the swapfile that was put in place during install (LEMP build) was showing up on the Dashboard as 100% used. Not a happy position. I noticed that it was very small for the system. I therefore shut down and rebooted with a bigger (4Gb) swapfile.
After reboot the dashboard was showing a much better swapfile use (7-8%).
However all sites hosted have gone off line !
I have checked that all the files are still intact.
It has taken all afternoon but I may have fixed it (for now?) the nginx server had stopped and not restarted for some reason.
Swap being 100% full is not necessarily a sign of problems (as long as there is some available memory of some kind), but there is no harm in increasing swap, if you want to and have the disk space to spare.
I don’t know what swap has to do with the other problems you’ve mentioned, though.
Joe,
thanks for the reassurance. I didn’t think it was the action of killing the old swapfile and recreating it bigger. (memory not a problem here) I have done it before on other servers.
What I was worried about was why the nginx server had stopped (or possibly not automatically restarted) following the reboot of the box following swap on. It remains a mystery as nothing I did should have affected nginx.