Cloud Linux Inc [Forum]

Kernel-Devel Problem

Sat, 18 May 2013 12:07:32 -0400

Kernel-Devel Problem 2.6.32-379.22.1.lve1.2.20.el6 in forum General Discussion.
Hi all,

I need to install a kernel module for my network card so im having to use the following rpm package:

kernel-devel-2.6.32-379.22.1.lve1.2.20.el6.x86_64.rpm

This�doesn't�work however, i think because the�kernel�installed when using cldeploy is:

2.6.32-379.22.1.lve1.2.17.el6

I cant yum upgrade with no network connectivity so im kind of stuck with that kernel.

Is the difference in versions likely to be causing this issue? im seeing the same error as reported here:

http://www.cloudlinux.com/solutions/forum/messages/forum18/topic285/message1345/?sphrase_id=76826#message1345

If so is there anywhere i can get the older kernel-devel package or upgrade the kernel manually with rpm?

Thanks in advance!
18.05.2013 12:07:32, sam hodgson.

MySQL Governor: Error: Unknown Kernel

Thu, 16 May 2013 19:11:21 -0400

MySQL Governor: Error: Unknown Kernel in forum General Discussion.
Hello,

I installed MySQL Governor, but got a lot of errors when installing. However, it seems to have installed.

But when I restart MySQL, I get this:

# service mysql restart
error: Unknown kernel
Shutting down MySQL....... SUCCESS!�
error: Unknown kernel
Starting MySQL. SUCCESS!�

It is very strange. What can I do?
16.05.2013 19:11:21, Brandon Olivares.

Kernel 0day exploit

Tue, 14 May 2013 15:34:25 -0400

Kernel 0day exploit in forum General Discussion.
Hello, is CloudLinux secure from this, has an updated kernel been released which fixes this vulnerability?

https://news.ycombinator.com/item?id=5703758
https://patchwork.kernel.org/patch/2441281/
14.05.2013 15:34:25, NWC IS.

network limits

Wed, 08 May 2013 10:28:52 -0400

network limits feature request in forum General Discussion.
Hi,

I found this discussion started some time ago�
here�http://www.cloudlinux.com/solutions/forum/messages/forum18/topic306/message1447/#message1447and here�http://www.cloudlinux.com/solutions/forum/messages/forum18/topic273/message1302/#message1302

I just wanted to know is there any progress for 'network limits' feature? According to Igor's posts fr om above threads we could get network limits with interface virtualization or packets marking with LVE id.

I'm asking because we are getting issues with DoS attacks originating from our users. These incidents sometimes lead to outage on our servers. The goal is to have an ability to lim it single tenant on network speed, bytes and packets per second. Possibly we can use owner match for iptables but we'd like to avoid creating endless list of iptables and qos rules. I think virtualized network interface fits here the best. Additionally virtual nic should allow ip binding for each user.

Please let us know is there any plan for that feature.
08.05.2013 10:28:52, Pawel Panek.

[Feature Request] Ability to disable swap usage

Tue, 07 May 2013 04:35:29 -0400

[Feature Request] Ability to disable swap usage in forum General Discussion.
Hello,
I know Cloudlinux use swap area as RAM, and you have tested this function. I know you say it works well, but on my system, there a lot of differences in speed time with CentOS and CloudLinux.

CentOS use swap only if RAM is not enough, and it's correct. Cloudlinux use swap if think that is better (with an algorithm I think).

But on my system, with 50% of ram free (1GB on 2GB), usage of swap make all the system slow.

I know you think it's a great feature, and maybe it works well with a lot of servers, but please, add the ability to disable this thing, and make CloudLinux use swap only if really necessary.

Thanks
07.05.2013 04:35:29, Nicola Bettini.

