Increasing the Size of tmpDSK (/tmp)

You may have noticed that cPanel’s default partition size for /tmp is 512 MB, which in some cases can be way too small.  The /tmp partition on cPanel servers, assuming it was the one cPanel created and not one you did yourself, is a file-base

d partition that can easily be resized.

By default on most servers, /tmp is the temporary dumping place for a lot of things, for example:

• PHP session files
• PHP temporary file uploads
• MySQL temporary files
• Cache files for certain Apache modules

Most software that uses temporary files or sessions will automatically prefer to use /tmp – this folder is usually set to 777  permissions and therefore writeable by every user on the server.

When your /tmp partition fills up, it can cause noticeable problems for your users. If you run a larger server, the /tmp folder can fill up quickly and be very annoying as far as maintenance is concerned. Fortunately there is a very easy way to increase the size of this partition on a standalone server.

1. Stop MySQL, Apache, and cPanel to prevent writing to the /tmp partition
2. Copy the contents of /tmp to another location, such as /home (cp -rfp /tmp /home)
3. Unmount /tmp. If you’re unable to, you can do an lsof (lsof |grep /tmp) to see what processes are still writing to it, and kill them off. Or do a lazy unmount (umount -l /tmp) .
4. Delete /usr/tmpDSK (rm -rf /usr/tmpDSK)

Now open /scripts/securetmp and look for this line:

my $tmpdsksize     = 512000;    # Must be larger than 250000

And change the “512000” value to your desired size in MB, and save the file. Now run the following script to recreate /tmp:

/scripts/securetmp

This will recreate your /tmp (tmpDSK) partition using the size you specified. While the securetmp script may be overwritten in a cPanel update, the size of /tmp will not be affected one you alter its size.

By thecpaneladmin.com

How to FSCK a Linux Filesystem

There comes a time in every sysadmin’s life where filesystem errors just…happen.  Luckily, these are somewhat easy to fix, assuming you don’t have a greater problem involving physical hardware damage.

First, you need to know the name of the disk device having the problem.  Do a quick df to see what device the affected partition is on:

Filesystem      Size  Used Avail Use% Mounted on
/dev/sda3       2.7T  2.6T  106G  97% /
/usr/tmpDSK     4.0G  1.7G  2.2G  44% /tmp

Look under the “Filesystem” column to see the device name for the partition in question.  Now, if this is any filesystem but “/”, your job is probably going to be easy.  Simply unmount the file system and run a fsck against it.  For example, if you have a separate /home partition listed as /dev/sda3, you would do:

umount /home
fsck -yC /dev/sda3

There are a number of options for fsck, but the above combination is my personal preference.  The ‘y’ tells the fsck to fix whatever error sees, which is preferable unless you feel that your index finger has the stamina to hit ‘y’ 500 times in a row, and the ‘C’ prints out a pretty little progress bar so you can keep an eye on it.  Ext4 fiesystems fsck rather quickly – typically less than an hour for a 2TB filesystem.  Ext3 takes significantly longer.

Now, unmounting a filesystem may not be straight-forward – if any services are actively using files on that partition, the OS will refuse to unmount it.  Doing a lazy unmount (umount -l) won’t work here either – you need to unmount it cleanly.  To see what processes are using the filesystem in question, use lsof.  From the above example:

lsof -p |grep /var/

Then stop any services or processes using it.

If the filesystem issue is on your primary partition, you have a little more work ahead of you.  You’re going to need to boot into a rescue image.  To do this, simply use a Netinstall image and boot to the CentOS installation screen, then type:

linux rescue nomount

You can skip networking and all that jazz, then run the shell.  From there, you’ll need to find the partition in question and run the same fsck command.  Do note that on CentOS 6+, the device name may be incremented since it will count the rescue image as the first device in most cases.  So your /dev/sda3 might be /dev/sdb3 now, or even /dev/sdc3.

Once the fsck is done, reboot and confirm your filesystem is clean:

dumpe2fs -h /dev/sda3

The “Filesystem state” line should read “clean”.  If it doesn’t, the fsck either didn’t complete correctly, or you have a larger problem on your hands.

