I need a specific version of MySQL (5.7) to be installed on my MacBook with M1.
I'm trying to do that with Homebrew.
brew install mysql#5.7
The output:
We've installed your MySQL database without a root password. To secure it run:
mysql_secure_installation
MySQL is configured to only allow connections from localhost by default
To connect run:
mysql -uroot
mysql#5.7 is keg-only, which means it was not symlinked into /opt/homebrew, because this is an alternate version of another formula.
If you need to have mysql#5.7 first in your PATH, run: echo 'export PATH="/opt/homebrew/opt/mysql#5.7/bin:$PATH"' >> ~/.zshrc
For compilers to find mysql#5.7 you may need to set: export LDFLAGS="-L/opt/homebrew/opt/mysql#5.7/lib" export CPPFLAGS="-I/opt/homebrew/opt/mysql#5.7/include"
To have launchd start mysql#5.7 now and restart at login: brew services start mysql#5.7 Or, if you don't want/need a background service you can just run: /opt/homebrew/opt/mysql#5.7/bin/mysql.server start
Right after that, I try to run:
echo 'export PATH="/opt/homebrew/opt/mysql#5.7/bin:$PATH"' >> ~/.zshrc
source .zshrc
mysql_secure_installation
And get the error:
Securing the MySQL server deployment.
Enter password for user root:
Error: Can't connect to local MySQL server through socket '/tmp/mysql.sock' (2)
Also, I've tried:
brew services start mysql#5.7
And also get the error:
Error: Permission denied # rb_sysopen - /Users/vivanc/Library/LaunchAgents/homebrew.mxcl.mysql#5.7.plist
Seems like there is a permissions-related problem.
Any advice is appreciated.
If you are running into issues like "Can't connect to local MySQL server through socket '/tmp/mysql.sock'" or "The post-install step did not complete successfully MySQL", and you installed a previous version of mysql (e.g. 8.x) previously, you may have been at the point where you need to clean everything before reinstalling your preferred version of mysql#x.x.
If you've already visited these to links:
Uninstall all those broken versions of MySQL and re-install it with Brew on Mac Mavericks (Coderwall) + Install MySQL 5.7 on macOS using Homebrew (github) and your're still having trouble with starting your mysql-service, you should try to also remove also /opt/homebrew/etc/my.cnf file.
Remove it together with all the related files too! Summary for an M1 Apple Silcion machine, after uninstalling via brew uninstall mysql or brew uninstall mysql#x.x, please remove:
/opt/homebrew/var/mysql
/opt/homebrew/etc/my.cnf
After this, everything worked like a fresh install (for me). Hope this saved someone's time.
While this question is specific to Apple Silicon and mysql 5.7, and I will address that in this answer, I would like to start by adding some general notes to save others time:
Homebrew supports mysql 5.7 for both intel and apple silicon.
https://formulae.brew.sh/formula/mysql#5.7
Homebrew supports mysql 5.6 for the intel chip, but not apple silicon:
https://formulae.brew.sh/formula/mysql#5.6
If you need 5.6 on apple silicon, it's probably a good idea to just settle with 5.7 since production will need to be upgraded eventually and the differences aren't that big.
In my situation I originally did brew install mysql and it gave me the latest mysql (currently 8.0). When I tried going back over it and doing brew install mysql#5.6, of course this didn't work due to not being supported on the m1 max (apple silicon). I ran into issues then trying to get brew install mysql#5.7 working. I followed some guides mentioning various suggestions. After playing with it for a while, it seems that running two versions of mysql at once will corrupt your mysql files and make it hard to work with and confuse homebrew.
I came across these guides:
brew install mysql on macOS
https://coderwall.com/p/os6woq/uninstall-all-those-broken-versions-of-mysql-and-re-install-it-with-brew-on-mac-mavericks
https://www.codegrepper.com/code-examples/shell/brew+uninstall+mysql
Before reading further, note that in my situation, I had a zip available for my whole local mysql database needs and could risk destroying what I have.
Warning! Reading further and executing these commands carelessly may delete your mysql storage. Make sure you have a backup.
brew remove mysql is a good command that got rid of my latest install.
I also needed brew remove mysql#5.7 even though this is the version I'm trying to install, but they were stepping on each other's foot...
brew cleanup is nice, I noticed it freed up some lock files.