Reducing Disk IO by Optimizing LVE & CageFS w/ SSD

Sun, 05 May 2013 18:10:36 -0400

Reducing Disk IO by Optimizing LVE & CageFS w/ SSD Attempting to reduce disk IO generated by CageFS in forum General Discussion.
Hello,

I'm working on a server with the following:
====code====

CloudLinux 5.9 (2.6.18-448.3.1.el5.lve0.8.64)

# df -h
Filesystem      Size� Used Avail Use% Mounted on
/dev/sda2       9.9G� 2.4G� 7.0G� 26% /
/dev/sda1       487M�� 38M� 424M�� 9% /boot
/dev/sda8       864G� 741G�� 80G� 91% /home
/dev/sda7       2.0G� 154M� 1.8G�� 9% /tmp
/dev/sda6      � 20G� 9.4G� 9.4G� 51% /usr
/dev/sda5      � 20G� 8.8G�� 10G� 47% /var
/dev/sdb1       1.8T� 1.2T� 549G� 69% /backup
/dev/sdc1      � 74G� 6.2G�� 64G�� 9% /var/lib/mysql

=============
While watching top and iostat, I see extended periods of 100% utilization of /home and /var. These seem to correlate with the following process running and staying near the top of the top output:

/usr/sbin/cagefsctl --force-update

/dev/sdc is an SSD drive. I'm curious if anyone has any suggestions on reducing the disk IO that the above and the lvestats-server process generate with moving certain files to the SSD as one potential method of accomplishing this goal. Other methods more than welcome.

I think this may be related: cldetect --cxs-installed
I couldn't find any documentation on what this does, but I just now ran it, as cxswatch is also running on this server. Does anyone have any info on what this process does and if it can help with disk IO?

Thanks for any insight!
05.05.2013 18:10:36, Nick Teeple.

Two Feature Requests

Sat, 27 Apr 2013 12:20:38 -0400

Two Feature Requests in forum General Discussion.
Two things I'd love to see in CloudLinux:

1. Burstable limits (soft limits), especially for RAM. For instance, I'd love to be able to specify a limit of 512 MB, but burstable up to say 1 GB or something like that.

2. Measure CPU in MHz. Not sure if this is possible at all, but give each account a certain measure of the CPU, in MHz instead of percent. Right now we're doing this by calculating the closest percent to the amount we want to give, but it'd be nice to be able to specify this.
27.04.2013 12:20:38, Brandon Olivares.

Limits for Resellers

Fri, 26 Apr 2013 23:28:56 -0400

Limits for Resellers in forum General Discussion.
Hello,

I there a way to limit a reseller so that all of their accounts must not exceed the total of that limit? For instance, if I allocate a reseller 2 GB RAM and 30% CPU, then all of that reseller's accounts cannot exceed as a whole those resources? I don't want to apply individual limits, but actually charge the reseller for the resources.

Any suggestions, or plans for implementing this kind of feature?
26.04.2013 23:28:56, Brandon Olivares.

CloudLinux with MariaDB and Litespeed

Thu, 18 Apr 2013 17:28:12 -0400

CloudLinux with MariaDB and Litespeed Getting CloudLinux to work with MariaDB and Litespeed in forum General Discussion.
Hello,

I've been using CloudLinux. I installed MariaDB to replace MySQL. However, now, MySQL Governor doesn't seem to work. dbtop shows all 0s.

I also have litespeed installed. When I run lvetop, only �nobody� is displayed.

What do I need to do to have these stats properly calculated so that resource usage can be done?
18.04.2013 17:28:12, Brandon Olivares.

CloudLinux 503 errors and Litespeed's automatic restarts on 503 error

Sun, 14 Apr 2013 06:06:08 -0400

CloudLinux 503 errors and Litespeed's automatic restarts on 503 error in forum General Discussion.
CloudLinux gives a 503 error when EP limit is reached or when the user goes well over the top with CPU limit.

Litespeeds listens for 503 errors and does an automatic restart of the web server.

This means that when the server is using alot of ressources beacuse of a few accounts hitting their maximum allowed ressources Litespeed starts restarting the webserver.

