Showing posts with label collation. Show all posts
Showing posts with label collation. Show all posts

Tuesday, March 27, 2012

Error publishing databases with not-server's default collation

Hello.
While trying to publish any database that uses a collation different from
server's default, I get this error in the step after selecting the articles
to be published:
Error 21423: You do not have sufficient privileges to view the publication
information. Your administrator may need to fix the PAL role on the publisher
for publication 'database_name'.
I've being trying to track down this error for a while now, and I've just
noticed it's related to the collation...
Any ideas about this?
Not sure if it's relevant here, but when I had a case-sensitive server
collation I noticed that passwords of logins became case sensitive on
authentication. I wonder if there is somehow a login/user case sensitivity
issue at work here also? Are the 2 collations the same in this regard or
different (a mixture of _CS and _CI)? Please can you post up the entire
collation names for the circumstance that doesn't work.
Rgds,
Paul Ibison SQL Server MVP, www.replicationanswers.com
(recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)
|||The collations are:
Chinese_Taiwan_Stroke_BIN for the dbs that fails
and:
SQL_Latin1_General_CP1_CI_AS for the dbs that are sucessfuly published.
In any case, the passwords for the accounts are all lowercase...
Thanks for your support,
RODOLFO
"Paul Ibison" wrote:

> Not sure if it's relevant here, but when I had a case-sensitive server
> collation I noticed that passwords of logins became case sensitive on
> authentication. I wonder if there is somehow a login/user case sensitivity
> issue at work here also? Are the 2 collations the same in this regard or
> different (a mixture of _CS and _CI)? Please can you post up the entire
> collation names for the circumstance that doesn't work.
> Rgds,
> Paul Ibison SQL Server MVP, www.replicationanswers.com
> (recommended sql server 2000 replication book:
> http://www.nwsu.com/0974973602p.html)
>
>

Wednesday, February 15, 2012

Error message

Hello,

When I try to run a query on a server, I get the following error:

"Cannot resolve collation conflict for equal to operation."

The same query, but on another server works without any problems.

Any idea how I can resolve this?You are joining two different coallations which each other. To solve the collation conflict you will have to either change the collation on the server / database / table / or expression server you are using (whereas the latter is the quickfix for your problem. It has be to implemented whereever you are joining different collations:

select ca.account_status, cs.code from customer as ca
inner join v_customer_status as cs on ca.account_status = cs.description COLLATE SQL_Latin1_General_CP1_CI_AI

HTH, Jens K. Suessmeyer.

http://www.sqlserver2005.de|||Thanks Jens.

But the query contains tables of the same database.|||That is not just related to different databases. Also just columns within a table can have a different collation. QUery the INFORMATION_SCHEMA.Columns to see which Collation they have and use the mentioned keyword and syntax to use a common collation.

HTH, Jens.|||txs Jens.

I saw some different entries in the syscolumns under the field collation.
Unfortunately I cannot change the collation name.

What can I do (apart from changing my queries)?|||Is there any specific reason why you have different collation settings for tables in the same database? Unless you're willing to "sweat it out" the easiest way is to change the collation settings of one table to be the same as the other.|||I copied the tables from another server, apparently with a different collation setting.

How can I change this setting per table?|||I think there is no command for changing every column within the table. YOu will have to change this by column with the command:

ALTER TABLE <tablename> ALTER COLUMN colname <datatype> COLLATE <collationname>

HTH, Jens K. Suessmeyer.

http://www.sqlserver2005.de