While working one of the clients server
We saw an issue where going from WHM > plugin > Softacalous and opening it resulted in blank page.
The issue is due to corrupt files in softacalous.
To fix this issue softacalous need to be reinstalled
to reinstall to go
cd /usr/local/cpanel/whostmgr/docroot/cgi/
rm -rf softaculous
cd /home
wget -N http://files.softaculous.com/install.sh chmod 755 install.sh ./install.sh Once done softacalous was working fine
While working on a server on installing php 5.6 on centos 6 OS
I was getting error “Error: Cannot retrieve metalink for repository: epel. Please verify its path and try again”
After googleing around I found this issue is due to https mentioned in the .repo file in /etc/yum.repos.d
[root@host yum.repos.d]# ls -la
total 28
drwxr-xr-x. 2 root root 4096 Oct 28 05:15 .
drwxr-xr-x. 80 root root 4096 Oct 28 05:16 ..
-rw-r–r–. 1 root root 954 Oct 28 05:15 epel.repo
-rw-r–r–. 1 root root 1056 Nov 4 2012 epel-testing.repo
-rw-r–r–. 1 root root 966 Jul 4 2015 webtatic-archive.repo
-rw-r–r–. 1 root root 862 Oct 28 05:15 webtatic.repo
-rw-r–r–. 1 root root 966 Jul 4 2015 webtatic-testing.repo
I checked in all files with .repo extension and an replaced all https with http and all repository started working fine
It would and could change hostname 🙂
LOG: MAIN cwd=/root 4 args: /usr/sbin/exim -i -v testemail@gmail.com LOG: MAIN REJECT F=<root@darkmatter.dnsprotect.com> rejected by non-SMTP ACL: failed to expand ACL string "${if eq{$originator_uid}{${perl{user2uid}{nobody}}}{1}{0}}": Undefined subroutine &main::user2uid called. LOG: MAIN cwd=/var/spool/exim 8 args: /usr/sbin/exim -v -t -oem -oi -f <> -E1UY3r2-000
root@server-web [~]# /scripts/buildeximconf Configuration file passes test! New configuration file was installed. /etc/exim.pl.local.build syntax OK Enabled system filter options: attachments|spam_rewrite .......... ... ........ /etc/exim.pl.local installed! razor2 is not installed, disabling it in SpamAssassin to save memory pyzor is not installed, disabling it in SpamAssassin to save memory SPF is disabled in exim or unavailable, enabling SPF for SpamAssassin Refreshing SMTP Mail protection.
root@server-web [~]# cp -rpf /etc/exim.pl.local /etc/exim.pl
We were having an issue where spams emails were sent from single cpanel account
We could not suspend the account and disable emails
So the solution to this was to keep website online and disable only emails
we use the below method to disable emails for a single cPanel account.
SSH to server and use below steps to disable cpanel for any account
The newwebho is the cpanel user name
root@online [~]# chmod 0 /home/newwebho/etc/
root@online[~]# chattr +ia /home/newwebho/etc/
root@eu [~]#
Softacalous license from GUI interface is just click of mouse work
However if you are doing it from cli
You need to run below cron to refresh the softacalous license
root@vps [~]# /usr/local/cpanel/3rdparty/bin/php /usr/local/cpanel/whostmgr/docroot/cgi/softaculous/cron.php
root@vps [~]#
While working on some issue on clients server I could
see large number of DNS request in /var/log/messages which were flooding DNS server
Jul 27 12:48:55 oklahoma named[14354]: client 74.125.72.2#33485: view external: query (cache) ‘192.245.129.104.in-addr.arpa/PTR/IN’ denied
Jul 27 12:48:55 oklahoma named[14354]: client 173.194.103.8#53106: view external: query (cache) ‘31.249.129.104.in-addr.arpa/PTR/IN’ denied
Jul 27 12:48:55 oklahoma named[14354]: client 74.125.72.147#39348: view external: query (cache) ‘27.249.129.104.in-addr.arpa/PTR/IN’ denied
Jul 27 12:48:55 oklahoma named[14354]: client 173.194.103.7#40721: view external: query (cache) ‘34.249.129.104.in-addr.arpa/PTR/IN’ denied
Jul 27 12:48:55 oklahoma named[14354]: client 173.194.90.9#63592: view external: query (cache) ‘192.245.129.104.in-addr.arpa/PTR/IN’ denied
The solution for such issues is by adding few parameter in /etc/named.conf which is bind configuration file
Add the following under options:
recursion no;
additional-from-auth no;
additional-from-cache no;
and restart named
and now I see no such request coming to DNS which has fixed DNS flood issue
On up-to-date RHEL5 or CentOS5 (currently that means 5.10 aka 5U10)
Add the following line to /etc/sysctl.conf:
|
1
|
net.ipv6.conf.all.disable_ipv6 = 1
|
On a live system you can disable it with:
|
1
|
echo 1 > /proc/sys/net/ipv6/conf/all/disable_ipv6
|
While working on WHMCS I was not able to generate tickets
I found out issue was due to mysql being in strict mode
I tried disabling mysql strict mode by adding entries in /etc/my.cnf
sql_mode=””
However it didnt work out.
I have found that there is a /usr/my.cnf file that is enabling strict mode on server.
This file contained the following:
====
sql_mode=NO_ENGINE_
====
I removed the STRICT_TRANS_TABLES option and saved the file. All that is left is to restart the MySQL server. You can do this at your convenience with the following command:
====
/scripts/restartsrv_mysql
====
Once it is restarted, strict mode should be disabled.
While creating of backup of mysql I was getting below error
mysqldump: Error: ‘Out of resources when opening file ‘/tmp/#sql_466b_2.MAI’ (Errcode: 24 “Too many open files”)’
The error is due to Mysql exceeding the limit of open files on mysql
So I just went ahead and saw the open file limit
MariaDB [(none)]> SHOW VARIABLES LIKE ‘open%’
->
-> ;
+——————+——-+
| Variable_name | Value |
+——————+——-+
| open_files_limit | 1024 |
+——————+——-+
1 row in set (0.00 sec)
I went ahead and added a parameter in /etc/my.cnf and restarted mysql to fix this issue
open_files_limit = 5000