Hello,

I have used ManageWP (http://managewp.com) for years, even while using nDeploy and 99.9% of the time it works great with and without nDeploy.

However, it appears that either Naxsi or Proxy_Cache when using the PROXY backend setup is causing ManageWP to report a 503 error when attempting to connect to the sites on my server.

Please help!

#PAGESPEED_NOT_ENABLED;
#Proxy Pass to a backend webserver
#proxy_cache enabled
location / {
    include /etc/nginx/conf.d/naxsi_learn.rules;

    proxy_send_timeout   800;
    proxy_read_timeout   800;

    proxy_buffer_size    32k;
    proxy_buffers     16 32k;
    proxy_busy_buffers_size 64k;
    proxy_temp_file_write_size 64k;

    proxy_connect_timeout 180s;

    proxy_pass   http://209.59.174.61:9999;
    proxy_set_header   Host   $host;
    proxy_set_header   X-Real-IP  $remote_addr;
    proxy_set_header   X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_set_header   X-Forwarded-Proto $scheme;
    proxy_redirect     off;
    proxy_cache            PROXYCACHE;
    proxy_cache_valid      200  1d;
    proxy_cache_use_stale  error timeout invalid_header updating http_500 http_502 http_503 http_504;
  }
  location /NaxsiRequestDenied {
         return 418;
     }

Hello,

Can you please advise as to how to go about using EasyApache4 with nDeploy? Can we go back to using system installed PHP versions?

l am uninstalled nDeploy but under cPanel still showing this option. please let me know, how to delete nDeploy files completely?

http://prntscr.com/7pyyqb

Hi @AnoopAlias

I have installed LetsEncrypt and created the yaml file as per the wiki instructions below. I have enabled this for a PROXY as well as PHP backend setup, neither appear to be working.

I'm not seeing any .well-known folders being created or anything either in the domains which is what I'd expect.

Please advise...

https://support.sysally.net/projects/ndeploy/wiki/LetsEncrypt_setup

Hello,

In latest nDeploy version, users with dedicated ips is not working.

Thanks

Hello,

I am getting error message when you try to install, If run : /usr/nginx/scripts/nginx-passenger-setup.sh, Please can you let me know, How to fixing this error ?

root@rg [~]# /usr/nginx/scripts/nginx-passenger-setup.sh
Setting up Phusion Passenger
gpg: requesting key D39DC0E3 from hkp server keys.gnupg.net
gpg: keyserver timed out
gpg: keyserver receive failed: Keyserver error
Downloading https://github.com/rvm/rvm/archive/1.26.11.tar.gz
Downloading https://github.com/rvm/rvm/releases/download/1.26.11/1.26.11.tar.gz.asc
gpg: Signature made Mon 30 Mar 2015 05:52:13 PM EDT using RSA key ID BF04FF17
gpg: Can't check signature: No public key
Warning, RVM 1.26.0 introduces signed releases and automated check of signatures when GPG software found.
Assuming you trust Michal Papis import the mpapis public key (downloading the signatures).

GPG signature verification failed for '/usr/local/rvm/archives/rvm-1.26.11.tgz' - 'https://github.com/rvm/rvm/releases/download/1.26.11/1.26.11.tar.gz.asc'!
try downloading the signatures:

sudo gpg2 --keyserver hkp://keys.gnupg.net --recv-keys 409B6B1796C275462A1703113804BB82D39DC0E3

or if it fails:

command curl -sSL https://rvm.io/mpapis.asc | sudo gpg2 --import -

the key can be compared with:

https://rvm.io/mpapis.asc
https://keybase.io/mpapis

/usr/nginx/scripts/nginx-passenger-setup.sh: line 5: rvm: command not found
/usr/nginx/scripts/nginx-passenger-setup.sh: line 8: /usr/local/rvm/rubies/ruby-2.2.2/bin/gem: No such file or directory
ln: target `/usr/local/rvm/gems/ruby-2.2.2/gems/passenger-5.0.11/' is not a directory: No such file or directory
Stopping nginx: [FAILED]
Starting nginx: nginx: [emerg] bind() to 194.145.208.207:80 failed (98: Address already in use)
nginx: [emerg] bind() to 194.145.208.207:80 failed (98: Address already in use)
nginx: [emerg] bind() to 194.145.208.207:80 failed (98: Address already in use)
nginx: [emerg] bind() to 194.145.208.207:80 failed (98: Address already in use)
nginx: [emerg] bind() to 194.145.208.207:80 failed (98: Address already in use)
nginx: [emerg] still could not bind() [FAILED]

