Back when I was still developing the application (CodeIgniter3) in my local computer using Xampp (windows 10) I can connect to the remote SQL Server database through my local computer without any issue. But, when I deploy my application in Centos 9 Apache web server can't seem to connect to the remote SQL Server database.
The message I got when I tried to query it through my ci models by using $this->db->error() command is:
array(2) { ["code"]=> string(11) "08001/10054" ["message"]=> string(73) "[Microsoft][ODBC Driver 17 for SQL Server]TCP Provider: Error code 0x2746" }.
I can ping the SQL Server IP without no issue from my web server, also checked that the required port for SQL Server already opened (1433) in the remote SQL Server by using Telnet. Things I've already tried based on a lot of googling but still won't resolve the issue:
enabling some the httpd_can_network_connect & httpd_can_network_connect_db in selinux
disabling selinux (changed to permissive mode)
changing the config/database.php settings. My current settings is :
$db['default'] = array( 'dsn' => '', 'hostname' => 'xxx.xx.xxx.xx',
'port' => '', 'username' => 'xxx', 'password' =>
'xxx', 'database' => 'xxx', 'dbdriver' => 'sqlsrv',
'dbprefix' => '', 'pconnect' => FALSE, 'db_debug' => FALSE,
'cache_on' => FALSE, 'cachedir' => '', 'char_set' => 'utf8',
'dbcollat' => 'utf8_general_ci', 'swap_pre' => '', 'encrypt' =>
FALSE, 'compress' => FALSE, 'stricton' => FALSE, 'failover' =>
array(), 'save_queries' => TRUE );
Modify /etc/ssl/openssl.cnf config file as follows sed -i -E 's/(CipherString\s*=\s*DEFAULT#SECLEVEL=)2/\11/' /etc/ssl/openssl.cnf since some reference said that it might be a problem with openssl vs. sql-server protocol/version (FYI, my web server openssl version is 3.0.1)
opening port 1433 in my web server through firewalld
Please help me, what exactly is the issue and how do I resolve this?
UPDATE 1
Already checked that i can telnet from my web server to the sql server machine on port 1433.
Regarding the probable cause of openssl tls version not supported, I also check tls version supported on the sql server using command openssl s_client -host ip -port 1433 -tls1 and it said that it is connected. But when i check tls version supported on my Centos Apache web server using the same command (different ip), it gives me error.
Hence, i tried to change the openssl.conf as instructed in https://askubuntu.com/questions/1233186/ubuntu-20-04-how-to-set-lower-ssl-security-level still didnt solve the issue
I believe that I've successfully deployed my (very basic) site to fortrabbit, but as soon as I connect to SSH to run some commands (such as php artisan migrate or php artisan db:seed) I get an error message:
[PDOException]
SQLSTATE[HY000] [2002] No such file or directory
At some point the migration must have worked, because my tables are there - but this doesn't explain why it isn't working for me now.
One of simplest reasons for this error is that a MySQL server is not running. So verify that first. In case it's up, proceed to other recommendations:
Laravel 4: Change "host" in the app/config/database.php file from "localhost" to "127.0.0.1"
Laravel 5+: Change "DB_HOST" in the .env file from "localhost" to "127.0.0.1"
I had the exact same problem. None of the above solutions worked for me. I solved the problem by changing the "host" in the /app/config/database.php file from "localhost" to "127.0.0.1".
Not sure why "localhost" doesn't work by default but I found this answer in a similar question solved in a symfony2 post. https://stackoverflow.com/a/9251924
Update:
Some people have asked as to why this fix works so I have done a little bit of research into the topic. It seems as though they use different connection types as explained in this post https://stackoverflow.com/a/9715164
The issue that arose here is that "localhost" uses a UNIX socket and can not find the database in the standard directory. However "127.0.0.1" uses TCP (Transmission Control Protocol), which essentially means it runs through the "local internet" on your computer being much more reliable than the UNIX socket in this case.
The error message indicates that a MySQL connection via socket is tried (which is not supported).
In the context of Laravel (artisan), you probably want to use a different / the correct environment. Eg: php artisan migrate --env=production (or whatever environment). See here.
