How does Cloud4Y technical support work?

If you visit our site , on it, like on many other sites, a feedback form from technical support will open with a proposal to talk. On the other side of the screen at this moment a whole team is working - support of the largest Russian cloud hosting. And today we want to tell you how it works from the inside.

By default, a sales specialist is responsible for online chat. In this case, of course, there you can ask any questions regarding sales, of a technical nature and general questions. If the questions are precisely on the technical side, then the request is transferred to technical support, which is already answered by phone, through the ticket system or by mail.

The support is divided into three lines:

• The first technical support line deals with accessibility and technical issues up to and including the OS level, starts tickets and advises clients. In the case of requests that do not affect the data and do not relate to finances, customer issues can be resolved by her using any of the available methods.
• The second technical support line deals with the issues of the level of hypervisors and OS, client software settings, customization, log analysis, integration issues and deep support for client information systems.
• The third line of support solves problems globally at the network level, the level of storage systems and is responsible for the architecture and accessibility as a whole.

If the support service receives a question related to finances, it transfers to the personal manager for this client, who identifies the client, advises if necessary and adds the required additional services. Additional services here are understood as some external things like L2 channel or USB key transfer to us, the client can change cloud resources independently and instantly. Almost all possible situations are written in our service contract, so when adding this or that additional service, additional agreements are not needed.

All three lines operate around the clock. Total support 12 people: five on the I and II lines, seven on the III. Five people during the day shift, and the most notorious owls, two on duty I and II lines, remain at the console at night . The duty III line works remotely.

There are several ways to ask a question: in addition to online chat on the site, these are tickets (“Client request” in your account), phone and mail. However, in the end, all requests flow into the ticket system. In particularly important issues, before processing the ticket, the personal manager contacts the client (the client’s contact person) and clarifies whether the client really wants to do this while specifying the name of the contact person, contact email and organization name.

Most often, users have queries related to the administration of virtual machines on VMware and XEN or the functioning of the cloud:

TOP3 of the most common questions:

• How to log in to a VM?
• How to expand resources (processor, memory, disk, network)?
• How to backup / restore from a copy?

That's all, of course, there are articles in our knowledge base with a detailed description and screenshots.

Unfortunately, typical solutions like “Tried to restart?” in our practice, almost never happens :) Such appeals are extremely rare.

As part of customer support, we often have to add users to access their terminal server (active directory, remote desktop), set up mailboxes, prescribe group policies, create dedicated private VLANs, install antiviruses, configure VPNs, forward USB keys, and administer MSSQL and 1C, plan and propose solutions to improve the architecture of client systems, etc. At the same time, we try to advise potential customers as fully as possible and offer such solutions in order to avoid difficulties in the future.

In the daytime, operators are on average 70-85% of their time, of which 40% -50% of II and III support lines are occupied by R&D (Research and Development). We use R&D to improve current services, expand the functionality of systems, implement / configure monitoring systems, update current systems, test and introduce new equipment, expand the cloud, etc.

At night, the average workload is - 20-25%.

The average time for phone communication with a client is about 5-7 minutes. The quickest conversation in less than a minute. The longest is 15 minutes.

On average, we process - 30 tickets per day, including those not yet closed, but awaiting any other clarifications from the client.


Ticket system through the eyes of the operator

The record number of calls is 58 tickets per day. We still do not know (since there were no problems with the infrastructure) what it was connected with - storms in the Sun or just such a day.

Before closing each ticket, the client can evaluate the solution to the request on a 5-point scale. In addition, personal managers contact customers to clarify the satisfaction of the solution to the problem and are interested in possible wishes.

What the?

Non-standard cases in our practice arise infrequently, but they do. At the end of last year, there was a serious DDoS attack with 5 million packets per second. Since we were not focused on the tasks of protecting against large DDoS attacks by our nature of business (deployment of infrastructure of corporate clients), we had a hard time. The attack was on our host, and on intermediate operators, watered everyone. We laid out a detailed description of the attack for our users on the site: cloud4y.ru/about/news/poyasneniya-k-intsidentu-svyazannogo-s-ddos-atakami-v-period-s-31-12-2013-po-05-01 -2014

But, as they say, that does not kill you makes you stronger. To fight off the attack, we:

• introduced additional communication channels
• introduced traffic filtering
• increased the number of uplinks
• added a secure feast, added the ability to transfer traffic through Voxility - the largest provider of Anti-DDoS solutions

In addition to DDoS, one can recall the case of one of our clients - a Russian system integrator and information technology provider for corporate customers, as well as a 1C solution provider. the client for some reason did not backup. Although we have repeatedly reminded him - you just need to put one tick in the cloud control panel. The client had meta data corruption on the ext3 file system and critical information was missing. As the employees of this company told us, if it had not been possible to restore the data, then they would have lost 1.5 million rubles at least.

Thanks to solutions from NetApp snap restore, we made snapshots and, after several attempts, ext3 was successfully restored. This method was the most suitable due to the extreme urgency of the issue. As a result, all customer data was restored. Admins in the end did not even get fired. They also came to our office to personally thank, and brought a very good beer in gratitude.

And today at our technical support Habr’s day. Ask questions and we will answer them in the comments. And thanks for watching :)