By thecpaneladmin.com

Changing Exim’s Sending IP

Anyone running a shared hosting server is probably now accustomed to dealing with constant complaints about blacklisting. It’s exim’s default setup on a cPanel server to use the shared IP of the server to send email, which means that all your clients on one server are sending out email on the same IP.  All it takes is for one spammer to send out a mailing list or one customer to get hacked and run a Darkmailer script, and suddenly everyone on your server seems to be complaining about getting their mail bounced.

You can change the server’s IP address for sending email. Assuming you already have an IP set up on your server with a valid PTR, you probably already saw on the cPanel forums or some other location that you can simply change the interface lines in your /etc/exim.conf file and restart exim:

remote_smtp:
driver = smtp
interface = ${if exists {/etc/mailips}{${lookup{$sender_address_domain}lsearch*{/etc/mailips}{$value}{}}}{}}
helo_data = ${if exists {/etc/mailhelo}{${lookup{$sender_address_domain}lsearch*{/etc/mailhelo}{$value}{$primary_hostname}}}{$primary_hostname}}

dk_remote_smtp:
driver = smtp
interface = ${if exists {/etc/mailips}{${lookup{$sender_address_domain}lsearch*{/etc/mailips}{$value}{}}}{}}
helo_data = ${if exists {/etc/mailhelo}{${lookup{$sender_address_domain}lsearch*{/etc/mailhelo}{$value}{$primary_hostname}}}{$primary_hostname}}
dk_private_key = "/var/cpanel/domain_keys/private/${dk_domain}"
dk_canon = nofws
dk_selector = default

In the above example, all you’d do is comment out the interface lines and replace them with:

interface = xx.xx.xx.xx

However, this tends to be a band-aid fix, as a cPanel update or any change made in WHM’s Exim Configuration Editor will regenerate the Exim config and overwrite your change. To make this more permanent, you’ll want to use the /etc/mailips file.

To set this up initially, go into WHM > Exim Configuration and enable this option:

** Send outgoing mail from the ip that matches the domain name in /etc/mailips (*: IP can be added to the file to change the main outgoing interface) [?]

Or, in /etc/exim.conf.localopts, add/change this line:

per_domain_mailips=1

Then run

/scripts/buildeximconf
service exim restart

*The /etc/mailips file should be root:exim, chmod 440  if it doesn’t already exist.

chattr +i /etc/mailips

Now for actually changing the IP:

Changing the IP Globally

If you want everyone on the server to send out on the same IP, just add the following to /etc/mailips:

*: xxx.xxx.xxx.xxx

Then add the IP and it’s matching PTR to /etc/mail_reverse_dns:

xxx.xxx.xxx.xxx hostname.tld

This will tell Exim to use that IP for any sender on the server.

Changing the IP Per Domain

If you want your users with dedicated IP addresses to be able to use those IPs to send email as well, you can add them to /etc/mailips. cPanel actually now has documentation on how to do this properly:

http://docs.cpanel.net/twiki/bin/view/AllDocumentation/WHMDocs/EximDifferentIps

If you have multiple dedicated IP domains already, I’ve devised a simple loop you can use to populate /etc/mailips automatically:

while read line ; do DOMAIN=`echo -e $line |awk '{print $2}'` && IP=`echo -e $line |awk '{print $1}' |cut -d: -f1` && echo "$DOMAIN: $IP" >> /etc/mailips ;done < /etc/domainips

You’ll always want the wildcard line to be in there to account for any domain not listed in the file, whether it is the main server’s IP or another that you have assigned for email:

*: xxx.xxx.xxx.xxx

Then:

cp /etc/domainips /etc/mail_reverse_dns

This will set all the existing sites on dedicated IPs to send out mail on those IPs.

By thecpaneladmin.com

SendGrid on cPanel/WHM

Setting up your cPanel server to send through a third-party mail server is very easy to do, if you understand the basics of how cPanel builds its Exim configs. You never want to edit your exim.conf file directly – your changes will be wiped out any time a cPanel update runs or someone makes a change via the Exim Configuration Editor in WHM.  Below is a quick guide on setting up a custom mail router with a provider like SendGrid, which in turn will route all outbound email through the external mail service.

The below instructions cover how to do this via command line.  If you prefer WHM, simply go to WHM -> Exim Configuration Manager -> Advanced Editor and alter the sections indicated.

