[ELGIS] /usr/lib64/libspatialite.so.2: undefined symbol: sqlite3_clear_bindings in Unknown on line 0

[Pavel Iacovlev]

Sorry I forgot to post the link to the related bug "
https://bugs.php.net/bug.php?id=40926" it's not the same problem because it
does not happen on connection close, but it gave a few clues of what might
be happening.

I am testing all this on a virtual machine, so only required software to
run our application was installed. I can start from a clean once again to
test it out (did this already once thought).

About that you can't reproduce this behavior, could you install one more
package that could affect this: php53-pdo (this should install pdo_pgsql
and pdo_sqlite).


On Tue, Jan 10, 2012 at 5:07 PM, Peter Hopfgartner <
peter.hopfgartner at r3-gis.com> wrote:

> **
> On 01/10/2012 03:41 PM, Pavel Iacovlev wrote:
>
> Separate problems, first problem in that php-mapserver from stable elgis
> does not work because it was compiled against spatialite from elgis-testing.
>
>  The second is in regard to that pg_connect crashes when php-mapserver is
> installed from elgis-testing, this can be fixed quite easily (my renaming
> the mapserver.ini) and if this fix should be integrated or not ?
>
>  Hi Pavel,
>
> I'm glad you were able to track this down. I still have not a clean
> picture what is happening. Where does the SSL-conflict come from? Is it,
> that some packages were compiled against different versions of openssl?
> As a reminder, I was not able to reproduce te problem, so it might be
> specific to your installation. Did you enable any other repository, except
> EPEL, ELGIS and the base distriution?
>
>  On Tue, Jan 10, 2012 at 4:34 PM, Mathieu Baudier <mbaudier at argeo.org>wrote:
>
>> > I found a similar bug but instead of php conflicting with mapserver's
>> SSL it
>> > was php conflicting with CURL's SSL.
>> >
>> > So I renamed mapserver.ini to z_mapserver.ini (this will make it load
>> after
>> > pgsql.ini) and the segmentation fault is gone.
>>
>>    I would consider to be rather hackish. If there is no better
> understanding of the problem, could consider this. But it's not really
> clean.
>
> Great work,
>
> thanks, Pavel
>
>    Great!
>> So, are you unblocked now?
>>
>> > Question 1: Should this fix be applied to ELGIS ?
>> > Question 2: What about php-mapserver that IS NOT in ELGIS-PLUS at the
>> moment
>> > it does not work at all, will be there any actions taken in regard to
>> this,
>> > or the package will be removed all together.
>>
>>  If we fix using what you suggest in Question 1, then is Question 2
>> still relevant?
>> Or are you talking about two separate problems?
>>
>> I have booked:
>> http://trac.osgeo.org/el/ticket/9
>> to follow this up
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/el/attachments/20120110/8380ab65/attachment-0001.html