I don't recall having any luck with:
launchctl unload -w ~/Library/LaunchAgents/homebrew.mxcl.mysql.plist
but you can try it.
rm ~/Library/LaunchAgents/homebrew.mxcl.mysql.plist has worked for me.
But note there will also be another file in there for your 5.7 setup.
I'd recommend:
cd ~/Library/LaunchAgents/ then look around to see if you have any mysql* in there. Then if you do, such as homebrew.mxcl.mysql#5.7.plist, then remove it.
These are some that worked for me:
sudo rm -rf /usr/local/var/mysql
rm /usr/local/etc/my.cnf
rm /usr/local/etc/my.cnf.default
Also:
cd opt/homebrew/var
Then if you notice a mysql directory in there, remove it.
It's possible you might have docker or something else similar also running mysql or mysqld.
ps -ax | grep "[m]ysql"
(Note the brackets and quotes in the above is to prevent the grep from showing up in the process list and matching itself unlike the lazier version: ps -ax | grep mysql which will give the impression there's an extra mysql process running)
Once you're ready, please run:
brew services list
double-check you don't have mysql in there.
Then double check you don't have any mysql process running:
ps -ax | grep "[m]ysql"
This can happen for example if you enter mysqld for example....
You might need to do brew services stop mysql or brew services stop mysql#5.7 or similar and repeat steps if you see anything on the service list or process list.
Finally, you should be ready for a fresh install.
The fresh installation process should be rather straight forward:
brew install mysql#5.7
brew link --force mysql#5.7
brew services start mysql#5.7
Then run brew services list to make sure your installation worked correctly.
If you see it's green and "started", your installation was successful!
You should also run:
mysql_secure_installation
Then choose a password for root and go through the list of questions like validation, etc.
Once you're installed, you might also run into mysql mode issues.
cd /opt/homebrew/etc and you should find a my.cnf file.
Edit it using either vim or nano.
You should see something like:
# Default Homebrew MySQL server config
[mysqld]
# Only allow connections from localhost
bind-address = 127.0.0.1
Edit it and set the mysql mode. For example, I don't want the no_zero_in_date mode, so I use:
# Default Homebrew MySQL server config
[mysqld]
# Only allow connections from localhost
bind-address = 127.0.0.1
sql-mode="ONLY_FULL_GROUP_BY,STRICT_TRANS_TABLES,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION"
Then restart mysql like so:
brew services restart mysql#5.7
Then double check you still have a green status "started":
brew services list
If you do, then you should have the no_zero_in_date mode disabled.
You can test it with a query:
SELECT ##sql_mode;
You should now have mysql 5.7 running on Apple Silicon and have the sql_mode set.
If you run into more troubleshooting, do your best to get into a known state such as uninstalled, doing a fresh install, or already installed with some version, etc.
I solved it after hours of searching, it was because I had not cleaned everything regarding my previous MySQL 8.X installation. I followed this and got suspicious when I realized I don't have most of the folders there. I then did a search for all folders named mysql on my computer and found a mysql folder in opt/homebrew/var. After removing it, reinstalling MySQL 5.7, and starting the server everything is working as expected.
When setting up my Apple M1 Macbook Pro I was experiencing the issue of the following error message being thrown:
Can't connect to local MySQL server through socket '/tmp/mysql.sock' (2)
My fix was to modify the MySQL config file:
nano /usr/local/etc/my.cnf
Append the following lines:
tmpdir=/tmp
user=root
Then run:
brew services restart mysql#5.7
This fixed my issue, and I was then able to connect successfully with my DB client.
Seems like I found the solution (or workaround).
After installing the MySQL 5.7 with Homebrew just run:
mysql.server start
MySQL is started and then you are able to run mysql_secure_installation and mysql commands.
I got this warning:
You should change the ownership of these directories to your user.
sudo chown -R $(whoami) /usr/local/share/doc /usr/local/share/man /usr/local/share/man/man1
And make sure that your user has write permission.
chmod u+w /usr/local/share/doc /usr/local/share/man /usr/local/share/man/man1
I run those two command then it works.
I have just installed Debian Lenny with Apache, MySQL, and PHP and I am receiving a PDOException could not find driver.