First, open up your /etc/exim.conf.local file in an editor and look for the @AUTH@ section.  Modify it to look like this:

@AUTH@
sendgrid_login:
driver = plaintext
public_name = LOGIN
client_send = : <user> : <password>

Of course, replace <user> with your SendGrid username and <password> with your SendGrid account password.  If you already have something in the AUTH section, simply add this block of text below it.

Now look for the @PREROUTERS@ section, and modify it to look like this:

@PREROUTERS@
send_via_sendgrid:
driver = manualroute
domains = ! +local_domains
transport = sendgrid_smtp
route_list = "* smtp.sendgrid.net::587 byname"
host_find_failed = defer
no_more

The last modification should be to the @TRANSPORTSTART@ section:

@TRANSPORTSTART@
sendgrid_smtp:
driver = smtp
hosts = smtp.sendgrid.net
hosts_require_auth = smtp.sendgrid.net
hosts_require_tls = smtp.sendgrid.net

Now save the file, and apply the changes:

/scripts/buildeximconf

service exim restart

To test whether things are working, send an email out from your server, and look for it in /var/log/exim_mainlog. You should see something like this in your log entry:

2013-10-08 19:37:29 1VTjeS-0000Ac-O3 -> my@email R=send_via_sendgrid T=sendgrid_smtp H=smtp.sendgrid.net [x.x.x.x] X=TLSv1:DHE-RSA-AES256-SHA:256

If you have SPF records, you’ll need to add the hostname of the SMTP server to the record itself to allow the third-party mail server to send email on behalf of your domain.  Sendgrid will provide the hostname you should use.

By thecpaneladmin.com

Mandrill on cPanel/WHM

Mandrill is a transactional email service run by MailChimp, comparable to SendGrid.  It comes stock with a powerful API for fast implementation into applications for sending email reliably over SMTP, but it can also be used as a smart host for all of your server’s outgoing email.

The below instructions cover how to do this via command line.  If you prefer WHM, simply go to WHM -> Exim Configuration Manager -> Advanced Editor and alter the sections indicated.

First, open up your /etc/exim.conf.local file in an editor and look for the @AUTH@ section.  Modify it to look like this:

@AUTH@
auth_login:
driver = plaintext
public_name = LOGIN
hide client_send = : your@email : api_key

Note:  If exim.conf.local doesn’t exist, create it.  There’s a template here.

The value of api_key is not your Mandrill account password – it’s your API key.  You can find it in your account settings under “SMTP & API Credentials”

Also, replace your@email with your Mandrill account email address.  If you already have something in the AUTH section, simply add this block of text below it.

Now look for the @PREROUTERS@ section, and modify it to look like this:

@PREROUTERS@
smart_route:
driver = manualroute
domains = ! +local_domains
ignore_target_hosts = 127.0.0.0/8
transport = auth_relay
route_list = * smtp.mandrillapp.com
no_more

The last modification should be to the @TRANSPORTSTART@ section:

@TRANSPORTSTART@
auth_relay:
driver = smtp
port = 587
hosts_require_auth = $host_address
hosts_require_tls = $host_address

Save this file.  Now open /etc/exim.conf.localopts and add this line (or modify it and append the hostname to the existing line)

smarthost_routelist=*: smtp.mandrillapp.com

If you’re using WHM, this is under Basic Editor -> Mail -> Smarthost support.

Save the file, and apply the changes:

/scripts/buildeximconf

service exim restart

To test whether things are working, send an email out from your server, and look for it in /var/log/exim_mainlog. You should see something like this in your log entry:

2014-03-11 21:59:50 1WNbGg-0006ud-2d => to@email R=smart_route T=auth_relay H=smtp.us-east-1.mandrillapp.com [x.x.x.x] X=UNKNOWN:ECDHE-RSA-AES256-GCM-SHA384:256 A=auth_login C="250 2.0.0 Ok: queued as 3529E193E178"

If you have SPF records, you’ll need to add the hostname of the SMTP server to the record itself to allow the third-party mail server to send email on behalf of your domain.  Mandrill will provide the hostname you should use underSettings -> Sending Domains, when you do an SPF test.

By thecpaneladmin.com