Nop 3.1 High Memory and CPU usage. Arvixe claiming website code.

nop 3.1 High Memory and CPU usage

This is a new issue that started about 2 months ago yet prior to this we never experienced this problem with 3.1 and we have been using 3.1 for about a year I believe. The CPU and memory will max out for 15 to 30 minutes and then slowly dissipate back to normal levels. When the CPU and memory spike the website is down. The website will run fine for several weeks and then out of the blue this issue pops up.

The problem is w3wp.exe *32 IIS worker process for the website in question which spikes to 95% to 100% CPU usage and from around .500GB ram to 1.9GB ram.

During this time website traffic is normal. We run RDPGaurd on the server to prevent brute force attacks.

Have called Arvixe many times on this and each time they cannot find an issue with the server and they say there is something wrong with the code in the application because the problem is originating from the w3wp.exe *32 IIS worker process for the website.

Is anyone else experiencing this issue? Is anyone else experiencing this issue with Arvixe or another host? I'm just trying to determine if there is an issue with the website code or if Arvixe is experiencing issues and just won't admit to it?

Is there anything else other than Arvixe issues, the website code or excessive traffic that could cause this in the w3wp.exe *32 IIS worker process for the website?

TIA

Which class of service are you using from Arvixe?  (Personal, Business, VPS, Dedicated)?
Did you make any custom changes?
Are you using any 3rd party plugins/templates, etc.?

Hi Greg,

Can you please PM me your ticket number?

Thank you,

Hello Greg,

Are you still having any issues with your website? or it is fixed?

I'm having a similar problem with 3.5 but haven't narrowed it down to any degree.  Same symptoms though.  Anyone with ideas?

Taz wrote:
I think similar issue. Once every couple of hours the site is very slow (> 20 seconds). After contacting my hosting provider, they say it might be a result of IIS running out of memory -> app pool recycle.
Not sure about CPU / Memory at that time, as they did not want to do further investigation...

I've shut down my nopCommerce site and application pool and I'm still having the same problem but the problem has only manifested after installing nopCommerce.  Not as frequent though. Still trying to locate the problem.

Taz wrote:

What hosting company? What kind of plan do you have?

Any errors in the Administration Logs?

Any kind of customization on the website (or using default nopCommerce out of the box)?

We have our own virtual servers in a data centre.  I use one as an application server and the other as the data server.

There are a stack of errors in the logs but the common one is "Error while running the 'Send emails' schedule task. An error occurred while executing the command definition. See the inner exception for details."  The email test works fine so I don't know why this would be failing except if it is trying to send while the server is hanging.  Nothing I can see looks like it would hang the server though, every error appears to result from a hung server not allowing a nopCommerce process to complete.

I have made very minor customisation to files in the 'View' directories.  Only things like adding 'clearfix' to the product specifications file so that FireFox would display specification data.  I've also uninstalled unused plugins so I'm using only one that is not part of the default install.

Looking at the server log files in the systems logs indicates that something is trying to access printers via terminal services and is generating an error when the drivers are not found.  These errors are occurring at nearly exactly the same time as server hangs.  I've raised a help request with our virtual server provider to investigate the server log errors.

Was there any resolution to this issue?

We host at a datacenter. The server is virtual. I have 3 different IIS implementations for the same store, live, qa, and development. The development (with debug on) and qa are fast. The live is slow. Even just pulling the contact us page takes 5 seconds in live. Less than a second in qa and dev.  I have compared the Application Pool to the other two and they are setup the same.

We are on Nopcommerce 3.2 with customization.