The short version
- The compromise. An attacker entered OpsShield's update distribution infrastructure with legitimate credentials and used the cPGuard update channel to push malicious versions of the Linux web-server security suite. OpsShield says the activity began at 02:06 UTC on 2 November and that it stopped the malicious distribution within 12 hours.
- What the script took. Servers that updated during the window received a script that collected Linux password hashes, root database configuration, root shell history and root SSH files. The data was uploaded to OpsShield's licensing server through a compromised account and retrieved from there by the attacker.
- Scale. OpsShield told BleepingComputer that 2,946 servers were affected and had information copied, and that its own billing system database was copied too. The vendor says it has not established what data was collected from each individual server.
- The fix. OpsShield advises updating every cPGuard server to version 5.88.25 or later, which removes the backdoor and staged files, then resetting passwords and rotating database credentials, cloud secrets and SSH keys. OpsShield says it will also ship an update to remove previously installed support-access keys.
Timeline
| Date | Event |
|---|---|
| 2 Nov, 02:06 UTC | The attacker gains access to OpsShield's update distribution infrastructure and begins pushing compromised cPGuard versions. |
| 2 Nov | OpsShield says it detected the compromise and stopped the malicious distribution within 12 hours of the compromised versions being pushed. |
| 3 Nov | OpsShield publishes an alert describing the incident and the affected version range. |
How the malicious updates were delivered
OpsShield, the developer of cPGuard, says an unidentified attacker got into its update distribution infrastructure, which it describes as a CDN, on 2 November at 02:06 UTC. The vendor says the entry was made with compromised credentials and that it does not know yet how those credentials were obtained. From that access the attacker pushed a malicious script out through the normal cPGuard update channel.
The affected releases are versions 5.88.12 through 5.88.19. OpsShield says it detected the compromise and halted the malicious distribution within 12 hours of the attacker pushing the bad versions, but it could not confirm whether every server received and ran the script, because installations check for updates at different intervals. cPGuard is a Linux web-server security suite offering malware scanning, a firewall and attack protection, commonly run by hosting firms and administrators alongside control panels such as cPanel, DirectAdmin and Plesk. Namecheap offers it on virtual and dedicated servers.
What was collected and who is affected
The script pushed through the update channel targeted material that would help an intruder on the same machine: Linux password hashes, root configuration for MariaDB or MySQL, root shell history, and root SSH authorised keys and known-hosts files. It uploaded what it gathered to OpsShield's licensing server using a compromised account, and the attacker then pulled it back from there, according to the vendor. OpsShield says its investigation shows 32 files were retrieved per customer, but it has not established which data was taken from each server.
A company spokesperson told BleepingComputer that the incident affected nearly three thousand servers, giving a figure of 2,946, and that the billing system database was copied as well. OpsShield says it is blocking access, rebuilding the affected systems and preparing a public announcement. It advises customers to update and then invalidate any credential that sat on a machine running a compromised version, since installing the fixed release removes the backdoor but does not undo the theft.
What is not known yet
- How the attacker obtained the credentials that gave access to the update distribution infrastructure.
- Who carried out the attack and what they intended to do with the collected data.
- Exactly what data was taken from each of the affected servers, which OpsShield says it has not established.
- What the copied billing system database contained, including whether it held customer personal or payment data.
- How many servers ran the malicious script and how many received a fixed version before executing it.
What to do
If you run cPGuard
- Update every server to version 5.88.25 or later, which OpsShield says removes the backdoor and staged files.
- Reset operating-system passwords and rotate MariaDB/MySQL root credentials, cloud secrets and SSH keys that were present on any server running versions 5.88.12 to 5.88.19.
- Review accounts and keys on OpsShield's licensing server and apply the vendor's follow-up update that removes previously installed support-access keys.
If you provide hosting on cPGuard
- Identify which customer servers ran the affected versions during the 2 November window and tell those customers to rotate credentials.
- Treat any shell history or SSH known-hosts data on those machines as exposed and check for signs of interactive access.
For everyone else
- Treat automatic update channels as privileged access into your systems, because a vendor-side compromise runs with the rights the updater holds.
- Keep a record of which versions each host is running so you can identify machines that updated during a vendor incident.
Update log
- First published.
Sources
New incident reports by email
One email a week with the reports published since the last one.