This is the specific line of code it is referring to:
$dbh = new PDO('mysql:host=' . DB_HOST . ';dbname=' . DB_NAME, DB_USER, DB_PASS)
DB_HOST, DB_NAME, DB_USER, and DB_PASS are constants that I have defined. It works fine on the production server (and on my previous Ubuntu Server setup).
Is this something to do with my PHP installation?
Searching the internet has not helped, all I get is experts-exchange and examples, but no solutions.
You need to have a module called pdo_mysql. Looking for following in phpinfo(),
pdo_mysql
PDO Driver for MySQL, client library version => 5.1.44
The dsn in your code reveals you are trying to connect with the mysql driver. Your error message indicates that this driver is unavailable.
Check that you have the mysql extension installed on your server.
In Ubuntu/Debian you check for the package with:
dpkg --get-selections | grep php | grep mysql
Install the php5-mysql package if you do not have it.
In Ubuntu/Debian you can use:
PHP5: sudo apt-get install php5-mysql
PHP7: sudo apt-get install php7.0-mysql
Lastly, to get it working, you will need to restart your web-server:
Apache: sudo /etc/init.d/apache2 restart
Nginx: sudo /etc/init.d/nginx restart
Update: newer versions should use php-sqlite3 package instead of php5-sqlite. So use this, if you are using a recent ubuntu version:
sudo apt-get install sqlite php-sqlite3
Original answer to question is here:
sudo apt-get install sqlite php5-sqlite
sudo /etc/init.d/apache2 restart
If your phpinfo() is not showing the pdo_sqlite line (in my case, on my Ubuntu Server), you just need to run the lines above and then you'll be good to go.
For newer versions of Ubuntu that have PHP 7.0 you can get the php-mysql package:
sudo apt-get install php-mysql
Then restart your server:
sudo service apache2 restart
I had the same issue. The solution depends on OS. In my case, i have debian, so to solve it:
Updated my php version from (php5 to php7)
Install php-mysql and php7.0-mysql
apt-get install php-mysql
apt-get install php7.0-mysql
I edited my php.ini locate at /etc/php/7.0/apache2/php.ini
uncomment the line : extension=php_pdo_mysql.dll
Then restart apache:
service apache2 restart
This solves my problem
On my Windows machine, I had to give the absolute path to the extension dir in my php.ini:
extension_dir = "c:\php5\ext"
Check if the module is available with php -m | grep pdo_mysql.
If not, for PHP 7.2, you can install relevant package with sudo apt install php7.2-mysql.
Use similar command on other PHP versions and package managers.
On Ubuntu just execute
sudo apt-get install php5-mysql
sudo apt-get install php-mysql
worked well on ubuntu and php 7
When adding these into your php.ini ensure the php_pdo.dll reference is first before the db drivers dlls otherwise this will also cause this error message too. Add them like this:
[PHP_PDO]
extension=php_pdo.dll
[PHP_PDO_MYSQL]
extension=php_pdo_mysql.dll
for Windows 8.1/10 in :\\php.ini file you should uncomment line "extension=pdo_mysql"
Did you check your php.ini (check for the correct location with phpinfo()) if MySQL and the driver is installed correctly?
For PHP 5.5 on CentOS I fixed this by installing the php55-mysqlnd package.
sudo yum -y install php55w-mysqlnd # For Webtatic
sudo yum -y install php55u-mysqlnd # For Remi
For help installing, write a comment as it depends on the way PHP is installed on your system. Available repo's are webtatic and remi.
Check if extension_dir in php configuration file set correctly. Try to comment/uncomment some extensions and see if it's reflected on phpinfo().
If it doesn't then either php config file cannot be loaded (wrong location) extension_dir is commented or set to the wrong location.
In my case my DSN string was incorrect, specifically it did not contain mysql://. I would have expected a different error message, perhaps something like 'DSN string does not specify driver/protocol.'
Adding mysql:// to the beginning of the DSN string resolved the issue.
I had the same problem during running tests with separate php.ini. I had to add these lines to my own php.ini file:
[PHP]
extension = mysqlnd.so
extension = pdo.so
extension = pdo_mysql.so
Notice: Exactly in this order
I spent the last day trying to figure out why I was getting the following error. I am running Ubuntu 14.04.
