Linux - ServerThis forum is for the discussion of Linux Software used in a server related context.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I have a dedicated server but am having some serious backend issues. It is a discussion forum and gallery mostly that run on the server with a 16gb mysql database. The load never goes above 5 but the server is dog slow. I request a ram upgrade to 16gb and a SSD drive added as a secondary drive for the mysql tables. Attached is my top and iotop out puts, it wasnt that busy when I took them though. Could I be missing something or is it all mysql doing this?
Try to run iostat -k 2 20 and see the i/o distribution accross disks.
Does the networking involved in the processing as well? It can slow down the performance as well...
Also what is the speed of your drives?
I can't spell mysql, but those numbers look seriously out of whack.
How can a system only using 168k of swap have tasks waiting on swapin ?. iotop uses the numbers from taskstats, so I'd presume they are correct, and kswapd is certainly being kicked into action. All bad signs I'd say.
Certainly looks like disk(s) are getting thrashed from those iowait numbers. Database and swap on the same disk(s) by any chance ?.
Which leaves mysql - maybe fixing buffers in RAM so it looks like normal "cached", but isn't as noone else can use it. I'm sure I've read something on Oracle (SGA ?) causing similar symptoms.
Try to run iostat -k 2 20 and see the i/o distribution accross disks.
Does the networking involved in the processing as well? It can slow down the performance as well...
Also what is the speed of your drives?
Here is what I got, I am doing a SQL operation that is quite heavy to show the load:
you're never going to tell what's wrong by looking at i/o rates as the real secret is in looking at what's happening at lower levels such as the disk service and wait times. One thing that's really easy to get fooled by is a disk that's not very busy. If it's doing random I/O it doesn't take a whole lot to saturate it. iostat -x will show that level of detail but quite honestly I can't handle reading the display - too much going on.
try collectl -sD --home and you'll see something much easier to read. if you have continual wait times over more than 100 msec or so, it could be that your disk(s) are flat out and nothing is going to help short of getting more disk or finding some other way to reduce the load on them.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.