Search results

If you're using Nginx-only (no proxying to Apache) then you'll likely want to use either the the single domain or all domain versions of this...

If that's the case, then it's likely an abusive or malicious bot/vulnerability sniffer that causes your memory usage to surge so much that OS invokes the OOM killer to avoid kernel panic so you might want to check access logs if you want to try and track it down. Almalinux 9 is solid and has a...

If the database (or more likely, a table within it) is corrupt or the service crashes during the dump process, that might result in odd behaviors so after checking the aforementioned path, try checking/optimizing the database, restart the database service, and check the database log to see if...

Your VPS doesn't have enough memory and the OOM killer selects the highest memory consuming process (which on your setup is invariably going to be MariaDB) to avoid a kernel panic state. If you can add more memory, that would help as more than likely you have a momentary increase in traffic...

That "InnoDB: Cannot allocate memory for the buffer pool" error usually indicates that there isn't enough free memory so one option is to restart your server with the allocation below as your cached memory might be taking up almost all of the free memory: If that fails, then check to ensure...

Those numbers are far too low - Trying clicking the "Revert All" button in your screenshot and then add these changes to the custom file in /etc/my.cnf.d/custom.cnf file before restarting MariaDB.

It's likely being set somewhere else; try looking in /etc/my.cnf

It sounds like your MariaDB setup is limited by memory - What are the values of these two set to? innodb_buffer_pool_size innodb_log_file_size (generally 25% of your innodb_buffer_pool_size) With 8 GB of total available memory, you might want to start with: This would give MariaDB up to...

You can either allocate more memory to the PHP process or disable PCRE entirely by editing the PHP ini via the Plesk control panel (Tools and Setting > PHP Settings > 8.3 FPM handler > php.ini ) and look for the pcre.jit variable, changing it to the following and setting a semicolon: pcre.jit=0

To follow-up on this thread, here is a quick way to take advantage of the built-in Plesk plesk-modsecurity jail which might handle requests a bit faster than scanning individual fail2ban regex/domain logs for bad bots user agent matches. First, build a file in a path such as...

Just based on that entry it was restarted 25 minutes ago so that indicates a crash/restart or possible OOM killer if you didn't kick off any updates. Have you checked your /var/log/mariadb/ log as well as the /var/log/messages log file for activity around that time? If Plesk and the websites...

What is the output of

This might be a bit more efficient, especially than using Fail2Ban which has to scan quite a few log files at scale (and there is a delay in looking at those). That script looks like it needs exact matches so it would be nice if it supported wildcard bot names.

Indeed, it would be nice if Plesk included this in the GUI MariaDB upgrade workflow.

The first place to start might be to check that the value that you're setting is actually in both places: [mysql] AND in the [mysqldump] sections. Then restart the MySQL or MariaDB service and try again.

Better question is, has your website back-end performance improved and has the MariaDB IO dropped a bit? It will still be present but writing to disk directly is slower than memory

Assuming that these are SSD drives, try setting your swapiness to 1 and this time, restart the server.

Now that you've given MariaDB more memory (disk IO, even with SSDs, can be clogged), by changing swapiness the goal is to lower swap access thrashing and prioritize memory. If you want to experiment without restarts of any kind, try: # sudo sysctl vm.swappiness=1 Hopefully throughout the day...

If it's an SSD array (or NVMe), try 1 first and then evaluate after a period. Personally I'd shoot for a reboot but you could use sysctl as an alternative.

Given that you've now given MariaDB more memory and assuming that you have SSDs or faster, try lowering vm.swappiness to a value such as 1 and the evaluate ensure that after you restart, that the value correctly survives the restart: Between the MariaDB changes above and this change, you...