GL.iNet confirms hackers used cloud flaw to infect routers and KVM devices
The vendor says the cloud hole is closed, but devices already infected stay infected until owners act, and the method of attack is still undisclosed.
By Mr Elite Published 7 min read
- Severity
- High
- Status
- Confirmed by the organisation
- Sources
- 1
- Last checked
The short version
- GL.iNet said in a 20 November advisory that an unauthorised party exploited its cloud remote management services to install malware on routers and KVM devices.
- The company says the attack path is blocked but that this does not remove malware from devices already infected.
- GL.iNet has not said how the attackers got in, how many customers were hit, or whether data was stolen.
- RemoteToHome Consulting reported suspicious activity on 152 of the 183 routers it analyses, though it cautions this may not reflect other customers.
- Stable firmware updates with further fixes were scheduled for 22 November 2026.
What happened
GL.iNet, a maker of home and travel routers and KVM devices, confirmed in a security advisory dated 20 November that an unauthorised party exploited a vulnerability in its cloud remote management services to install malicious software on customer devices, BleepingComputer reported on 22 November.
A KVM device lets a person see and control another computer remotely, as if sitting at its keyboard. GL.iNet's GoodCloud platform lets customers manage supported devices from afar, including controlling computers connected to its KVM products.
The company said it fixed the cloud vulnerability and blocked the access used in the incident. It also said devices can no longer be newly infected through this path, but that these measures do not remove malware from devices already compromised.
After the disclosure, customers and security researchers shared evidence of custom malware built for GL.iNet devices, BleepingComputer reported. They also shared root-level reverse shells that connected routers to the IP address 170.39.226[.]219. A reverse shell is a connection in which the infected device calls out to the attacker, who can then run commands on it. The reverse shells used the Linux nc utility to make those outbound connections.
In a separate forum report, the owner of a compromised GL-RM1 KVM shared network logs showing the device downloading a malicious script called ni.sh and further executable files from 66.245.219[.]252 on 19 November.
GL.iNet has pointed customers to a community-built scanner in its official support instructions. It has asked users who run the scanner to upload the anonymous report so developers can work out how many devices, and which types, are affected.
Why it happened
GL.iNet has confirmed only that a vulnerability in its cloud remote management services was exploited. It has not said how the attackers used it, how many customers were affected, or whether information was taken from compromised devices.
RemoteToHome Consulting, a research firm, identified a potential command injection flaw in the GoodCloud client. It involves a handler used for AstroWarp/SD-WAN functionality, which the researchers say passes cloud-supplied values into shell commands run with root privileges without proper validation.
GL.iNet has not confirmed that this flaw was the one used in the attacks. The link between the researchers' finding and the incident therefore remains unproven.
How this kind of failure works
Command injection happens when a program builds a system command out of text it receives and fails to check that text. If an attacker can control part of that text, they can add their own instructions, and the system runs them along with the legitimate ones.
Picture a form at a print shop that says 'print file: [name]'. If the clerk obeys whatever is written in the name box, someone who types a filename followed by 'and also unlock the till' gets both done. A device that runs such commands as root, the all-powerful account, hands over complete control.
Cloud management adds another layer of risk. A router trusts instructions from the vendor's cloud because that is how remote management works. If an attacker can reach the cloud service, or send values through it, the device may act on them without any login to the router itself.
Persistence is the other half of the problem. Malware that writes itself into configuration files can survive an ordinary firmware update, which is why fixing the cloud side does not clean up devices that were already compromised.
Who is affected
GL.iNet has not said how many customers are affected. Its advisory covers routers and KVM devices managed through its cloud services. For KVM products, it recommends firmware 1.10.3 or later for the RM10, RM1, RM1V2, RM1PE and RM4PE, 1.10.3 beta or later for the RM10RC, and 1.8.7 or later for the RMQ1.
To check a device, GL.iNet's initial instructions told customers to look for a process running as crond -i /tmp/.log, which it called an indicator of compromise. Forum members warn this check may miss devices infected with other backdoors or running reverse shells.
For KVM owners the stakes extend beyond the device. The malware may have allowed access to the computers the KVM controls, and GL.iNet is investigating whether any were compromised. RemoteToHome said it found suspicious activity on 152 of 183 routers it owns or manages, including 24 with root reverse shells, 47 with an older persistent backdoor and 81 showing attacker-supplied commands had run. It says its results may not match other customers' and the older backdoors may be unrelated to the November attacks.
How to defend against it
Tick each step as you check it. Your progress is saved on this device.
What to learn from this
The concept: Trust in the management plane
The management plane is the channel used to configure and control a device, as opposed to the traffic the device carries. Cloud-managed gear usually trusts that channel completely, so a flaw there can bypass every local protection, such as a strong router password.
The same pattern appears in remote monitoring tools, mobile device management and software update services. Whoever controls the channel controls every device that listens to it.
A vendor's cloud is part of your device's attack surface, so a fix on the vendor side does not clean devices that were already compromised.
Quick check
Sign in to earn 10 XP for a right answer
GL.iNet says it has fixed the cloud flaw. Why should owners of possibly infected devices still act?
GL.iNet said its measures stop new infections but do not remove malware from devices already infected, so owners must check and clean them.
Timeline
- 19 November 2026 A compromised GL-RM1 KVM was logged downloading ni.sh and other payloads from 66.245.219[.]252, according to a forum report.
- 20 November 2026 GL.iNet published a security advisory confirming exploitation of its cloud remote management services.
- After the advisory Customers and researchers shared evidence of custom malware and root reverse shells to 170.39.226[.]219.
- 22 November 2026 BleepingComputer reported the story; GL.iNet scheduled stable firmware updates for routers and KVM devices.
What is not known yet
- How the attackers exploited the cloud platform.
- How many customers and devices were infected.
- Whether data was stolen from compromised devices.
- Whether the command injection flaw found by RemoteToHome was used in the attacks.
- Whether computers connected to infected KVM devices were compromised.
- Who is behind the attacks.
Questions people ask
What did GL.iNet say happened?
In an advisory dated 20 November, GL.iNet said an unauthorised party exploited a vulnerability in its cloud remote management services to access devices and install malicious software. The company says it has fixed the flaw and blocked the access used.
Does the cloud fix remove the malware?
No. GL.iNet said the measures stop new infections through this path but do not remove malware from devices already infected. Owners need to check their devices and update the firmware.
How can I tell if my GL.iNet device is infected?
GL.iNet's initial guidance was to look for a process running as crond -i /tmp/.log. Forum members say this may miss some infections, so the company also links to a community scanner, whose developers say it cannot catch everything.
Is a normal firmware update enough to clean an infected device?
Possibly not. The scanner's developers warn that malicious files in configuration directories could survive an ordinary upgrade, and that a firmware reinstall without keeping existing settings may be needed.
Sources
- BleepingComputer: Hackers exploit GL.iNet cloud flaw to infect routers, KVM devices