Litespeed say to set CloudLinux's EP limit higher than Litespeeds Suexec max conn limit which is a pain as we currently have different EP limits per type of account.

I do not however see a solution for a user going well over his CPU limit�

Is there a way to change Cloudlinux's error message code to replace 503 Service unavailable with something like 408 Request Time out to prevent Litespeed from restarting their webserver ?

Also 408 sounds more user freindly that 503 that souds more link an internal error�
14.04.2013 06:06:08, Richard Hordern.

Wordpress brute attacks

Fri, 12 Apr 2013 12:23:00 -0400

Wordpress brute attacks in forum General Discussion.
As you might be aware of the current HUGE brute force attacks worldwide going on.

More info:
http://www.webhostingtalk.com/showthread.php?t=1255387

Is there anything that CloudLinux can help to mitigate such attacks :)
12.04.2013 12:23:00, Rajesh Kumar.

mod_ruid and 11.38

Sat, 18 May 2013 04:08:42 -0400

mod_ruid and 11.38 in forum CloudLinux and cPanel.
Does this:
http://releases.cpanel.net/releases/11-38/jail-system-update/

"Users of mod_ruid2 can now enable �Jailed apache� �support which will chroot() each virtual host into their virtfs."
Makes us a�little�safer now ? I still remember Igor saying it was the best bad idea ever. Anyway. (to use mod_ruid)

So we can run it more "safe" or not now ?
Because I was re-considering trying ruid with DSO.

Accelerators like APC / Xcache / Eaccelerator can run in such environment by the way (mod_ruid with dso)
18.05.2013 04:08:42, Christos Panagiotakis.

Upgrade to MySQL 5.5

Fri, 17 May 2013 13:38:29 -0400

Upgrade to MySQL 5.5 in forum CloudLinux and cPanel.
Hello,

I just installed CloudLinux on a new server yesterday. However, the server had MySQL 5.1, so mysql-governor installed 5.1 as well. However, I want to upgrade it to 5.5. How can I do this?
17.05.2013 13:38:29, Brandon Olivares.

use lve wrapper to reduce I/O for cpbackup

Thu, 09 May 2013 06:08:02 -0400

use lve wrapper to reduce I/O for cpbackup in forum CloudLinux and cPanel.
As many cPanel administrators know, the backup process (cpbackup) can be a real resource hog.�
Especially the disk i/o can bring a busy server to a grinding halt.

Would it be possible to use lve_wrapper or lve_suwrapper to wrap the cpbackup process within an LVE, and thus gaining control of the choking I/O usage of the backup process?
If so, how do I change the cron for cpbackup to allow for lve wrapping?

information about the wrappers on�http://docs.cloudlinux.com/index.html?lve_wrappers.html
09.05.2013 06:08:02, Ernest Bruggeman.

extended root partition

Wed, 08 May 2013 22:03:54 -0400

extended root partition in forum CloudLinux and cPanel.
Dear All,

currently, I'm using Cloudlinux + Cpanel. unfortunately, my root�partition�still have free space only 1.5GB/20GB.

Do you have any idea that can extend my root partition.

my OS running on file system ext3

Best Regards,
08.05.2013 22:03:54, steven elvisda.

virtual memory vs physical memory

Thu, 02 May 2013 21:23:12 -0400

virtual memory vs physical memory in forum CloudLinux and cPanel.
i believe cloudlinux allow us to specify how many virtual memory and physical memory can be assign to one account.� my question is, can i set virtual memory to unlimited which i can set as 0 in the ve.cfg and physical memory to 2 or 3 GB? if this can be done.

