This can be a long time because it is a problem that I have been fighting for a long time, but I desperately need help and hope that someone there will have ideas. for me. The technicians on my server do not have any other solution than some ideas on the software change.
We have a dedicated machine with about 150 active websites (all wordpress) and e-mail addresses for some of them. Every day, around 4 pm, all the websites start to slow down, the mail is not delivered on time (my tests show that they arrive an hour later) and when SSH arrives on the server and runs under Top, it seems to be almost all A single website locks the server at the same time (even development sites on which there should not be real traffic). We have been dealing with this for a long time and we have tried a number of things that we thought were the culprit.
Each site has the same suite of plugins, Wordfence for security, Backupbuddy for backups. So, in our first set of changes, we found that Wordfence was doing analysis at this time of day. So we went to each website and turned off the WordFence analysis feature (which is not ideal). We thought it helped, but it could have been coincidence that the problem started again.
We then began to take a closer look at the top command and found that every day, all sites were running cron jobs and some of them appeared to be connected to the backup plug-in. So we went to all the websites and disabled our backup plug-in. It seemed to help for a day, but maybe another coincidence. As the problem is still topical, and even worse in recent days.
Now, the guys on my server do not stop saying that these are random processes and that the server is just overloaded, but why every day at the same time? You can see the current processes go from 14 to 50,60,80, etc.
Now, when the top command is active, it does not seem to be a cron job. It simply indicates random tasks such as /index.php and /wp-login.php. The normal things you expect to see, but this only really worsens when this happens, and on sites I know do not get visitors.
The guys on my server suggest going to Cloud Linux, which I am not opposed to, but I wanted to get some tips here first, because it seems like nobody has a clue and just runs in the dark.
Has anyone ever seen anything like this before? My knowledge of all this is limited and self-taught, but any help is greatly appreciated.
The restart of HTTP and SQL solves the problem, but often it restarts after a few seconds.
I will try to give more details about our specific situation.
150 sites (ish), all running wordpress, not all of them up to date (some with custom plugins or obsolete themes and requiring work that the client is not ready to implement to fix the update). Most work under PHP 5.6 (something else that I have to fix, but it takes time, we do not have it).
Drive is 90% full, which I also know is not an idea, but it is expensive to remedy the situation.
The server specifications are:
16 GB of RAM
800 GB ssd
mirror drive for backup
Cpanel / WHM / MySQL
I can provide any other necessary information, I really have to get to the bottom of things because my clients are not satisfied.
Should I experience this type of problem with these server specifications and that many sites? Or is there something else going on?
Is obsolete PHP my problem, or websites, or disk space? I just do not know what would cause more than 100 sites to hit the server at once and slow it down.
Thank you in advance for any help you can provide.