The Problem:
I noticed that my PHP-CLI version was running php7.0 but php_info() (the web version) was displaying php 5.5.9. Even though php_info() said pdo was enabled, using the command line (CLI) wasn't recognizing the pdo_mysql command. It turns out that mysql was enabled for my old version but not the CLI version. All I did was install mysql for php7.0 and it was able to work.
This is what worked:
To check the version:
php -v
To install mysql for php7.0
sudo apt-get install php7.0-mysql
1) make sure your CLI version is the same as your web version
2) If they are different, make sure your CLI version has the mysql plug-in since it doesn't come with it as a default.
Incorrect installation of PHP was being called
I was experiencing the same problem. And I hope this would help someone who is having a similar issue as me.
Scenario
OS = Windows 10
Platform = XAMPP
PHP Version = 7 (Multiple Version seem to have been installed in the PC)
I created phpinfo.php file in the public folder and run the phpinfo() to look for the location of my php.ini file.
PHP.ini Location = c:\xampp\php\php.ini
Problem
Calling c:\xampp\htdocs> php -v returned PHP 7.2.3 but phpinfo.php showed PHP 7.2.2.
Solution
Instead of calling
php artisan migrate:install
which gave me this error, I used
c:\xampp\php\php artisan migrate:install
and it worked.
The problem is a missing php to mysql library. In CentOs i fixed it by running
# yum install php-mysql and then restarting apache with # /bin/systemctl restart httpd.service Note that the naming is slightly different from debian/ubuntu based distros, php->php5 and httpd->apache2.
I extremely recommend mysqllnd instead of mysql because of you would have a lot of problems like number converting and bit type evaluates problem with mysql extension.
on ubuntu install mysqllnd with following command:
sudo apt-get install php5-mysqlnd
In my case, I was using PDO with php-cli, and it worked fine.
Only when I tried to connect from apache, I got the "missing driver" issue, which I didn't quite understand.
A simple apt-get install php-mysql solved it. (Ubuntu 16.04 / PHP7. credits go to the selected answer & Ivan's comment)
Hope it can help.
PHP Fatal error: Uncaught PDOException: could not find driver
I struggled and struggled with "apt install php-mysql php7toInfinity and don't forget sqlite-what-ever's" and just could not get rid of this error message until I went back to basics and reset the file-permissions on the web-site in question.
These 3 commands reset file and folder permissions on the web-site and got it to work again.
cd /var/www/web-site-name.com/web/
# find (sub) directories and change permissions
find . -type d -exec chmod 755 {} \;
# find files and change permissions
find . -type f -exec chmod 664 '{}' \;
I Fixed this issue on my Debian 6.
Normally I just had installed php5-common package. After installation, you have to restart your web server (apache or nginx depending on which one you installed).
Then I just do an lsof on the apache process id (lsof -p process_id) as followed :
sudo lsof -p 1399 #replace 1399 by your apache process id
apache2 1399 root mem REG 254,2 80352 227236 /usr/lib/php5/20090626/xmlrpc.so
apache2 1399 root mem REG 254,2 166496 227235 /usr/lib/php5/20090626/suhosin.so
apache2 1399 root mem REG 254,2 31120 227233 /usr/lib/php5/20090626/pdo_mysql.so
apache2 1399 root mem REG 254,2 100776 227216 /usr/lib/php5/20090626/pdo.so
apache2 1399 root mem REG 254,2 135864 227232 /usr/lib/php5/20090626/mysqli.so
As you can see above, the modules are installed on a file path not known or guided by common library path: /usr/lib/php5/20090626/. For your installation, it may be different, but only the path of pdo_mysql.so, pdo.so, mysqli.so. So, this is why Drupal or any other php engine couldn't find the library and shows that error: PDOException: could not find driver
I just don't know why it is installed on such a weird path, for me it's just a bug in the library package installation script in debian 6.
I solved the issue by creating a symbolic for all the files under /usr/lib/php5/20090626/ to
/usr/lib/php5/ with this command :
ln -s /usr/lib/php5/20090626/* /usr/lib/php5/
$DB_TYPE = 'mysql'; //Type of database<br>
$DB_HOST = 'localhost'; //Host name<br>
$DB_USER = 'root'; //Host Username<br>
$DB_PASS = ''; //Host Password<br>
$DB_NAME = 'database_name'; //Database name<br><br>
$dbh = new PDO("$DB_TYPE:host=$DB_HOST; dbname=$DB_NAME;", $DB_USER, $DB_PASS); // PDO Connection
This worked for me.