1. what will happened if the pysical memory has been fully utilize?
2. will the unlimited virtual memory i set just now will affect actual physical memory which i install on the server?
3. is there any specific calculation that i can use to calculate the maximum concurrent connection that my server can handle.

please advise.
02.05.2013 21:23:12, faizal karim.

cpanel - centos 6.4 new install fails

Tue, 30 Apr 2013 03:59:01 -0400

cpanel - centos 6.4 new install fails in forum CloudLinux and cPanel.
hi guys,

i am trying to install the trial on a new server with cpanel/whm installed, and i am getting this:

conversion=true
components=true
Loaded plugins: fastestmirror, presto
Cleaning repos: base extras rpmforge updates
Cleaning up Everything
Cleaning up list of fastest mirrors
0 delta-package files removed, by presto
You have third-party kernel module rpm installed.
It's not compatible with CloudLinux kernel.
Please contact CloudLinux support at https://helpdesk.cloudlinux.com

any idea what's up?
30.04.2013 03:59:01, stefanita rares.

lveps error

Mon, 29 Apr 2013 06:52:14 -0400

lveps error in forum CloudLinux and cPanel.
Hello, I just noticed this was showing alot last night one my server in�//usr/local/cpanel/logs/error_log

error: invalid time
lveps version 1.2 release 11.el6.cloudlinux
usage: lveps [-p] [-n] [-o ] [-d] [-c ] [-s ] [-t] [-h]
� � � �-p to print per-process/per-thread statistics
� � � �-n to print LVE ID istead of username
� � � �-o to use formatted output (fmt=id,ep,pid,tid,cpu,mem,io)
� � � �-d to show dynamic cpu usage instead of total cpu usage
� � � �-c to calculate average cpu usage for seconds (used with -d)
� � � �-r to run under realtime priority for more accuracy (needs privileges)
� � � �-s to sort LVEs in output (cpu, process, thread, mem, io)
� � � �-t to run in the top-mode
� � � �-h to print this brief help message
error: invalid time
lveps version 1.2 release 11.el6.cloudlinux

Should I ignore this or is this a bug?
29.04.2013 06:52:14, NWC IS.

No cache info available. APC does not appear to be running.

Sun, 28 Apr 2013 00:20:17 -0400

No cache info available. APC does not appear to be running. I've installed apc but its not working in forum CloudLinux and cPanel.
Hello there
i am facing problems on running apc on cloud linux cpanel server.�
i never faced such thing while running cent os.�
I've installed apc and i've added .so�extension�and also apc options to my php.ini but it seems apc is not running.
i'm using cagefs also. is there any particular thing that has to be done in order to get apc working on cloudlinux?
thank u in advance.
28.04.2013 00:20:17, Mahdi Hazaveh.

Help me, How to login on VMware Images CloudLinux?

Wed, 24 Apr 2013 07:45:59 -0400

Help me, How to login on VMware Images CloudLinux? login, aws-cl6-cpanel.vmdk in forum CloudLinux and cPanel.
Sorry I don�t good English
I download file http://dropbox.cloudlinux.com/images/cl6/aws-cl6-cpanel.vmdk
I run it but I cannot login in tty1. Because I don�t know password login
It here: http://docs.cloudlinux.com/index.html?vmware_and_kvm_images.html

[IMG]http://i.imgur.com/pHGjWLA.png[/IMG]

Pic: http://i.imgur.com/pHGjWLA.png

Please Help me, thank you! :)

PS/
And I download file ISO CloudLinux 6.4 ISO
http://docs.cloudlinux.com/index.html?vmware_and_kvm_images.html
http://repo.cloudlinux.com/cloudlinux/6.4/iso/x86_64/CloudLinux-6.4-x86_64-DVD.iso

I setup ok, but I cannot connect SSH CloudLinux�.
(root and every user cannot login ssh�):(
24.04.2013 07:45:59, thanh thanh.

PHP Selector custom PHP without modules ghost option

Wed, 17 Apr 2013 19:11:10 -0400

PHP Selector custom PHP without modules ghost option PHP Selector givesa ghost module option called 'directory' in forum CloudLinux and cPanel.
(SOLVED)

I am setting up PHP selector with a few custom build PHP versions. All pre-configured, modules loaded as needed.
Now when a customer (or the WHM admin user through the LVE manager) selects one of those PHP version's I would have expected there to be no modules to sel ect (to enable or disable). Just as when you select 'native'.

