to a posting of mine in which I report a
bug in the the test suite, and provide a fix. This fix was
incorporated into the released version of DBI 1.50.

It is the fixed test which is now failing. As the code says,
the purpose of the test is

# basic check that we can fork subprocesses and wait for the status

t/10general.............NOK
1
# Failed test 'system exit 1 should return 256'
# in t/10general.t at line 31.
# got: '-1'
# expected: '256'
t/10general.............NOK
2
# Failed test 'system exit 0 should return 0'
# in t/10general.t at line 32.
# got: '-1'
# expected: '0'

The result of the test suggests that we cannot Fork subprocesses
and wait for the status. 'perldoc -f system' says:

Return value of -1 indicates a failure to start the program
(inspect $! for the reason).

Something is very seriously wrong if system('exit 0;') is failing.

I am afraid I do not have the time to investigate this myself at
the moment, but it does need investigating. A good start would be
to modify the test to print the value of $! after the failure.

--
Charles Jardine - Computing Service, University of Cambridge
c...@xxxxxxxxx Tel: +44 1223 334506, Fax: +44 1223 334679

Thanks to Charles and John.

I then took this as a bug and went ahead to have the DBD installed.
Seems now it recognizes the DBD module, but my script met another
problem about connection: ORA-12154: TNS:could not resolve the connect
identifier specified (DBD ERROR: OCIServerAttach). I guess it's a
problem in the Oracle installation or configuration itself and will
seek for solutions in other groups I guess.

Thanks again, and please let me know if you have suggestions or
updates.

Alan

.

Relevant Pages

  • Re: use assert and defensive together
    ... It would be better if you used different return codes for each type of failure, but I'm not going to fix al of your code. ... because the calling program is notified and will report to the user that it failed and if the calling program is any good at all will report the reason for the failure. ... But by confusing programming errors with incorrect paths, you are bug hiding. ... The function is therefore incorrect because it should use them or incorrect because it takes them. ...
    (comp.lang.c)
  • Re: just a warning?
    ... connection can have this problem if there's a transient failure ... connection, but haven't seen it since installing dapper. ... grep - GNU grep, egrep and fgrep ... the bug will be fixed in breezy one day. ...
    (Ubuntu)
  • Re: What to do with a failed automated test
    ... one of then is failing. ... After discussion, the bug is ... report that was produce base on the initial failure. ...
    (comp.software.testing)
  • Re: buggy optimization levels...
    ... > the optimization that caused the observed failure is thus also hard. ... Note that it is not necessarily gcc which is at fault for such ... It may be a bug in gcc, but it may also be a bug in the code ...
    (freebsd-questions)
  • Re: gets(); was: Re: Why does rewind() ignore errors?
    ... When you can't make failures completely impossible, the best you can do is compare the probablility a failure and the cost of its consequences with the cost of trying to minimize the chance of a failure, and draw the line somewhere. ... Costs of a line too long because of some program bug using gets: ... The debugging costs are trivial since a clear error ...
    (comp.std.c)