Finally I made the necessary changes in Proxmox configuration for this issue.
See this: https://forum.proxmox.com/thre….26012/page-3#post-142829
The problem was related with the IO performance.
The LXC containers must be configured as chroot format and it's needed to make some changes un the fstab file where the MVs are stored.
I think I've found the source of the problem.
It's related to Proxmox 4.2 and Linux Containers.
Doing some more testing I found out that an import of a dump sql gets more than 7 minutes with a cpu load of 4,50 (last minute) in the Virtual Container.
The same import in an openvz container with other host with Proxmox 3 gets 8 seconds.
For those with the same problem... I found two posts regarding to this problem
They talk about the I/O write speed.
https://forum.proxmox.com/thre…-the-lxc-container.25773/
https://forum.proxmox.com/thre…grade-to-proxmox-4.25128/
I've not tested yet the solutions proposed there. I'll write back with my experiences.
You surely mean 4 Cores
Yes. sorry 
I'll try later some other testings, and will contact you if I can't manage with the problem.
Thanks.
You can install 1.3.0 even on LXC where bind mounts is not allowed. You must only think to deactivate the mount of httpd logs directories. See github.com/i-MSCP/imscp/blob/1…t_customer_logs-parameter
@Nuxwin Thanks for the tip. I've tested a new installation with a Debian MV and Imscp 1.3.0. But the problem is still there. The system is quite slow and cpu load increases up to 4.5
Concerning to the first MV configuration:
Display MorePHP version in use?
APCu ? Opcache in use?
How many memory assigned to the LXC container?
How many CPU assigned to the LXC container?
i-MSCP server implementation in use (ITK, Fcgid, PHP-FPM)?
PHP version in use? 5.5.9
APCu ? Opcache in use? Default drupal cache. Database cache.
How many memory assigned to the LXC container? 6 GB
How many CPU assigned to the LXC container? 4 GB
i-MSCP server implementation in use (ITK, Fcgid, PHP-FPM)? Fcgid (At the beginning I also tested php-fpm)
OS: Ubuntu Server 14.04.4
Well. I've tested a new Drupal 7 site installation with dummy content.
The pages loads a little faster but the problem is still there. Cpu load average increase till 3 or 4 when I open one or two more tabs and try to open other pages.
I have a server for development (also virtualized with proxmox, but openvz instead LXC) with the same Drupal site, and is quite fast.
I don't understand the reason of having a server with better hardware working worse.
The imscp panel is fast and it doesn't increase the cpu load average too much.
About memory consumption apache 2 has allocated 2GB and mysqld 1GB.
First upgrade to 1.3.0 please. Second try with a fresh drupal installation without any plugins...
Thanks for your reply.
I tried 1.3.0 with Ubuntu 16.04, but the panel failed when adding a new domain. I had a bind mounting error (read only), Because of this I ended installing 1.2.x.
I'm going to try a fresh drupal install and test it on 1.2.x
Hello.
I have a new install in a virtualized box. A OVH (So you start) server with Proxmox 4.2, Debian 8.5, the containers are LXC.
The main host is a Intel(R) Xeon(R) CPU E3-1245 V2 @ 3.40GHz, 8 cores, 32GB RAM.
The virtual container has a fresh install of Ubuntu Server 14.4 with Imscp 1.2.17.
The panel has only one site running, with a Drupal 7 site.
When I try to access a page it takes about 7 to 10 seconds to load. The cpu load increases to 2.00 or 3.00 in the container and the main host.
Accessing to a cached page is very fast.
Watching the processes I see php5-cgi and mysqld that are on top.
I've tested both php-fpm and fcgi.
I'm thinking that I there is a bottleneck in somewhere but don't know where is the problem.
I have other server with an old ispcp panel but similar hardware, with about 20 sites running and I don't have any similar issues.
Any idea of what is the problem?
Thanks.