But whatever I try I always end up with one module option called 'directory'?

Just once did I manage to get rid of it by making an exact copy of the PHP directory in /opt/alt/ and adding it with a different name. But I am unable to replicate this 'solution'.

For customers, the option does not do anything. They can switch between the custom versions just fine.
Using the LVE Manager, when I select it to make it default and click save an error pops up stating "Alternatives directory not found".
I have compared my PHP's with the default folders fr om the alt-php packages but can't find anything wrong. I cannot find anything named 'directory' either somewhere (nothing in my php.ini either).

Anyone who has an idea wh ere this comes fr om and/or how to get rid of this last ghost option?
17.04.2013 19:11:10, Thijs Stuurman.

PHP Selector, CageFS and Litespeed Issues

Wed, 17 Apr 2013 08:40:49 -0400

PHP Selector, CageFS and Litespeed Issues in forum CloudLinux and cPanel.
I have been having issues trying to get PHP Selector to work on Litespeed. It was basically changing the opt php.ini files correctly but not changing the PHP version itself.

I followed the following document :
http://docs.cloudlinux.com/index.html?litespeed_support.html

and got everything exactly as it should be.
It was originally pointing to the /opt files but now goes to default location at:�/usr/local/lib/php.ini
I then decided to do the following:

Change everything back on litespeed. 2 settings only. So didn't think I need to reinstall it.

Uninstalled CageFS and PHP Selector and reinstalled both and enabled all users on CageFS.

Still no luck. No one points to /opt php.ini files. Any ideas what could be wrong?
17.04.2013 08:40:49, Tasneem Jacobs.

PHP Handler per account

Tue, 16 Apr 2013 18:08:00 -0400

PHP Handler per account in forum CloudLinux and cPanel.
Just wondering, is it possible in the near future
to sel ect php handler with the same way as we select php version fr om selector? (or other way)

For example, virtualmin does it, and I can choose if I want mod_php, cgi or fcgi per domain/account.

Is (or "Will") something like this be possible ?
16.04.2013 18:08:00, Christos Panagiotakis.

Can you specify a CPU/RAM value per cPanel package?

Mon, 15 Apr 2013 03:55:05 -0400

Can you specify a CPU/RAM value per cPanel package? in forum CloudLinux and cPanel.
As per post title, is it possible to specify RAM/CPU/etc LVE values for different cPanel packages?

ie) Basic package gets 10% CPU, advanced package gets 20% CPU

I can't really see any indication of such a feature, but just wanted to check.
15.04.2013 03:55:05, Alan Harris.

PHP Selector php.ini settings

Sun, 14 Apr 2013 15:15:56 -0400

PHP Selector php.ini settings in forum CloudLinux and cPanel.
One objection, correct me if I am wrong, but
post_max_size shouldn't be equal with max upload ?
Because one has as highest value 1G and the other 128MB.

And�disable_functions can easily implemented on all ? It's handy and useful both ways.
If a) someone want to shut something down for security reasons, or b) if he wants to have an empty line
in case I shut something down in global php.ini for security reasons.

And I don't know if that's doable but, if for example someone use xcache, eaccelerator, suhosin or apc,
shouldn't have an area (custom or not), (could start a textbox with fixed "suhosin." for example)�
for any custom module that needs customization ?

For example if someone will check suhosin as a module he might also need to adjust variables
like�

suhosin.get.max_value_length or
suhosin.max_input_vars and
suhosin.request.max_varname_length to make thing work propery (like forums, ipb or vbulletin, etc)

or ajdust opcode caching like
xcache.* ? or
eaccelerator.xx.xxx ?
and so on...

Of course if my thoughts are crap, where can I find the per user's generated php.ini to change / add something like this
(suhosin changes, or disable_functions etc) without touching the global php.ini ?
14.04.2013 15:15:56, Christos Panagiotakis.