root@rg [~]# /opt/nDeploy/scripts/cpanel-nDeploy-setup.sh enable
httpd restarted successfully.
Stopping nginx: [FAILED]
Starting nginx: nginx: [alert] Unable to start Phusion Passenger: Support binary PassengerAgent not found (tried: /usr/local/rvm/gems/ruby-2.2.2/gems/passenger-5.0.11/buildout/support-binaries/PassengerAgent and /root/.passenger/support-binaries/5.0.11/PassengerAgent). This probably means that your Phusion Passenger installation is broken or incomplete, or that your 'passenger_root' setting contains the wrong value. Please reinstall Phusion Passenger or adjust the setting (see: https://www.phusionpassenger.com/documentation/Users%20guide%20Nginx.html#PassengerRoot). (-1: Unknown error)
[ OK ]
Shutting down Filesystem event daemon (incrond): [FAILED]
Starting Filesystem event daemon (incrond): [ OK ]
Stopping nginx: [ OK ]
Starting nginx: nginx: [alert] Unable to start Phusion Passenger: Support binary PassengerAgent not found (tried: /usr/local/rvm/gems/ruby-2.2.2/gems/passenger-5.0.11/buildout/support-binaries/PassengerAgent and /root/.passenger/support-binaries/5.0.11/PassengerAgent). This probably means that your Phusion Passenger installation is broken or incomplete, or that your 'passenger_root' setting contains the wrong value. Please reinstall Phusion Passenger or adjust the setting (see: https://www.phusionpassenger.com/documentation/Users%20guide%20Nginx.html#PassengerRoot). (-1: Unknown error)

Everytime I update nDeploy I have to go back into nginx.conf and update the client_max_body_size value to 128m from 10m as I often times need to upload larger files to WordPress sites.

So, when attempting to visit https://support.sysally.net/projects/ndeploy/wiki on Windows 10 I'm suddenly getting this warning from Avast.

I can still pull up the page via Mac & Linux without an issue, however, these obviously are not running Avast Antivirus.

Any ideas?

access log must be buffered .

Additionally an open log file descriptor cache must be defined.

What is the supported way to install memcache for NginX & PHP-FPM?

Deleting a wildcard domain will also delete the nginx nDeploy settings for the subdomains configured for the domain other than the wildcard

Hi Anoop,

I'm attempting to use Rocket-Nginx for a few of my sites (https://github.com/maximejobin/rocket-nginx) and am having trouble with getting it setup.

Then, in your configuration file, you must include the configuration. If your websites configurations are in /etc/nginx/sites-available, you need to alter your configuration:

server {
  ...

  # Rocket-Nginx configuration
  include rocket-nginx/rocket-nginx.conf;

  ...
}

Where would I put this for it to apply to a single domain using NGINX via nDeploy?

Hello,

I "still" (?) have the issue where deleted accounts are still present in php-fpm's config :

root@web-01:~# grep rusth /opt/nDeploy/watcher.log
[03-Feb-2016 13:07:03] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 13:07:07] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 13:07:11] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:02:49] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:02:51] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:02:55] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:02:56] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:00] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:01] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:05] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:07] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:11] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:13] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'
[03-Feb-2016 17:03:17] ERROR: [pool rusthisp] cannot get uid for user 'rusthisp'


root@web-01:~# grep rusth /var/cpanel/accounting.log
Sat Oct 24 20:52:49 2015:CREATE:root:root:domain.es:5.135.103.187:rusthisp
Wed Feb  3 00:00:34 2016:REMOVE:root:root:domain.es:rusthisp

And so some domains (??, why not All ?) are giving 503 Errors

When running the nDeploy_watcher restart command, everything is fine again.

I'm using the last version (updated daily) .. Do you have any clue/need more information ?