I got the same problem and I'm running Mac OS X 10.10 Yosemite. I have enabled the Apache Server and PHP that already comes with the OS. Then I just configured the mCrypt library to get started. After that when I was working with models and DB I got the error:
[PDOException]
SQLSTATE[HY000] [2002] No such file or directory
The reason I found is just because PHP and MySQL can't get connected themselves.
To get this problem fixed, I follow the next steps:
Open a terminal and connect to the mysql with:
mysql -u root -p
It will ask you for the related password. Then once you get the mysql promt type the next command:
mysql> show variables like '%sock%'
You will get something like this:
+-----------------------------------------+-----------------+
| Variable_name | Value |
+-----------------------------------------+-----------------+
| performance_schema_max_socket_classes | 10 |
| performance_schema_max_socket_instances | 322 |
| socket | /tmp/mysql.sock |
+-----------------------------------------+-----------------+
Keep the value of the last row:
/tmp/mysql.sock
In your laravel project folder, look for the database.php file there is where you configure the DB connection parameters. In the mysql section add the next line at the end:
'unix_socket' => '/tmp/mysql.sock'
You must have something like this:
'mysql' => array(
'driver' => 'mysql',
'host' => 'localhost',
'database' => 'SchoolBoard',
'username' => 'root',
'password' => 'venturaa',
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
'unix_socket' => '/tmp/mysql.sock',
),
Now just save changes, and reload the page and it must work!
I encountered the [PDOException] SQLSTATE[HY000] [2002] No such file or directory error for a different reason. I had just finished building a brand new LAMP stack on Ubuntu 12.04 with Apache 2.4.7, PHP v5.5.10 and MySQL 5.6.16. I moved my sites back over and fired them up. But, I couldn't load my Laravel 4.2.x based site because of the [PDOException] above. So, I checked php -i | grep pdo and noticed this line:
pdo_mysql.default_socket => /tmp/mysql.sock => /tmp/mysql.sock
But, in my /etc/my.cnf the sock file is actually in /var/run/mysqld/mysqld.sock.
So, I opened up my php.ini and set the value for pdo_mysql.default_socket:
pdo_mysql.default_socket=/var/run/mysqld/mysqld.sock
Then, I restarted apache and checked php -i | grep pdo:
pdo_mysql.default_socket => /var/run/mysqld/mysqld.sock => /var/run/mysqld/mysqld.sock
That fixed it for me.
The answer from #stuyam solved the "No such file or directory" issue for me
Short answer: Change "host" in the /app/config/database.php file from "localhost" to "127.0.0.1"
But then I had a "Connection refused" error. If anyone had the same issue, my solution for this was to update the app/config/local/database.php file so the port is 8889:
'mysql' => array(
'driver' => 'mysql',
'host' => '127.0.0.1',
'port' => '8889',
'database' => 'databaseName',
'username' => 'root',
'password' => 'root',
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
),
In my case i had no problem at all, just forgot to start the mysql service...
sudo service mysqld start
If you are using Laravel Homestead, make sure you're calling the commands on the server.
homestead ssh
Then simply cd to the right directory and fire your command there.
It worked after I change from DB_HOST=localhost to DB_HOST=127.0.0.1 at .env file
This is because PDO treats "localhost" host specially:
Note: Unix only: When the host name is set to "localhost", then the
connection to the server is made thru a domain socket. If PDO_MYSQL is
compiled against libmysqlclient then the location of the socket file
is at libmysqlclient's compiled in location. If PDO_MYSQL is compiled
against mysqlnd a default socket can be set thru the
pdo_mysql.default_socket setting.
(from http://php.net/manual/en/ref.pdo-mysql.connection.php)
Changing localhost to 127.0.0.1 will "force" the use of TCP.
Note: mysqli_connect is working fine with localhost.
Add mysql.sock path in database.php file like below example
'unix_socket' => '/Applications/MAMP/tmp/mysql/mysql.sock',
Eample
'mysql' => [
'driver' => 'mysql',
'unix_socket' => '/Applications/MAMP/tmp/mysql/mysql.sock',
'host' => env('DB_HOST', 'localhost'),
'port' => env('DB_PORT', '8889'),
Mamp user enable option Allow network access to MYSQL
Building on the answer from #dcarrith ...
