Thread: HEAD version of initdb fails on Win32

HEAD version of initdb fails on Win32

From
Thomas Hallgren
Date:
I get the following error message when I try to do an initdb with CVS 
HEAD on a Win32 platform:

creating configuration files ... ok
creating template1 database in data-head/base/1 ... FATAL:  access 
method "btree" does not support unique indexes
child process was terminated by signal 1

Regards,
Thomas Hallgren


Re: HEAD version of initdb fails on Win32

From
Tom Lane
Date:
Thomas Hallgren <thhal@mailblocks.com> writes:
> I get the following error message when I try to do an initdb with CVS 
> HEAD on a Win32 platform:

> creating configuration files ... ok
> creating template1 database in data-head/base/1 ... FATAL:  access 
> method "btree" does not support unique indexes
> child process was terminated by signal 1

Try "make distclean" and a full rebuild.
        regards, tom lane


Re: HEAD version of initdb fails on Win32

From
Thomas Hallgren
Date:
Tom Lane wrote:

>Thomas Hallgren <thhal@mailblocks.com> writes:
>  
>
>>I get the following error message when I try to do an initdb with CVS 
>>HEAD on a Win32 platform:
>>    
>>
>
>  
>
>>creating configuration files ... ok
>>creating template1 database in data-head/base/1 ... FATAL:  access 
>>method "btree" does not support unique indexes
>>child process was terminated by signal 1
>>    
>>
>
>Try "make distclean" and a full rebuild.
>  
>
Much better :-) Thanks!

- thomas



Re: HEAD version of initdb fails on Win32

From
Andrew Dunstan
Date:

Tom Lane wrote:

>Thomas Hallgren <thhal@mailblocks.com> writes:
>  
>
>>I get the following error message when I try to do an initdb with CVS 
>>HEAD on a Win32 platform:
>>    
>>
>
>  
>
>>creating configuration files ... ok
>>creating template1 database in data-head/base/1 ... FATAL:  access 
>>method "btree" does not support unique indexes
>>child process was terminated by signal 1
>>    
>>
>
>Try "make distclean" and a full rebuild.
>
>    
>  
>

Incidentally, thanks to Jim Buttafuoco, we now have a Windows box 
(goose) on the buildfarm building every day, so when things look screwy 
you can check there to see if other people are having the same problem.

cheers

andrew


Re: HEAD version of initdb fails on Win32

From
Tom Lane
Date:
Thomas Hallgren <thhal@mailblocks.com> writes:
> Tom Lane wrote:
>> Try "make distclean" and a full rebuild.
>> 
> Much better :-) Thanks!

Always the first thing to try if you get a weird failure after CVS update.
(Personally I never even try to do a partial rebuild after updating;
I can overlap recompiling with doing other things, and the time wasted
tracking even one false bug isn't worth it.)
        regards, tom lane


Re: HEAD version of initdb fails on Win32

From
Tom Lane
Date:
Andrew Dunstan <andrew@dunslane.net> writes:
> Incidentally, thanks to Jim Buttafuoco, we now have a Windows box 
> (goose) on the buildfarm building every day, so when things look screwy 
> you can check there to see if other people are having the same problem.

Indeed, I checked the buildfarm status before opining Thomas probably
just needed to rebuild ;-)
        regards, tom lane