Forum

We had 1C base for development in file mode, but now we need to install 1C enterprise server.

We installed 1C server and PostgreSQL database server on CentOS. Everything works fine when we creating 1C database with english or russian language in regional settings field in 1C database properties.
But the problem appears when we are trying to cre ate database with spanish language.
When we chose Spanish(Ecuador) in regional settings in database creation dialog we have an error message:

Error creating infobase:
Infobase error
Error setting or changing Infobase regional settings
Required collation is not supported by database.

Error setting or changing Infobase regional settings
Required collation is not supported by database
because:
Required collation is not supported by database

We are investigating your problem.

PostgreSQL does not support creating databases with collations which differ from the PostgreSQL's cluster collation. So the cluster's collation must be Spanish (Equador) too.

I initialized PostgreSQL cluster as

initdb -D /var/lib/pgsql/data --locale=es_EC.UTF-8

 X List of databases
   Name    |  Owner   | Encoding |  Collation  |    Ctype    |   Access privileges
-----------+----------+----------+-------------+-------------+-----------------------
 1c_db_    | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 |
 1c_db_ec  | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 |
 1c_db_ec0 | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 |
 postgres  | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 |
 template0 | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 | =c/postgres          +
           |          |          |             |             | postgres=CTc/postgres
 template1 | postgres | UTF8     | es_EC.UTF-8 | es_EC.UTF-8 | =c/postgres          +
           |          |          |             |             | postgres=CTc/postgres

The problem is that localeName es_EC corresponds to collate  eModern_Spanish while pglang es corresponds to collate eLatin1_General.
When those collations are compared the error is displayed.

Thank you, but could you give me some advice how to fix it?
As I understand, 1C is trying to cre ate database in UTF8 encoding, but ModernSpanish and Latin1_General are kinds of windows-1252 encoding? And in postgresql show all I can see that lc_collate = es_EC.UTF-8. When I initialize cluster with locale=es_EC I have another error: Error creating infobase: encoding UTF8 doesn't match locale es_EC.

And more interesting question: why infobases are being created successfully only from the second try?

Sorry for the delay, our engineers are still investigating your problem.

Hello, Mikhail!

The temporary solution is to try to create the infobase with es_AR (spanish language + Argentina country) or es (spanish language, no country) encoding.

Hello. I tried to compile postreSQL 9.1.2 from sources on CentOS 6.5 x64 and i managed to initialize postgreSQL cluster with es_EC.UTF8 collation.