Instead of editing the config files, I created an alias in the location that PHP is looking that connects to the real mysql.sock. (source)
Just run these two commands (no restart needed):
mkdir /var/mysql
ln -s /tmp/mysql.sock /var/mysql/mysql.sock
Step 1
Find the path to your unix_socket, to do that just run netstat -ln | grep mysql
You should get something like this
unix 2 [ ACC ] STREAM LISTENING 17397 /var/run/mysqld/mysqld.sock
Step 2
Take that and add it in your unix_socket param
'mysql' => array(
'driver' => 'mysql',
'host' => '67.25.71.187',
'database' => 'dbname',
'username' => 'username',
'password' => '***',
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
'unix_socket' => '/var/run/mysqld/mysqld.sock' <-----
),
),
Hope it helps !!
I'm running on MAMP Pro and had this similar problem when trying to migrate (create db tables). Tried a few of these mentioned suggestions as well but didn't do it for me.
So, simply (after an hour googling), I added two things to the /config/database.php.
'port' => '1234',
'unix_socket' => '/path/to/my/socket/mysqld.sock'
Works fine now!
Just i do one change in .env file
I have following line of code.
DB_CONNECTION=mysql
DB_HOST=localhost
DB_PORT=3306
DB_DATABASE=database_name
DB_USERNAME=root
DB_PASSWORD=
Change host name localhost to 127.0.0.1
DB_CONNECTION=mysql
DB_HOST=127.0.0.1
DB_PORT=3306
DB_DATABASE=database_name
DB_USERNAME=root
DB_PASSWORD=
That is work in my case because that can't find any hostname like localhost
And after changing hostname write following command
php artisan config:clear
php artisan migrate:install
php artisan migrate
I had this problems when I was running my application using docker containers.
The solution was put the name of the MySQL service container I was using in docker_compose.yml on DB_HOST. In my case, it was db :
DB_HOST=db
Hope it helps.
I ran into this problem when running PHPUnit in Elixir/Gulp, and Homestead as my Vagrant enviroment.
In my case I edited the .env file from DB_HOST=localhost to DB_HOST=192.168.10.10 where 192.168.10.10 is the IP of my Vagrant/Homestead host.
Check your port carefully . In my case it was 8889 and i am using 8888.
change "DB_HOST" from "localhost" to "127.0.0.1" and vice versa
I had similar problems accessing my Drupal website. I fixed it by opening the command line, and restarting my MySQL server or service:
service mysqld restart
This should work. If it doesn't, restart your local webserver:
service httpd restart
That should be enough. Hope it works for other environments, too. Note that these commands generally require superuser privileges.
I had the same problem using Docker and MySQL service name db in docker_compose.yml file:
I added the following in the .env file:
DB_HOST=db
you should also assure that your host is discoverable from the php app.
It was because PHP didn't figure out which host to use to connect.
I got the same problem in ubuntu 18.04 with nginx. By following the below steps my issue has been fixd:
First open terminal and enter into mysql CLI. To check mysql socket location I write the following command.
mysql> show variables like '%sock%'
I got something like the below :
+-----------------------------------------+-----------------------------+
| Variable_name | Value |
+-----------------------------------------+-----------------------------+
| mysqlx_socket | /var/run/mysqld/mysqlx.sock |
| performance_schema_max_socket_classes | 10 |
| performance_schema_max_socket_instances | -1 |
| socket | /var/run/mysqld/mysqld.sock |
+-----------------------------------------+-----------------------------+
4 rows in set (0.00 sec)
In laravel project folder, look for the database.php file in the config folder. In the mysql section I modified unix_socket according to the above table.
'mysql' => array(
'driver' => 'mysql',
'host' => '127.0.0.1',
'database' => 'database_name',
'username' => 'username',
'password' => 'password',
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
'unix_socket' => '/var/run/mysqld/mysqld.sock',
),
Now just save changes, and reload the page and it worked.