Hello,

I got a few errors during install, I don't have them now, but I know one of them was because I already had a /usr/local/apache/conf/includes/pre_virtualhost_global.conf and it didn't add a newline before adding:
SetEnvIf X-Forwarded-Proto https HTTPS=on

Are we supposed to have a directory for fpm pools configuration per user? where do I find that?

I see large directory trees at /opt/remi/ what are those about?

where do I find the apache config for how it hooks into php-fpm socket?

Do we need to use nginx? Is there a way to just use straight apache with fpm? What does nginx buy us?

How can I tell if a page is being served up by fpm?

thanks.

I'm planning to purhase a cpanel plugin that enables automatic installation of Let's Encrypt SSL on my WHM.
https://letsencrypt.org/

Here is their link: https://letsencrypt-for-cpanel.com/

I have asked them about a conflict with nDeploy but they are not so sure so I wanna ask the same question to you.

Can I able to use their plugin or should I manually install Let's Encrypt SSL.
If you suggest manually, can you tell me how ?

I dont wanna mess up with the configuration on nDeploy.

Hello,

The wiki is broken :-( https://support.sysally.net/projects/ndeploy/wiki

Are there any reasons on why you're not using the Wiki hosted by Github ? :-)

Regards !

EDIT: looks like it was the case before https://github.com/AnoopAlias/nDeploy/blob/0ec6c939ca459bce326c92756d00a8dfd4998431/rpm_buildtree/nDeploy-pkg/opt/nDeploy/README.md

What's going on here

https://github.com/AnoopAlias/nDeploy/blob/master/scripts/event_trigger.sh#L10 (after &&)

Why you execute domain-data file ?

• lfd on [redacted hostname]: Suspicious process running under user [redacted]

User:dockdogs PID:31493 PPID:26660 Run Time:298(secs) Memory:393972(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31498 PPID:26660 Run Time:297(secs) Memory:401960(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31500 PPID:26660 Run Time:297(secs) Memory:401060(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31501 PPID:26660 Run Time:296(secs) Memory:394948(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31502 PPID:26660 Run Time:296(secs) Memory:393924(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31506 PPID:26660 Run Time:296(secs) Memory:393924(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31507 PPID:26660 Run Time:296(secs) Memory:393672(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs
User:dockdogs PID:31512 PPID:26660 Run Time:294(secs) Memory:424312(kb) exe:/usr/local/phpbrew/php/php-5.5.20/sbin/php-fpm cmd:php-fpm: pool dockdogs

Looks like the conf file is mangling with the page html as the conf is not properly html escaped.

Escaping html seems to complicate things as well.

Looks like we may need to provide a hint and save the current config in user home so users can download ,edit and update with the custom config update script.

It would be a step backward on the usability side..but looks like that is the only option available.

CentOS 7, CloudLinux + cPanel:

After executing /opt/nDeploy/scripts/attempt_autofix.sh

Attempting to restart ndeploy_watcher daemon
Job for ndeploy_watcher.service failed because the control process exited with error code. See "systemctl status ndeploy_watcher.service" and "journalctl -xe" for details.
Redirecting to /bin/systemctl restart ndeploy_watcher.service
Job for ndeploy_watcher.service failed because the control process exited with error code. See "systemctl status ndeploy_watcher.service" and "journalctl -xe" for details.

systemctl restart ndeploy_watcher

Job for ndeploy_watcher.service failed because the control process exited with error code. See "systemctl status ndeploy_watcher.service" and "journalctl -xe" for details.

systemctl status ndeploy_watcher.service
● ndeploy_watcher.service - nDeploy Watcher
Loaded: loaded (/usr/lib/systemd/system/ndeploy_watcher.service; enabled; vendor preset: disabled)
Active: failed (Result: exit-code) since Tue 2015-12-22 09:36:35 AEDT; 5min ago
Process: 410716 ExecStart=/opt/nDeploy/scripts/watcher.py start (code=exited, status=1/FAILURE)

Dec 22 09:36:34 systemd[1]: Starting nDeploy Watcher...
Dec 22 09:36:35 systemd[1]: ndeploy_watcher.service: control process exited, code=exited status=1
Dec 22 09:36:35 systemd[1]: Failed to start nDeploy Watcher.
Dec 22 09:36:35 systemd[1]: Unit ndeploy_watcher.service entered failed state.
Dec 22 09:36:35 systemd[1]: ndeploy_watcher.service failed.

