Doctor walks in, smiles and picks up the chart. He starts examining the man and as he writes a prescription advises he keep sanitary and wash his hands several times a day.
Do you trust this doctor…especially after he prescribes vigorous hand-washing, but forgot to wash his own before poking and prodding during the exam.
Obviously this doctor loses credibility. This moral is a lesson that MSPs (managed service providers) must heed. MSPs frequently tell their clients to treat data in a secure fashion lest it compromised-by hack, by carelessness or by lax standards. It is only natural to expect your MSP to abide by certain security practices to prevent client data from becoming exposed.
Now don’t read this as an indictment of any MSP. The great majority are quite vigilant and do practice what they preach. In fact many are excellent proponents of security best practices. However, MSPs have a lot of responsibility for the IT assets of their clients. It is expected that MSPs will take certain precautions to safeguard client data and infrastructure. In this case, it’s not just the client and their sphere of access that touch the data, but oftentimes the employees/consultants/techs of the MSP themselves – and if their back door is open, the risk can be transferred to the client.
If you currently use an MSP, in the market for one, or are one yourself, there is a quick hit list you review to ensure assets are properly secure on both sides of the server.
Some of the following items may help you begin to think about what security protocols you should have in your managed services practice: Of the six, four of them are validated by the MSP Alliance (in fact there is a very good document titled Universal Best Practices for Managed Services by Charles Weaver) that includes a security section. The other line items I cover come from 30 years of security and most recently cloud-based options. They are:
Real time monitoring: Just because you are running back-up tapes, running compliance reports or simply sitting on a calm Saturday afternoon watching a ballgame, your network is active. It’s being pinged, accessed, surveilled, and penetrated—some innocuous, some business related and, many to many, suspicious. It’s obvious the last one we care about. Many companies use a log management or archiving solution which captures all the suspect activity. Problem is…when is the designated person going to review the reams of machine code? Days, weeks later? By then, the issue has festered; the former employee has made a copy of your sales database, the hacker has left a nasty Trojan gift or the customer’s sensitive data has flown the coup. However, if you apply the best practice of having SIEM or another intrusion detection protocol working in conjunction with your log, in real time, then any suspicious activity is immediately alerted and the risk can be cut at the knees before it has a chance to do any damage.
Human Resource safeguards: The ability to provision and de-provision users immediately not only saves time, but regulates access to only those who need to see certain files and/or applications. When a company (or a client if managed by a MSP) hires a new employee, there must be a simple way to add them to the network and automatically provide them with necessary access based on that employee’s role. Conversely, when an employee leaves the company, there must be an equally simple way to retire the account and prevent anyone from using that account to access corporate information. It also prevents usage of unsanctioned applications from a corporately controlled computer. These functions are typically handled by an Identity Management solution.
Strong authentication: This is more than telling clients they can’t use “password” or “12345” as passwords. Credentialing is the core of security. And multi-factor authentication is the best practice needed to apply that core. Yes there are passwords, and Multi-factor authentication can mean multiple passwords, however, attackers and would-be impersonators wouldn’t have to know just one shared secret, but two! (Or three, or four!) However, multiple passwords suffer the same frailties as traditional passwords — we’re awash in passwords already, adding more won’t help. So multi-factor authentication methods usually rely on something you have (like a device)in addition to something you know. For instance you log in to a single sign on application centrally controlled in the cloud. You successfully enter in your password. The system then sends verification to another device which can then be used to gain permitted access.
Unique User IDs: This is simple…no shared admin accounts!
Encryption of data: For MSPs and clients for are beholden to regulatory compliance and data governance, this is not only a “should have,” but a “must have.” It covers data that is sent via email (in transit) but also the millions a Terabytes sitting behind a firewall. Encryption isn’t the only method that can protect your data. Tokenization is an up-and-coming technique to remove sensitive data from applications and storage and replace it with placeholder characters called tokens. The benefit of tokens is that they are completely random and there is no algorithm that can turn them back into the real data they represent. As many companies have cloud-based applications there are additional considerations each company must make based on their specific business needs. But remember, in the end, the client/end user is the one responsible (read that SLA!!!) for the data.
Unification: You can all the best tools across the Net. You might have excellent processes. BUT, if the left hand doesn’t know what the right is doing, you’re operating at a disadvantage. By unifying the solutions as a centralized and layered initiative you enhance your network visibility and better understand the context of certain activities. For instance…after three failed attempts, an account is accessed. Is this a problem? Well, that depends. Identity management says it is a legitimate user—they used the right credential (eventually). Log Management says this access happened at 3:00 a.m. On it’s own, that’s feasible. SIEM picks up that the IP address is traced back to a site in China. All three independently and without context, could be a legitimate user. But unified and analyzed from a central station in real time tells a much different story.
These six best practices require a variety of technologies (and processes). Some MSPs have the leverage to have them all in house or partnered through various developers. For others, it is not as affordable—until you investigate that all these functions, capabilities and power can be deployed, managed and controlled from the cloud. And this makes it a very affordable proposition for any managed service provider AND any of their clients who realize enterprise-class security might be out of their grasp. Cloud-based security is one of the fastest growing
There are many other security measures you can implement; including technologies you can employ to help you safeguard your practice. Whichever path you take, make sure you at least consider all the ways you need to protect yourself, both from internal and external threats. If you protect yourself it is a lot easier to protect your clients.
Who’s cloud-based security solution company DOES practice what it preaches