As of Laravel 5 the database username and password goes in the .env file that exists in the project directory, e.g.
DB_HOST=127.0.0.1
DB_DATABASE=db1
DB_USERNAME=user1
DB_PASSWORD=pass1
As you can see these environment variables are overriding the 'forge' strings here so changing them has no effect:
'mysql' => [
'driver' => 'mysql',
'host' => env('DB_HOST', 'localhost'),
'database' => env('DB_DATABASE', 'forge'),
'username' => env('DB_USERNAME', 'forge'),
'password' => env('DB_PASSWORD', ''),
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
'strict' => false,
],
More information is here https://mattstauffer.co/blog/laravel-5.0-environment-detection-and-environment-variables
solved
in my case it was a logic issue in code, the connection values are in a if statement:
if($_SERVER['HTTP_HOST'] == "localhost")
so the solution was to add a pipe and add 127.0.0.1, that solved the problem for me
if($_SERVER['HTTP_HOST'] == "localhost" || $_SERVER['HTTP_HOST'] == "127.0.0.1")
If you are using Laravel Homestead,
here is settings
(include Vagrant-Virtual Machine)
.bash-profile
alias vm="ssh vagrant#127.0.0.1 -p 2222"
database.php
'mysql' => [
'driver' => 'mysql',
'host' => env('DB_HOST', '127.0.0.1'),
'database' => env('DB_DATABASE', 'homestead'),
'username' => env('DB_USERNAME', 'homestead'),
'password' => env('DB_PASSWORD', 'secret'),
'charset' => 'utf8',
'collation' => 'utf8_unicode_ci',
'prefix' => '',
'strict' => false,
],
Terminal
vm
vagrant#homestead:~/Code/projectFolder php artisan migrate:install
Attempt to connect to localhost:
SQLSTATE[HY000] [2002] No such file or directory
Attempt to connect to 127.0.0.1:
SQLSTATE[HY000] [2002] Connection refused
OK, just comment / remove the following setting from my.cnf (on OS X 10.5: /opt/local/etc/mysqlxx/my.cnf) to obtain:
[mysqld]
# skip-networking
Of course, stop and start MySQL Server.
If anyone are still looking for the answer, just check your .env file. For some reason laravel create a .env.example file, so all this answers didn't work for me. I fixed my issue renamming .env.example to .env
This happened to me because MySQL wasn't running. MySQL was failing to start because I had a missing /usr/local/etc/my.cnf.d/ directory.
This was being required by my /usr/local/etc/my.cnf config file as a glob include (include /usr/local/etc/my.cnf.d/*.cnf).
Running mkdir /usr/local/etc/my.cnf.d, and then starting MySQL, fixed the issue.
In my case, I was running php artisan migrate on my mac terminal, when I needed to ssh into vagrant and run it from there. Hope that helps someone the headache.
In my case I had to remove the bootstrap/cache folder and try it again.
My cenario was after a server migration.
When using a VirtualMachine make sure you ssh into that machine and navigate to your App folder and call the php artisan migrate command from there.
I can´t send email on digitalOcean droplet with ubuntu, php7 and cakephp3
In my localhost it works but in production I get timeout error.
I've done everything and many ways and no works.
see my app.config
'EmailTransport' => [
'default' => [
'className' => 'Smtp',
'host' => 'smtp.gmail.com',
'port' => 587,
'timeout' => 30,
'username' =>'MYuser',
'password' => 'MYpwd',
'client' => null,
'tls' => true,
'url' => env('EMAIL_TRANSPORT_DEFAULT_URL', null),
],
],
Can someone help me?
thanks
I found the answer.
DigitalOcean blocks for 60 days new Droplets to send email for "security reasons".
If you need a server for host a site whos send email, don't choice DigitalOcean.
Apparently, Digital Ocean blocks outgoing SMTP ports by default. You'll have to ask them to unblock them, and they may wait a while before doing so.
https://www.digitalocean.com/community/questions/not-able-to-connect-to-gmail-smtp
Look like you have to enable port 25. According to their support, they mentioned like this;
sudo ufw allow 25
sudo ufw reload
Or you may not have enough money for support SLA.