However, when running
/opt/nDeploy/scripts/watcher.py restart
it does not provide output

/opt/nDeploy/scripts/watcher.py start
pidfile /opt/nDeploy/watcher.pid already exists. Daemon already running?

ps waux | grep watcher
root 410498 0.0 0.0 433748 12216 ? Sl 09:35 0:00 python /opt/nDeploy/scripts/watcher.py start

I did update ndeploy to the latest version today, with this outcome. Any ideas of what could be wrong?

Hello,

I am receiving an error on a new cpanel account I created. php-fpm is failing to start for that account alone. The php-fpm interface in the cpanel for that account is showing:

Apache PHP-FPM plugin not enabled
(!) click on the php-fpm icon above to restart the configuration process anytime

This is the php-fpm error log for that account

[proxy:error] pid 382647No such file or directory: AH02454: FCGI: attempt to connect to Unix domain socket /opt/fpmsockets/poconosg.sock (*) failed
[Thu Sep 17 19:38:02.552395 2015] [proxy_fcgi:error] [pid 382647] [client xxx.xxx.xxx.xxx:57043] AH01079: failed to make connection to backend: httpd-UDS, referer: http://domaincreated.com/

(First of all thanks for making nDeploy possible ! :) We've ran into some issues (like, for instance, on account deletiong -> the .conf file was kept) but looks like it's solved now (I've pulled the scripts/ folder))

I have a small issue, though, with the current setup: when someone installs a SSL certificate, it is not taken in account by nginx, and we have to manually restart nginx. (not just reload, really restart)

I haven't found the right place to change this behaviour, and I don't know if it's a tested feature or not ... But I would like to get some help, please :)

I'd like to see the default_server.conf changed to also listen on the host's shared IPv6 address if applicable, and also have a default https server (also listening on IPv6 and IPv4 shared address) in order to ensure that all web requests coming in go somewhere, even if that's just to the WHM "Default Web Site" page.

When a cPanel account is removed ;the nginx config for the slave servers are not deleted

The problem lies in the fact that SSL certificate install does not provide any hooks in their API by cPanel.

Currently we use inotify and incron on /var/cpanel/users/ file to detect cpanel events with no hooks and fire a reconfiguration . The problem is that cPanel doesnt change files in SSL cert installation. But cPanel do change yaml files in /var/cpanel/userdata for all our usecases .

Why not use inotify to watch /var/cpanel/userdata ? - Well inotify doesnt support recursive file events and /var/cpanel/userdata has subdirectories where events happen .

So the solution ? - Write a custom inotify daemon to listen recursively on /var/cpanel/userdata or use ready-made scripts like https://github.com/gregghz/Watcher .

This is targeted for the next release of nDeploy plugin .

Hello,

As per (https://support.sysally.net/projects/ndeploy/wiki/Mod_remoteip_) I have installed Mod_RemoteIP via the provided Yum command and when I execute

vim /etc/apache2/conf.modules.d/360_mod_remoteip.conf it has the following lines in it by default:

# Enable mod_remoteip
LoadModule remoteip_module modules/mod_remoteip.so

• I believe you're stating that I need to add the lines below to this file with my Servers IP address, correct?
• What if I have multiple IPs on my server?

RemoteIPHeader X-Forwarded-For
RemoteIPTrustedProxy x.x.x.x/24 #substitute x.x.x.x with yout IP address

Thanks!

Running the latest version of nDeploy on a 64 bit CentOS 7 server, I'm getting the nginx worker processes segfaulting fairly frequently. This is on a server with only a single test site with just me loading pages. This happens on the stock CentOS 7 kernel as well as a GRsec enabled kernel. Any thoughts where to start looking?

