Skip to content
eastbaycyber

Cybersecurity Threat Digest — October 22, 2026

Threat digests 11 min read
EC
East Bay Cyber Editorial Team Updated
Week of 22 OCT 2026

TL;DR - PHP2Shell vulnerabilities reportedly have working exploits and active exploitation. - Internet-facing PHP, Plex, Cisco Expressway, and vulnerable WordPress systems need urgent review. - Hunt for web shells, command execution, persistence, and compromised third-party infrastructure today.

This cybersecurity threat digest October 22 2026 covers PHP2Shell exploitation, Plex attacks, critical vulnerabilities, and cybercrime disruption.

Top Stories#

Operation Serengeti III disrupts nearly 11,000 cybercrime networks

Law enforcement agencies disrupted nearly 11,000 cybercrime networks and arrested more than 1,400 people during Operation Serengeti III, according to BleepingComputer.

The operation targeted business email compromise, ransomware, and online scams across Africa. The arrests do not eliminate the underlying threat: criminal infrastructure, stolen credentials, affiliates, and replacement operators can reappear quickly.

Defensive takeaway: Continue treating payment-change requests, mailbox compromise, and credential theft as high-probability attack paths. Require out-of-band verification for financial transactions and privileged-account changes.

Chrome adds Gemini-powered scam detection and ClickFix warnings

Google Chrome is adding Gemini-powered protections to detect deceptive websites and warn users about suspicious commands associated with ClickFix attacks, BleepingComputer reports.

ClickFix campaigns typically persuade users to copy and paste commands into PowerShell, a terminal, or another execution interface. The attack relies on social engineering rather than a browser exploit, so browser warnings should complement, not replace, user training and endpoint controls.

Defensive takeaway:

  • Tell users never to paste commands into a terminal or PowerShell window because a website instructed them to do so.
  • Monitor endpoint telemetry for browser-spawned shells and scripting engines.
  • Block or alert on suspicious command interpreters launched by office applications, browsers, or document viewers.
  • Ensure browser security controls are centrally managed and updated.

Onamae Desktop Cloud compromise creates a large hosting risk

Attackers compromised Japan’s Onamae Desktop Cloud infrastructure and installed malware on as many as 26,739 customer servers, according to BleepingComputer.

This incident shows the blast radius of a managed hosting or infrastructure-provider compromise. Customers may have limited visibility into the initial intrusion, administrative actions, or malware deployment process.

Defensive takeaway: Organizations using affected services should obtain provider-specific incident details, verify server integrity independently, rotate credentials and tokens where appropriate, and review outbound connections and persistence mechanisms. Do not assume that provider-side remediation proves every customer system is clean.

Lazarus uses post-quantum cryptography for malware delivery

North Korean Lazarus operators are reportedly using a post-quantum cryptographic mechanism to protect malicious payload delivery across campaigns involving multiple malware families, BleepingComputer reports.

The immediate concern is not that post-quantum cryptography makes malware invulnerable. Attackers are adapting delivery and payload-protection techniques to complicate network inspection and analysis.

Defensive takeaway: Detection programs should focus on behavior and infrastructure as well as payload signatures. Hunt for unusual encrypted downloads, newly registered or low-reputation domains, abnormal outbound traffic, and execution chains involving scripts, loaders, and temporary directories.

CISA warns of active Plex2Shell attacks

CISA warned that attackers are chaining four Plex Media Server vulnerabilities in attacks referred to as Plex2Shell, according to BleepingComputer.

Internet-exposed media servers are attractive targets because they often run continuously, have broad access to local storage, and may be managed outside formal enterprise patching processes.

Defensive takeaway:

  • Patch Plex Media Server using the vendor-recommended release.
  • Remove unnecessary internet exposure.
  • Restrict administrative interfaces to trusted networks or VPN access.
  • Review process creation, authentication, and outbound network logs.
  • Treat unexpected media-server child processes or new scheduled tasks as potential compromise indicators.

Critical Vulnerabilities#

PHP2Shell: CVE-2026-109618 and CVE-2026-109619

PHP released emergency patches for two critical vulnerabilities tracked as PHP2Shell:

  • CVE-2026-109618: CVSS 10.0. Improper handling of out-of-memory errors may allow crafted requests to trigger memory corruption and execute arbitrary code.
  • CVE-2026-109619: CVSS 9.8. A use-after-free and stale global-state condition may be triggered through multiple strategies to achieve remote code execution.

The reported details and emergency patch release are documented by BleepingComputer. Additional technical analysis is available from the PHP security advisory for CVE-2026-109618, the PHP security advisory for CVE-2026-109619, Openwall, and Wiz.

