cPanel/WHM slow on VMware Server

[iggi]

Has anyone noticed a long delay on VMware Server 2.0 for certain scripts i.e. when loading roundcube webmail, loading support center, etc. Restart helps this issue, but turns worse in a few hours.

Here is my Setup:
Host OS:
Centos 5.2 x86
Core 2 Quad 2.4GHz
4GB RAM
Gigabit connection

Guest OS:
2x 3.125 GHz
2GB RAM

Load is about 0.90 on VPS and clicking support center took about 5 min to load, webmail took about the same time. Any ideas, im running the CURRENT install on cPanel/WHM

[_jman]

It could be a myriad number of things, not enough swap, a slowly failing HD, some runaway process...

Try top (or htop if you have it installed) to see if anything is hogging your CPU. Do that right after re-boot, then in an hour or two, and look for any process taking up an excessive amount of resources.

Could also be other users on the box, one bad thing about Virtuozzo is its ability to be oversold. Xen servers don't have that problem.

HTH

[iggi]

there are 2 VM's on the VMWARE machine, one for cpanel and one windows box that is not setup yet, essentially the vmware virtual machine has all the resources it needs. I checked top while the issue happened on the VM and seeing no issues, load is at about 0.40 and nothing using a lot of resources, but still taking 5 minutes to load a page. I left top open while I was loading the WHM page and I noticed top freezes while page is loading. Any idea's? possibly a network issue? I am running VMware Server 2.0 with vmware tools installed.

[cPanelDavidG]

At this time, Roundcube is known to cause high CPU utilization. If Roundcube is the source of these issues for you, I recommend reading:

http://forums.cpanel.net/f5/upcoming...nt-132629.html

[mikelegg]

I've noticed this on a brand new VMWare VPS too.

There is very little load on the CPU (0.0 to 0.10) but when I log in to WHM, although the login popup appears almost instantly, the main page takes about a minute to load.

The websites on the server run like lightning, so it's not the machine itself that is struggling, it's just WHM.

[cPanelDon]

I've noticed this on a brand new VMWare VPS too.

There is very little load on the CPU (0.0 to 0.10) but when I log in to WHM, although the login popup appears almost instantly, the main page takes about a minute to load.

The websites on the server run like lightning, so it's not the machine itself that is struggling, it's just WHM.

To help better diagnose the symptoms it will help greatly to know additional details:

1. Please provide the output from the following command (via root SSH access):
   

   # uname -a && grep -H '' /etc/*release /usr/local/cpanel/version /var/cpanel/envtype

2. Do you have "root" access to the hardware node and VMware configuration?
3. Is the issue exhibited in any other virtual servers with cPanel/WHM on the same hardware node?
4. What is the total number of virtual servers that are hosted on the same hardware node?
5. Where does the access attempt originate when testing WHM; for example, are you accessing via a remote system, external or outside of the same network, or are you accessing WHM locally from the same system running VMware?
6. Is any delay experienced when logging-into SSH via root access from a remote connection (in an external network)?
7. When the issue is exhibited via WHM, are you logged in as "root" or are you logged-in as a reseller?
8. Are any errors or warnings displayed or logged while testing cPanel, as an end-user, and while testing WHM, as both "root" and as a reseller? When testing both cPanel and WHM areas of the control panel, please use the following command, via root SSH access, to monitor various log information for possible clues as to why the difficulty may be occurring:
   

   # tail -fvn0 /usr/local/cpanel/logs/*log /var/log/messages /var/log/secure

[_Dejan_]

Hi,

I have too VMware VPS but I don't have any problems with it. Before this vmware VPS Im have virtuozzo VPS with very similar configuration and all has been so slow. For example forum pages sometime has need 2-20sec. to open now opening is instant. Also WHM is much faster...
VMWare ESX 4.0 Update1 is latest(Build 208167 <- from vSphere Web Access info).
Maybe your HDD's are to slow or you have network problems???
My node using SSD's for:
-ESX installation
-Host(CentOS 5.4 x64) have /boot, /tmp and swap on SSD Array and / on HDD array

and as Im say all work very fast...

[mikelegg]

Hi Don

Interestingly, while I was doing the things below, WHM was running fine.

1. # uname -a && grep -H '' /etc/*release /usr/local/cpanel/version /var/cpanel/envtype

Linux hyperion.sslserver.net.au 2.6.18-164.11.1.el5 #1 SMP Wed Jan 20 07:39:04 EST 2010 i686 i686 i386 GNU/Linux
/etc/redhat-release:CentOS release 5.4 (Final)
/usr/local/cpanel/version:11.25.0-RELEASE_43473
/var/cpanel/envtype:vmware

2. Do you have "root" access to the hardware node and VMware configuration?

No, I only have root access to the virtual machine itself. But I can forward information/requests to the NOC staff who manage the hardware.

3. Is the issue exhibited in any other virtual servers with cPanel/WHM on the same hardware node?

This is my first time running WHM on VMware, so I can't really answer that question.

4. What is the total number of virtual servers that are hosted on the same hardware node?

I'm not privy to that information, but I do have faith that the hardware is not overloaded.

5. Where does the access attempt originate when testing WHM; for example, are you accessing via a remote system, external or outside of the same network, or are you accessing WHM locally from the same system running VMware?

Remotely.

6. Is any delay experienced when logging-into SSH via root access from a remote connection (in an external network)?

No, not at all.

7. When the issue is exhibited via WHM, are you logged in as "root" or are you logged-in as a reseller?

Root

8. Are any errors or warnings displayed or logged while testing cPanel, as an end-user, and while testing WHM, as both "root" and as a reseller? When testing both cPanel and WHM areas of the control panel, please use the following command, via root SSH access, to monitor various log information for possible clues as to why the difficulty may be occurring:

# tail -fvn0 /usr/local/cpanel/logs/*log /var/log/messages /var/log/secure

The only weird thing I saw was this ...

==> Loaded RRDs: version 1.3006
==> WARNING: The configured processor count does not match the
==> actual processor count (2)! Running log analysis programs
==> on this system may cause excessive load! You should set "extracpus"
==> to "0" in /var/cpanel/cpanel.config if this is not ok.

So I've changed extracpus to 0 and run /usr/local/cpanel/whostmgr/bin/whostmgr2 --updatetweaksettings.