From /var/log/nginx/error.log:

2016/02/15 18:49:51 [notice] 9582#9582: signal process started
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8054 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8147 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8049 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8051 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8082 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8194 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8050 exited on signal 11
2016/02/15 18:49:53 [alert] 1487#1487: worker process 8052 exited on signal 11

And from syslog:

Feb 15 18:49:53  kernel: nginx[8194]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel: nginx[8052]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8049]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8051]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8054]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8147]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8050]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel: nginx[8082]: segfault at 0 ip 00000000004a5d9b sp 00007ffc36362738 error 4
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]
Feb 15 18:49:53  kernel:  in nginx[400000+b97000]

Zend opcache is enabled by default for NginX-PHP-FPM
My application support Xcache only!
How to enable Xcache instead of Zend Opcache?

@AnoopAlias

Please help!

When run easy_php_setup.sh

[11-Jan-2016 17:36:00] NOTICE: PHP message: PHP Warning: PHP Startup: Unable to load dynamic library '/opt/remi/php56/root/usr/lib64/php/modules/intl.so' - libicui18n.so.50: cannot open shared object file: No such file or directory in Unknown on line 0
[11-Jan-2016 17:36:00] NOTICE: PHP message: PHP Warning: PHP Startup: Unable to load dynamic library '/opt/remi/php54/root/usr/lib64/php/modules/intl.so' - libicui18n.so.50: cannot open shared object file: No such file or directory in Unknown on line 0
[11-Jan-2016 17:36:00] NOTICE: PHP message: PHP Warning: PHP Startup: Unable to load dynamic library '/opt/remi/php55/root/usr/lib64/php/modules/intl.so' - libicui18n.so.50: cannot open shared object file: No such file or directory in Unknown on line 0

• But packages allready installed on system.
• latest version of nDeploy.

Hello,

When you installed nDeploy on my server the YUM repo wasn't available. Can you please advise on what I need to do to to now use the YUM based install method?

Thanks,
BK

Cuisine is a wrapper over Fabric and therefore does not work well with Python 2.6 which is the default Python is CentOS 6.x . If we can switch the remote script execution part (which is not a big part in there) to Ansible (or salt for that matter ) ;cluster functionality can be easily brought to CentOS-6 .

Although CentOS7 is the future . I think 80 percent of cPanel installations as of now are on CentOS6

By default all of the EasyApache4 builds are configured with MPM_Worker, is there a reason the REMI builds or phpbrew builds always suggest being used with MPM_EVENT?

If I need to use MPM_Event how can I switch to this by default?

Thanks!

@AnoopAlias I finally have the PHPBrew 5.5.20 build working with all the correct modules and configuration to work with my main sites.

However, I'd like to update to a newer version of 5.5.X with the same modules compiled in. How can I do this?

Thanks!

The cluster deploy script now use mixed bash scripting and ansible cli for remote tasks . I think its better to convert the entire stuff to an ansible playbook

@AnoopAlias -- HELP!

I installed on a server hosting more than 1000 websites.
When i run this /opt/nDeploy/scripts/apache_php-fpm_setup.sh enable the sites return internal server errors so i have to disable it. I have tried it several times but it is not working.
I am running cloudlinux. Any help would be appreciated. Thanks

When modifying an account with an IPv6 address assigned, the following error is generated from generate_config.py

Domain::Data::Modify thenumber
Traceback (most recent call last):
  File "/opt/nDeploy/scripts/generate_config.py", line 568, in <module>
    nginx_confgen(cpaneluser, main_domain)  # Generate conf for main domain
  File "/opt/nDeploy/scripts/generate_config.py", line 437, in nginx_confgen
    ipv6_addr = str(yaml_parsed_cpaneldomain.get('ipv6').keys()[0])
AttributeError: 'str' object has no attribute 'keys'

There's only a single ipv6 key in the userdata yaml for this account, and it looks like you're not handling that case correctly?

nginx vhost for /etc/nginx/sites-enabled/DOMAIN_SSL.conf not working

@AnoopAlias

How can I add the directives I submitted in #35 so that they are applied to both PROXY & PHP based configurations?

Thanks!

I'm not seeing in your documentation. Is there a way to uninstall it after installation if we don't like it?

Hello,

Without visiting https://support.sysally.net/projects/ndeploy/news is there a way to be notified of the need to run the update command when a new release is made?

Thanks!

@AnoopAlias -- Can you please check on a site on my server (user dockbiz) and see why it's forcing to go to https://domainname instead of http://domainname?

I'm attempting to install https://github.com/snipe/snipe-it