I had an “error connecting to database” when accessing one of my wordpress website hosted on Digital Ocean.
ps ax | grep mysqld
mysqld isn’t a running process. Clearly mysql was stopped for some reason.
cat /var/log/mysql/error.log
Checking the error log, it shows that there is a problem starting InnoDB. Seems like insufficient memory.
140610 2:04:05 [Note] /usr/sbin/mysqld: Shutdown complete
140610 2:04:06 InnoDB: Initializing buffer pool, size = 128.0M
InnoDB: mmap(137363456 bytes) failed; errno 12
140610 2:04:06 InnoDB: Completed initialization of buffer pool
140610 2:04:06 InnoDB: Fatal error: cannot allocate memory for the buffer pool
Hence i check my available memory:
free -m
Indeed, I have not much free memory to start mysql..
total used free shared buffers cached
Mem: 490 456 34 0 1 76
-/+ buffers/cache: 378 111
Swap: 0 0 0
To recover, I restart apache2 to free up some memory
sudo service apache2 restart
And with that, my free memory increased from 34MB to 153MB, which is enough to restart mysql.
sudo service mysql restart
Question now is:
How to prevent this error again (when memory runs low)?
Swap Memory
The tiny droplet I have has 512MB RAM, and 20GB on SSD harddisk.
It’s a waste not to create swap memory out of the 20GB SSD!
So, add a few GB of swap memory to your Ubuntu.
Steps:
# Confirm you have no swap
sudo swapon -s
# Allocate 1GB (or more if you wish) in /swapfile
sudo fallocate -l 1G /swapfile
# Make it secure
sudo chmod 600 /swapfile
ls -lh /swapfile
# Activate it
sudo mkswap /swapfile
sudo swapon /swapfile
# Confirm again there's indeed more memory now
free -m
sudo swapon -s
# Configure fstab to use swap when instance restart
sudo nano /etc/fstab
# Add this line to /etc/fstab, save and exit
/swapfile none swap sw 0 0
# Change swappiness to 10, so that swap is used only when 10% RAM is unused
# The default is too high at 60
echo 10 | sudo tee /proc/sys/vm/swappiness
echo vm.swappiness = 10 | sudo tee -a /etc/sysctl.conf
That’s it.
Without spending a cent, memory increased!