I faced the same issue after I removed the php5 package (that includes all the drivers as well) in order to install php7 package. I actually installed php7 package without a mysql module.
I managed to solve it by typing in the terminal:
1) $ apt-cache search php7
which lists all the modules, looking through the modules I found,
php7.0-mysql - MySQL module for PHP
2) $ sudo apt-get install php7.0-mysql
That's it. It worked for me in my linux system.
(use the appropriate php version, yours could be php5)
Just one other thing to confirm as some people are copy/pasting example code from the Internet to get started. Make sure you have MySQL entered here:
... $dbh = new PDO ("mysql: ...
In some examples this shows
$dbh = new PDO ("dblib ...
For those using Symfony2/3 and wondering why you're getting this error. If you're using "mapping_types", you might encounter this error. The reason is that "mapping_types" is placed at the wrong level. For instance :
doctrine:
dbal:
mapping_types:
set: string
This "mapping_types" must be placed at this level :
doctrine:
dbal:
#To counter the error caused by 'mapping_types'
connections:
default:
server_version: %database_server_version%
mapping_types:
set: string
I hope this helps
I found the solution here : https://github.com/doctrine/DoctrineBundle/issues/327
Everywhere I go I read that the path of extension_dir should be changed from ext to an absolute path. It worked for me. However, when trying to build a server of my colleague's PC, I had to let the value to ext instead of putting an absolute path.
If you did put an absolute path and it does the extension is still not found, considerer trying both with the absolute path and ext.
Check correct path in extension_dir in you phpinfo().
Had the same issue, because I forgot to go into my virtual machine. If I go to my local directory like this:
cd /www/homestead/my_project
php artisan migrate
that error will appear. But it works on my virtual machine
cd ~/homestead
vagrant ssh
cd /www/homestead/my_project
php artisan migrate
I am trying to install mysql server in my AWS EC2 instance. I googled and executed the below command in the shell.
sudo yum install mysql-server
It throws the below connection timed out exception
Loaded plugins: priorities, update-motd, upgrade-helper
http://packages.us-west- 2.amazonaws.com/2014.03/main/201403504e47/x86_64/repodata/repomd.xml?instance_id=i-c82c93c3®ion=us-west-2: [Errno 12] Timeout on http://packages.us-west-2.amazonaws.com/2014.03/main/201403504e47/x86_64/repodata/repomd.xml?instance_id=i-c82c93c3®ion=us-west-2: (28, 'Connection timed out after 10001 milliseconds')
Trying other mirror.
Is this the right way to install mysql server?
Thanks.
installing mysql
From the current AWS EC2 documentation:
To install and start the LAMP web server
Connect to your instance.
To ensure that all of your software packages are up to date, perform a quick software update on your instance. This process may take a few minutes, but it is important to make sure you have the latest security updates and bug fixes.
Note
The -y option installs the updates without asking for confirmation. If you would like to examine the updates before installing, you can omit this option.
[ec2-user ~]$ sudo yum update -y
Now that your instance is current, you can install the Apache web server, MySQL, and PHP software packages. Use the yum groupinstall command to install multiple software packages and all related dependencies at the same time.
[ec2-user ~]$ sudo yum groupinstall -y "Web Server" "MySQL Database" "PHP Support"
Note
Non-Amazon Linux instances may have subtle differences in their group names. If the above command fails because of an invalid group name, use the yum grouplist command and scan the output for similar groups, such as "MySQL Database server" instead of "MySQL Database", and use the appropriate group name for your distribution.
Install the php-mysql package.
