Office 365-Focused Botnet Puts Spotlight on Security of System Accounts
    The recent discovery of a botnet aimed at Office 365  customers puts a spotlight on a commonly overlooked category of system  accounts.
Skyhigh Networks this month reported a botnet it dubbed "KnockKnock"  that it discovered several weeks ago. Active since at least May, and especially  active from June through August, the relatively small botnet seems to have been  highly targeted in both the types of accounts it attacked and the types of  organizations it went after. 
"The reason this is interesting is not that a botnet is  trying to get into accounts, but the fact that it is trying to get into system  accounts," said Sekhar Sarukkai, chief scientist at Skyhigh Networks,  in  an interview.
What the attack does, according to Skyhigh's description, is  go after the system accounts that are commonly used to connect the Exchange  Online e-mail system with marketing and sales automation software. In cases  where the system accounts were compromised, KnockKnock exported data from the  inbox, created a new inbox rule and began a phishing attack from the account  against the rest of the organization.
Skyhigh picked up evidence of the botnet through its Cloud  Access Security Broker (CASB) Threat Protection engine when the company's  customers were attacked. Skyhigh says the traffic came from 63 networks and 83  IP addresses, with 90 percent of traffic coming from IP addresses in China. In  all, the attacks came from 16 countries.
The attacks averaged only five e-mail addresses per customer.  Additionally, the organizational targeting was extremely specific -- aimed at  infrastructure and Internet of Things (IoT) departments within the  manufacturing, financial services, health care and consumer products industries,  as well as U.S. public sector agencies.
"It just seems like it's orchestrated in a controlled  manner, rather than a free-for-all, get-what-you-can kind of campaign," he  said.
Sarukkai said that what is helping the effectiveness of the  attack is that non-human system accounts are less likely to be protected by  multi-factor authentication or security policies, such as recurring  password reset requirements. "Once these accounts have been provisioned,  they're really sort of forgotten," he said. "I think these actors  have a pretty good understanding of the weakest link in Office 365 and in  general the security infrastructure -- almost like the hidden weakness."
 
	Posted by Scott Bekker on October 12, 2017