Separate MySQL server and CL / db_governor

Fri, 12 Apr 2013 05:36:26 -0400

Separate MySQL server and CL / db_governor in forum CloudLinux and cPanel.
First of all I didn't try it so I was wondering if anyone is using separate server for mysql if it really worth it.
(the "Setup Remote Mysql Server" in WHM) using cloudlinux in front cpanel server.

And if it worth it (faster loading, less cpu, more accounts on the same servers etc), can db governor work
in the separate mysqld system (cloudlinux on the mysql server) ?
12.04.2013 05:36:26, Christos Panagiotakis.

yum SSL: certificate failure

Thu, 11 Apr 2013 05:19:24 -0400

yum SSL: certificate failure in forum CloudLinux and cPanel.
Hi,

I keep getting errors like this, if i want to download the package readline-devel, which is needed for compiling EasyApache.

Error Downloading Packages:
� readline-devel-6.0-4.el6.x86_64: failed to retrieve getPackage/readline-devel-6.0-4.el6.x86_64.rpm from cloudlinux-x86_64-server-6
error was [Errno 14] PYCURL ERROR 51 - "SSL: certificate subject name 'cl-mirror.uk2group.com' does not match target host name 'cl-mirror.pidginhost.com'"
11.04.2013 05:19:24, Andreas Rademacher.

Fcgid settings

Tue, 09 Apr 2013 02:52:26 -0400

Fcgid settings in forum CloudLinux and cPanel.
Based on the settings on the blog does this looks like ..sane?

====quote====

FcgidMinProcessesPerClass 0
FcgidMaxProcessesPerClass 8
FcgidMaxProcesses 200
FcgidIdleTimeout 180
FcgidProcessLifeTime 320
FcgidIdleScanInterval 60
FcgidMaxRequestLen 536870913
FcgidErrorScanInterval 30
FcgidFixPathinfo 1
FcgidIOTimeout 99

=============

Keep in mind that, there are a few sites that are running in-browser cron jobs and other weird stuff
and they need to be alive constantly to update database / sending mails etc.

Any thoughts, advice, if something needs changing, adding or modifying anything, is welcome :-)
09.04.2013 02:52:26, Christos Panagiotakis.

Select PHP Selector not displayed in DA

Wed, 08 May 2013 06:19:22 -0400

Select PHP Selector not displayed in DA in forum CloudLinux and DirectAdmin.
Hello,
When installing PHP Selector not displayed in DA..
Any ideas what could be wrong?

Thank you all for your help...

check the screenshot

08.05.2013 06:19:22, webdomsk.

Cloudlinux and Plesk panel execution mode apache, fastcgi and cgi

Tue, 09 Apr 2013 07:23:44 -0400

Cloudlinux and Plesk panel execution mode apache, fastcgi and cgi in forum CloudLinux and Plesk.
Hello,
I have installed a trial version for testing purpose, we'll going to migrate 5000 domains to 5 servers with cloudlinux so we are testing with a wordpress installation. I notice some issue or things that i dont understand:

- cl-selector: Only work if the domains is in fastcgi or cgi execution, i think because the apache mode execution is with apache user so i have to change the apache user properties with cl-selector. Is right?

- cagefs: If i execute the wordpress with the "localhost" as database url only work if i execute in apache mode, when i change to fastcgi or cgi mode i get an error if i execute in apache mode not. The error in fastcgi and cgi mode is "Cant connect to database", if i disable the cagefs to the user that run wordpress i can connect but when i enable again i get the error again. Why?�

I have to add two rpm to cagefsctl, "psa-mod-fcgid-configurator" and "psa-mod_fcgid". Without this rpm cant execute the domains in fastcgi or cgi mode with plesk panel. Can be relation the error to the database connection in fastcgi or cgi with this?

Someone else with this problems?

Best Regards.
09.04.2013 07:23:44, Jose Joaquin.