The Planet Blog

Hi there. Laurence Simon here. I work in The Planet’s call center. I do triage on weekday mornings, sweeping the ticket queues and picking off calls. Nice to meet ya.

You know, all things considered, it’s rather nice in here. Good ambient light, and there’s a great view of Buffalo Bayou and the Ferris wheel at the Downtown Aquarium. The chairs are pretty comfortable. And, I. Love. The Tunnels. (I highly recommend Treebeards‘ red beans and rice.)

However, there’s one thing that irks me: While we work 24/7, 365 days a year (or 366 this year!), come hurricane or hot air from the folks on television getting us all worked up over named storms, we can still make your experience even better.

I talked with my fellow trench-mates (pictured below), cobbled together a few notes, and jotted down a few suggestions, tips and thoughts that can give you a better, faster and more efficient experience with Technical Support:

1. Reboots

It takes just as long to set up a Manual Reboot Request (MRR) as it does to submit a trouble ticket or call in a reboot request when you’re at a computer. If you need a reboot for a server, use the ROD/Remote Reboot functions. If that fails, then use the Manual Reboot Request. You’ll see much quicker results this way.

Submitting a normal ticket with “Please reboot my server” requires that a technician go into Orbit, submit a reboot request for you, and then close out the original ticket. It takes much longer to get a server rebooted this way than by using ROD or Manual.

2. Initial Setup

The first thing you should do with your server is set the hostname, get that hostname resolving in DNS, and then request it as the Reverse DNS. This will prevent a lot of common network issues from the get-go.

The hostname should be a fully qualified domain name that points to the server. For example: server.exampledomain.com.

To get a Reverse DNS entry set up, just submit a trouble ticket with Reverse DNS as the subject line or a DNS Change Ticket. Make sure to include the IP address and the hostname you want to point it at.

3. Escalation Procedures

Prepare and post a set of escalation procedures for common situations so that we don’t need to ask you to perform basic recovery tasks.

Let us know in advance what to do if an alert comes up and the server requires a reboot, if it is OK to run an FSCK on the drive when it’s needed, or to replace failed drives the moment they lock up.

4. Readily Accessible Information

Keep a card with your account number and password — with bogus characters — handy in case you’re away from your desk.

For instance, if your password is:

1L0v3Kev1nH4z4rd

Besides the fact that you’ve got some serious issues to work through (or the fact that you are actually Kevin Hazard), you can bogus this up with some XJQP magic…

x1L0jv3Keqv1nH4zp4rd

If someone steals your wallet, they won’t have the actual password. Just remember to remove the X, J, Q and P characters and you have your password.

For additional security, you can require that we ask a security question, like your mother’s maiden name. That’ll also throw off anyone who gets a hold of the account.

5. Keep a Calendar

Keep a wall calendar handy with reminders when SSLs and domains are up for renewal. Also, program them as appointment reminders in your cell phone.

A multimillion-dollar operation can be brought to its knees for a day when its domain expires. Renew it for as long in advance as you can afford, and be sure to keep the account information for logging into your registrar handy. (Or, if you want to make sure it’s all in one place, make The Planet your domain registrar.)

(NOTE: According to Whois, ilovekevinhazard.com is available)

6. Notification Address

Do not set your notification address to one on your server. When you use an external contact address, you can get updates when the server’s down or if there is an abuse or legal issue.

If you lost your cell phone, you wouldn’t want people calling your cell phone to tell you they’ve found your cell phone, right?

Be sure to check that mailbox frequently, or have it alert you.

7. Up to Date

Keep all your contact information up to date. This tip is a continuation of No. 6, but it’s important enough to reiterate on its own.

8. Plan Ahead

If I were to tell you, “Your server is down and we cannot recover any data off of it,” what would you do?

Disaster planning and testing your recovery procedure on a regular basis is critical to any online business. Spending some time and money with our sales department exploring backup and redundancy will prevent you from losing clients, money and sleep.

As I like to say, “Terminator 2: Judgment Day was just a big advertisement for offsite backups.”

9. Cut out the Middle Man

Simple changes to the A Records of DNS entries can be done in Orbit instead of via a ticket.

Just go to DNS Administration, bring up the domain, make the changes, and save them.

This is actually easier to do than submitting a ticket requesting the change, and it’s a lot quicker.

10. Help Us Help You

Keep your passwords updated in Orbit and let us know what port SSH is listening to. Include any wheel users or special instructions for login.

This will save us time in addressing your issue because it will let us get into your server the moment we get the ticket.

It’s like keeping a key in a fake rock in the garden so the neighbor can feed your cat (or James Caan if you’ve got a Kathy Bates Misery thing going on).

11. Firewalled

If you have a firewall, please allow the Technical Support group access to your server.

It’s important to have a firewall on your server for security purposes, whether it’s one of our dedicated external firewall products or just a set of iptables or Windows Defender. But it’s also important that we are able to access your server quickly should you raise the alarm.

12. Be Specific

Please make the ticket summaries descriptive and unambiguous. “Help!” and “Urgent!” and “I am losing my business!” make it difficult to quickly assign issues. After the issue is resolved, they also don’t help us track down previous issues with a server to establish patterns of problematic hardware, software or networking.

We understand that it’s an emergency, but please don’t let your panic prevent us from being able to resolve the problem quickly or learn from that incident in to assist you in the future.

13. Include as Much Information as Possible

Always include error messages, exactly what you did before getting those errors and any relevant sections of log files to demonstrate what problems you are running into. If it’s with a specific domain or account, provide those, too. A ticket with “This doesn’t work!” doesn’t tell us much, since we need to know more about “this” and how it’s not working.

The more information we have about the problem, the better. Otherwise, there may be some back-and-forth on the ticket that ends up taking hours instead of minutes.

Here’s the biggest one that folks have told me about:

14. Trust Us, We’re Here for You

After a ticket is submitted, a lot of customers will speed-dial Technical Support, asking them to read the ticket and address it immediately.

We’re here to assist, but jumping the queue by calling in isn’t necessarily effective in escalating your ticket. We triage issues based on the severity of the incident, which technicians are best equipped to handle a given problem, and if the ticket needs to be handled by any of our other teams (Networking, Data Centers, Professional Services, Advanced Services, Abuse, etc.).

And, unlike those pesky elevators which actually do show up quicker when you press the button multiple times, calling in to check if a reboot has been completed will not get a server rebooted faster. Our call center is downtown while the Houston data centers are at least 30 miles away. (Dallas is even farther than that!) Data center technicians work through the queues as quickly as possible, and they are generally able to respond to tickets within minutes of submission.

We understand that you’re frustrated when your business depends on your server, and sometimes that materializes in blowing off some steam and passing along some of the heat that your own clients are dumping on you.

One odd thing I’ve noticed in my time here is that I’ve actually become a lot nicer and more patient with service representatives on the telephone. I’ve been reading the FAQs and submitting trouble tickets when possible. I know what it’s like on the other end of the line. I know what it’s like to be them.

OK, so that’s all I’ve got for this time around. I’ll have more next time.

Until then, save me the last flagon of iced tea in the break room, and thank you for choosing The Planet!

-Laurence