Cloud providers renting out bare-metal servers must make sure they scrub every last byte of writable storage on their boxes between deployments, infosec outfit Eclypsium has urged. Otherwise, malicious customers could stash malware in motherboard flash memory that activates when the next user of a machine powers it up.
The biz today emitted a case study , centering around IBM's SoftLayer, as evidence of how giving customers superuser-grade access to dedicated off-premises hardware can expose folks to attacks from firmware-level infections. Eclypsium, we should point out, just so happens to sell firmware security solutions.
The problem, explains Eclypsium, is that a miscreant could rent a bare-metal server instance from a provider, then exploit a firmware-level vulnerability, such as one inUEFI orBMC code, to gain persistence on the machine, and the ability to covertly monitor every subsequent use of that server. In other words, injecting spyware into the server's motherboard software, which runs below and out of sight of the host operating system and antivirus, so that future renters of the box will be secretly snooped on.
Ideally, this shouldn't be allowed to happen, that the firmware storage be completely wiped and restored to how it should be before a new customer is allocated a freed-up node, but apparently this sort of attack is possible.
The BMC in OpenBMC stands for 'Burglarize My Computer' – thanks to irritating security flawREAD MORE
To prove their theory, Eclypsium's team looked at SoftLayer, the IBM-owned cloud platform, and spotted some of Big Blue's bare-metal servers were vulnerable Supermicro firmware that could be leveraged to cause further mischief.
"Our goal was to acquire access to a device, make a small change, release it back to IBM for reclamation, and then reacquire the same device from a different user account to see if our changes survived the reclamation process," the Eclypsium team explained on Tuesday. "In our initial investigation, we identified a particular SoftLayer data center that seemed to have a small supply of a particular type of hardware."
Indeed, the researchers found they could acquire, in the Softlayer cloud, a server, modify the underlying BMC firmware, release the box for someone else to use, and then, by tracking the hardware serial number, wait to re-provision server to see if their firmware change was still intact. And it was. BMC is the Baseband Management Controller, the remote-controllable janitor of a server that has full access to the system.
This showed that after the server had been given up, wiped, and rented out to the next customer, and then obtained again, the underlying firmware had not been reset.
In the wild, this would potentially allow an attacker, or more likely a group of attackers, to stake out and observe specific cloud compute services to steal corporate data. In the shorter term, a scumbag could simply sabotage the firmware to brick servers en masse.
Eclypsium says it notified IBM of the vulnerability back in September, and claimed it had not heard back from Big Blue since December. The day before Eclypsium was due to go live with its findings this week, IBM published an advisory online , noting it has addressed the infosec upstart's concerns:
IBM has responded to this vulnerability by forcing all BMCs, including those that are already reporting up-to-date firmware, to be reflashed with factory firmware before they are re-provisioned to other customers. All logs in the BMC firmware are erased and all passwords to the BMC firmware are regenerated.
IBM has found no indication that this vulnerability has been exploited for malicious purposes.
Even though this has seemingly been resolved in SoftLayer, the underlying issues are much bigger than a single IBM service, and extend to all bare metal providers, it is claimed. "While our case study was based on IBM SoftLayer technology, this is not an issue limited to any one service provider," Eclypsium explained.
"The issues of firmware vulnerabilities and threats apply to all service providers, and we expect this area of research to remain very active based on the growing importance of cloud services."
Customers worried about these scenarios would be well advised to do a bit of homework on their bare-metal box provider, including running a test instance to check for firmware vulnerabilities and persistence, and, if possible, reflash every server's motherboard flash with a clean image and monitor for any changes during use. We suspect the big players do this anyway, but it's good to check.
Service providers, on the other hand, should develop procedures to reflash the firmware on machines before every bare metal redeployment. ®