[ec2-user ~]$ sudo yum install -y php-mysql
I suppose you only need to install mysql, so you just need to run the 1st command, adapt the second to only grab mysql from the repositories, and skip the last one:
[ec2-user ~]$ sudo yum update -y
[ec2-user ~]$ sudo yum groupinstall -y "MySQL Database"
Depending on which version of linux you have installed (Amazon, Red Hat or CentOS), you may need to change the string "MySQL Database" to something else. The following command:
[ec2-user ~]$ sudo yum grouplist
Will list all the possible group install you may perform, so you need to filter it to get the right string:
[ec2-user ~]$ sudo yum grouplist | grep -i mysql
And choose the most appropriate one in that filtered result set.
network connectivity
Now, regarding your specific issue, the FAQs describes your problem as follow:
If the AMI cannot access the yum repositories, it will timeout and retry multiple times before completing the boot procedure. Possible reasons for this are restrictive firewall settings or VPC settings, which prevent access to the Amazon Linux AMI package repositories.
If you encounter this issue you can either modify your environment so that the Amazon Linux AMI can connect to its package repositories [...]
Unfortunately, no explanations are given as yo how you can change your environment to enable connectivity.
Apparently, by using the dashboard, you should be able to open outbound access to the port the yum servers are using:
AWS dashboard > ec2 > security groups > default group > outbound rule: ??? (HTTP*) 0.0.0.0/???
The usual port for yum server is 80, so just replace ??? above by 80 to have access to the repos. You also should replace 0.0.0.0 with the actual repository server address.
If this solution does not work, you might also try disabling the firewall altogether for the duration of the update:
[ec2-user ~]$ sudo /etc/init.d/iptables stop
<...Update commands as listed above...>
[ec2-user ~]$ sudo /etc/init.d/iptables start
Or
[ec2-user ~]$ sudo /sbin/service iptables stop
<...>
[ec2-user ~]$ sudo /sbin/service iptables start
I just delete all the files in /etc/resolv.conf dir
I have previously installed, connected and uploaded to a WordPress MySQL database via SSH no problem. The people who run my server made a few configuration changes, and now I get a "bash: mysql: command not found" error when I try to log into MySQL via the command line on the same server to access the same database.
I am relatively new to all of this, so I am really not sure what to do. When I run "which mysql" I get a message that says there is no MySQL executable in the /usr/bin directories. I can verify that MySQL is running between the fact that my site is still live and functioning and when I ran a command to test MySQL, I got a message that said "MySQL works!"
I find MySQL files in several directories, but I'm not really sure what I am looking for and how I connect to it when I do find it. I am also not sure if this is user error, or if somehow someone moved or hid MySQL from me -- likely user error??
A MySQL server is not a MySQL client.
Check if MySQL is running by executing this command:
ps aux | grep mysql | grep -v grep
And install the MySQL client:
sudo apt-get install mysql-client # Or your distribution command
If you have MySQL server up and running on your server, it does not mean you have a MySQL client installed on this server.
Try
ls -l /usr/bin/mysql*
Do you see MySQL binaries exactly present on system?
One more problem: Linux has very strange behaviour when you run binaries for another architecture.
For example, if you run a 32-bit executable on 64-bit system you will get an error message like "command or file not found", even if the binaries are actually present!
Now mysql-client has changed to default-mysql-client.
Hence the command:
sudo apt-get install default-mysql-client
Although MySQL is running, if you are not able to connect to the server using the mysql command, then you might be missing to provide soft links:
sudo ln -s /usr/local/mysql/bin/mysqladmin /usr/bin
sudo ln -s /usr/local/mysql/bin/mysql /usr/bin
This should enable you to connect.
In order to get MySQL CLI working or to access the mysql command from anywhere, there are a set of steps to be done to add it to the $PATH variable.
First, open a terminal, and run the following command
echo 'export PATH=/usr/local/mysql/bin:$PATH' >> ~/.bash_profile
Then to reload the Bash profile, run
. ~/.bash_profile
Now run,
mysql -u root -p
Enter the password which you gave during installation. You should see the following result:
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 8
Server version: 8.0.23 MySQL Community Server - GPL
You sometimes need to install MySQL on the local machine as well. That means, if you have a host running a MySQL Docker container, the strange thing happened to me that I needed to install mysql-server on the host as well, not only in the container.
The reason was that there was a shell script that needed to check the right database name from a query. In your case, it might be something different. On your host machine (which might even be your local computer, depending the your setup), if on Linux, try:
sudo apt-get update && sudo apt-get install mysql-server
Then you will get rid of
/home/.../some_bash_script.sh: line 123: mysql: command not found