I run a hip-hop music discussion forum. We get 8-10,000 unique visitors a day. A few days ago, we ditched Cpanel/EasyApache and did a fresh install on our CentOS Xen dedicated VPS using eva2000's modified Centmin NMP server install script. Everything appears fine, but I can't can't get the site back up due to problems with MySQL. We're getting these errors: The SQL server is MariaDB 5.2.6, done with the secure installation script. As of yesterday I was able to login to MySQL, however today this happens: Centmin suggested this: However I'm unable to do that right now since I can't login to MySQL (the server IS confirmed to be running, however). I've already contacted Centmin, my hosting provider, vbulletin.com tech support and another tech support provider but none of them have been able to help me so far. My site has already been down for several days, and staff/members are starting to get really frustrated. I don't know where to turn. At this point I need to get these problems fixed ASAP, and am more than willing to pay any willing freelancer their hourly rate, anything to get my forums working again. If anyone has any suggestions, or is willing to offer their services or recommend someone, please let me know.
It was up and running in the beginning, however since it has appeared to disappear. I've uninstalled/reinstalled our RPMs several times and tried editing my.cnf, to no avail. MySQL was also returning errors about being unable to find the pid file, and it wasn't generating the socket file like it's supposed to. My first inclination is that it's a permissions problem, but I've been unable to fix it. It's a fresh install on the exact hardware the Centmin script was written for. In any case it's been 4 days now, I'm out of time. Money is not an issue at this point, getting my site online is my primary concern. I'd rather have this dealt with quickly by a professional, or anyone with a greater level of experience than me for that matter. The alternative is a bleak and soul-destroying prospect: reinstalling the OS again, enduring further down time, and returning to the bloat of WHM/Cpanel.
Try starting mysqld manually and see if it throws an error... or check the error log in your mysql folder.
Shawn, Thanks so much for the speedy replies. I just wanted to update this to say I contacted my hosting provider and they were able to resolve the issue, so everything is working smoothly now. Cheers
if this happens again just use the command #killall mysql. Im pretty sure that is what your host did. Go ahead and run a db optimize after you get that error just in case.
Thanks for the info, all. Smooth sailing continues so far. Also, I think it's worth noting that since switching from EasyApache to nginx, our webpagetest score has gone from 67 to 84, and our average load time has dropped from nearly 7 seconds down to 4.25. All this, without a single custom performance tweak.