Tried to start the db:
csql German -u dba

This results in:
ERROR: Failed to connect to database server, 'German', on the following host(s): localhost

And the csql.err file reads:
Time: 05/23/13 16:37:24.227 - ERROR *** file ../src/communication/network_cl.c, line 4224 ERROR CODE = -191 Tran = -1, EID = 1
Cannot connect to server "German" on "localhost".

Time: 05/23/13 16:37:24.227 - ERROR *** file ../src/transaction/boot_cl.c, line 1561 ERROR CODE = -677 Tran = -1, EID = 2
Failed to connect to database server, 'German', on the following host(s): localhost

*** The previous error message is the last one. ***

What am I doing wrong?

I try to insert data into shards with PHP, but connect using PHP PDO with error:

Error: SQLSTATE[HY000] [-10100] CAS, Unknown error
Code: 4294957196

PHP code:

try {
    $db =new PDO('cubrid:dbname=shard1;host=localhost;port=45511', 'shard', 'shard123' );
    echo "PDO connection created ok!"." ";
} catch(PDOException $e) {
   echo "Error: ".$e->getMessage()." "."Code: ".$e->getCode()." ";
}

But with JAVA example ( http://www.cubrid.org/manual/91/en/shard.html ) everything works fine. JAVA code:

connection = DriverManager.getConnection("jdbc:cubrid:localhost:45511:shard1:::?charSet=utf8", "shard", "shard123");

If in PHP i use port 30000 - everything works fine, data inserted, but without sharding.

What's the problem?

I'm running Ubuntu 12.04 and have downloaded all the php drivers, etc.  Just says unable to connect.  Other than this, I believe I can run the script.

I'm running the command line:  php5 ./imap_redbean.php

and I'm getting the above error.

If you have questions or encounter any issue, feel free to post in this thread.

What is the purpose of first volume when creating database which is generic volume? Why it is not data volume?

What is the usage of generic volume? It means it can hold both data and index?

I'm confused about difference between checkpoint_interval_in_mins and page_flush_interval_in_msecs? They are both writing dirty pages in buffers to data volume at specific interval. 

Does table partitioning cause index partitioning too? Or we have one big index and bunch of small data tables

I'm asking this because in PostgreSQL we should create index for each partition separately. 

Partition pruning section says:

In versions lower than CUBRID 9.0, partition pruning has been executed at the query compiling stage. However, in version of CUBRID 9.0 or higher, it is executed at the server side at the query execution stage

 Does this mean we can not benefit partition pruning when using parametrized query because at compile time arguments are not available?

create table t (

    ID_ varchar(64) not null,

    EMP_ID varchar(64) not null,

    COM_ID varchar(64),

    NAME varchar(64) not null

);

alter table t

   add constraint UNIQ_T_EMP_COM

   unique(EMP_ID , COM_ID );

these inserts should not violate unique constraint
insert ('12', '23', null,  'blah');

insert ('13', '23', null,  'blah');

Is it possible to define that kind of unique constraint? Please point me out.

Thanks

Carter

Want to try Cubrid but.. I cant get the PDO driver to work:

When I run php-cli I get:

I have no clue what this means. I installed the PDO version as described and put the extension .so ref in my php.d/cubrid.ini file.

Any ideas?

It seems like the getUpdateCount method doesn't return an updated value if a select statement follows an update statement. I would expect the updateCount variable to have -1 at the end of the loop but instead it has the result from the update statement.

you are providing a sample database 'demodb' with your installers as well with your virtual machine images. This raises the expectations to have a quick and easy posibility to start and get issue the first queries within a few minutes. But in the end I needed four hours (also caused by other documentation issues posted in this forum).

You do not provide the user with the necessary information to access it (or do not provide links to it). It took a long time to find out, that not my JDBC connection data was wrong, but that I had to start the demodb explicitly (cubrid server start demodb) though the CUBRID service was running. Finally I found user and password ('dba', <empty>) only through a forum question related to accessing a custom instance, not by regular documentation. You can save evaluators a lot of time providing a short 'Getting started with demodb' page or so.

Regards,

Volker

I have the following table:

CREATE TABLE "users"(
"user_id" integer AUTO_INCREMENT,
"user_name" character varying(4096) NOT NULL UNIQUE,
"user_email" character varying(4096) NOT NULL UNIQUE,
"user_nickname" character varying(4096) NOT NULL,
"user_firstname" character varying(4096) NOT NULL,
"user_lastname" character varying(4096) NOT NULL,
"user_points" integer DEFAULT 0,
"user_rank" integer DEFAULT 0,
CONSTRAINT pk_user_user_id PRIMARY KEY("user_id")
);

I want to be able to search users like this:

SELECT
    users.user_name AS username,
    users.user_nickname AS nickname
FROM
    users
WHERE
    users.user_name LIKE 'Joh%' OR users.user_nickname LIKE 'Joh%'

Questions:

1. Is there a better way to do this?

2. How can make this operation case insensitive? If I write 'joh%' I get no result.

Thanks,
Dragos

I looked into the documentation, but I only found support for SHOW CREATE VIEW functionality.

I've found out that for 8.4.0 is not supported, but I cannot find the "New features" for 8.4.1. Could you please point out the correct link?

Thanks,

Cristian

We want to build relationships with the open source communities. The CUBRID Affiliates Program is waiting for your support. If you have questions or would like to discuss about it, please leave your comments below.

I am working with CUBRID and python CUBRID in a W7 environment for over a month and this is the first time when I encounter a problem like this. I can reproduce it all over again.

I have created a view that extracts data from 2 tables:

SELECT
    actions.action_id AS action_id,
    actions.user_id AS user_id,
    actions.page_id AS page_id,
    actions.action_type_id AS action_type_id,
    actions.action_when AS action_when,
    users.user_name AS username,
    users.user_nickname AS nickname
FROM
    actions LEFT JOIN users ON actions.user_id = users.user_id
ORDER BY actions.action_when ASC

In the users table are 3 records:

In the actions table are 5 records:

And here is the view with 5 records:

Now, anytime I want to query the view like this:

SELECT username, nickname FROM actions_users WHERE page_id = 100274

the database stops working:

1. from CUBRID Manager:

2. from web application via python driver

After that, I have to manually start the database, then it works for a few queries and stops working again. An important thing to be considered is that this happens mostly after an INSERT operation in the actions table - the sql can be viewed in the above capture.

Right now I don't know how to handle this. Any ideas? I tried a lot of things: delete view and recreated with a different name, rename columns but the same result: when query the view the database stops.

If is necessary I can provide the full database schema + data and also the SQL commands executed through the python driver.

Thanks a lot,
Dragos

I have downloaded the following VM Images multiple time. When I extract the files, it gives CRC error.

I am using download accelerator plus to download. So, I feel the download is perfect. Please letr me know what needs to be done.

The images

=========

centos-6.2-cubrid-8.4.1-vbox.tar.gz.001

centos-6.2-cubrid-8.4.1-vbox.tar.gz.002 

Regards

- from development server

- from the production server

The records come in the reverse order. Why is that?

Thanks,
Dragos

How can I clone a database? What I want is an exact replica of a database but under a different name.

Thanks,
Dragos

What I need is a DIGG style pagination like here - http://metropotam.ro/La-zi/. There is a python library the implements DIGG pagination but in order to use that with CUBRID I have to do 2 queries instead of one:

1. first count and find out how many records are

2. instantiate digg pagination in python and get the range (e.g. is page number 3 and record from 201 - 300 using 100 records per page)

3. call database again using LIMIT

Is there another solution?

Thanks,
Dragos