#StackBounty: #windows #memory #apache-http-server #godaddy Why is RAM usage increasing continuously in GoDaddy VPS Win 2016

Bounty: 100

I have a .NET website hosted with GoDaddy VPS.

It’s a Windows 2016 server and it is also running a SQL express instance and I’m using Plesk.
It also has mysql installed, which I think is used by Plesk.

Initially I chose a 2GB RAM server config, but it crashed twice and I noticed the RAM would start at about 1.5GB usage and continue to go up to 2GB and then the server would sort of crash.

I say sort of because I could still access MS-SQL, still RDP, but my website would not work and the Plesk dashboard would be down.

So I just upgraded to the 32GB of RAM server because the website is live and my client is paying me.

But since I upgraded a few weeks ago, the RAM is still growing slowly over time. It is now on about 40% usage, which is about 13GB of RAM!!

I am running 2 basic IIS websites. In Task Manager, there are 2 IIS processes, each using 175MB of RAM.
I also see SQL and DrWeb using about 450MB each. There is a DNS service and a non-sucking service manager using another 350MB of RAM. Besides that, there is nothing using more than 10-40MB of RAM, and if I add up all the processes, I get to about 2-3GB.

On the performance tab of task manager, it shows 7.8GB of cached committed memory. I don’t know what that means, but I think that’s where the RAM is going.

I am not a server admin but I do know the basics and I’ve checked many things.

I have other websites using GoDaddy’s shared hosting and the websites are very similar (clones in fact), and none of them have issues like this, so I don’t think it’s my website.

I personally think it is related to GoDaddy’s virtualisation software or whatever. The ugprade from a 2GB server to a 32GB server took 3 hours. That to me is a clear sign of something weird going on.

Whatever the cause, how I can go about to find what is causing the RAM increase over time??

I’ve read it’s normal, but when it gets that high, my website crashes, so that can’t be normal.

I’ve ran perfmon logs to save working memory, but I have no idea how to analyze that info as it is hundreds of lines and many different processes.

Some strange activity on the server I do see is that it runs various python/powershell/conhost scripts. They appear to use about 40MB of ram, then die, and then a new one starts. But they remain as greyed out in task manager, so i’m not sure if that is something to look into or not.

I just want to know what is going on, but i don’t know enough about VM’s and servers to know how to diagnose this problem.

I would appreciate any help. I’m willing to pay for help but I also don’t know where to find this sort of service, if someone knows of not-too-expensive companies that can help with this that would be great too.

GoDaddy’s support is utterly terrible. I’ve already tried their paid diagnosis and I honestly believe they just sit there for 1 hour and say it’s my software and I need to contact my server admin.

UPDATE 1
I read that the page file size should be between 1.5 and 4 times the RAM.
I plan on trying this soon.
I have to wait for my client to approve because the server is live and the client depends on the server. Will update once that’s done.

UPDATE 2
Changing the page file size has done nothing for me.
The RAM is still going up at about 1% every 9 hours.
Very frustrating.


Get this bounty!!!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.