KrebsOnSecurity reports that working exploits reportedly exist and that exploitation is occurring in the wild. Neither vulnerability is currently listed as added to CISA’s Known Exploited Vulnerabilities catalog in the supplied data, but the reported exploitation makes this an emergency remediation priority.

Analyst’s Take: Treat PHP2Shell as the first response priority for internet-facing environments because working exploits are reportedly circulating and exploitation is occurring in the wild. If patching must wait, isolate affected systems and restrict inbound access while checking for crashes, child processes, and suspicious outbound connections. The absence of a current CISA Known Exploited Vulnerabilities listing does not outweigh the reported exploitation.

Prioritize:

  1. Internet-facing PHP applications and APIs.
  2. Systems that cannot be rapidly patched or isolated.
  3. PHP-FPM, Apache, Nginx, and application stacks processing unauthenticated requests.
  4. Shared hosting and build images containing vulnerable PHP runtimes.
  5. Systems showing crashes, unexpected child processes, or suspicious outbound connections.

CVE-2026-76466: Cisco Expressway Series

CVE-2026-76466 carries a CVSS score of 9.8 and affects Cisco Expressway Series database services. The supplied description says an unauthenticated remote attacker with network access may execute operating-system commands as root by abusing persistent cookies and insufficient client-certificate validation.

Use the Cisco security advisory as the authoritative source for affected releases and remediation guidance.

Defensive actions:

  • Apply Cisco’s security update or mitigation.
  • Restrict access to Expressway services from untrusted networks.
  • Review persistent-cookie activity and unexpected administrative sessions.
  • Investigate root-level process creation and outbound connections from the appliance.
  • Rotate credentials and certificates if compromise is suspected.

CVE-2026-111832: WP Mailster

CVE-2026-111832 carries a CVSS score of 10.0 and affects WP Mailster versions through 1.9.1.0. The vulnerability permits unrestricted dangerous-file uploads that may allow an attacker to place a web shell on a WordPress server.

Details and remediation guidance are available from Patchstack.

Defensive actions:

  • Update or remove the affected plugin.
  • Inspect upload directories for PHP and other executable files.
  • Disable script execution in user-uploaded content where feasible.
  • Review WordPress administrator accounts and authentication logs.
  • Rotate credentials if unauthorized files or accounts are found.

CVE-2026-111834: Mang Board WP

CVE-2026-111834 carries a CVSS score of 9.9 and affects Mang Board WP versions through 2.4.4. The issue permits dangerous-file uploads that may be used to deploy a web shell.

See the Patchstack vulnerability record for available product-specific guidance.

Defensive actions:

  • Patch or remove Mang Board WP.
  • Search web roots and upload directories for newly created executable files.
  • Review web-server access logs for unusual multipart upload requests.
  • Check for new WordPress users, modified theme files, and unexpected cron jobs.
  • Rebuild compromised systems from trusted sources rather than deleting only the suspected shell.

What Defenders Should Do Today#

1. Patch and reduce exposure

Prioritize the following order:

  1. PHP systems exposed to the internet.
  2. Plex Media Server instances reachable from untrusted networks.
  3. Cisco Expressway systems requiring network access for remote administration or service use.
  4. WordPress sites running WP Mailster or Mang Board WP.
  5. Hosted infrastructure potentially affected by the Onamae Desktop Cloud compromise.

Where immediate patching is not possible, isolate the system, restrict inbound access, disable unnecessary services, and place compensating controls in front of the application.

2. Review logs for exploitation and post-compromise activity

Look for:

  • PHP worker crashes or repeated out-of-memory events.
  • Web requests immediately followed by shell or scripting-process creation.
  • New files in upload directories or web roots.
  • Unexpected PHP files with obfuscated code.
  • Web-server processes launching shells, download tools, or interpreters.
  • New administrator accounts, scheduled tasks, services, or startup entries.
  • Outbound connections to newly registered or low-reputation infrastructure.
  • Plex, Expressway, or hosting-provider administrative activity outside normal change windows.

3. Audit WordPress file uploads

For affected WordPress systems, inspect upload paths and web roots. A basic defensive review can identify recently modified PHP files:

find /var/www -type f \( -name '*.php' -o -name '*.phtml' -o -name '*.phar' \) \\
  -mtime -14 -printf '%TY-%Tm-%Td %TH:%TM %p\\n' 2>/dev/null

Treat this as triage, not proof of compromise. Compare results with known application files and deployment records. Preserve suspicious files for analysis before deleting them.

4. Hunt for suspicious web-server child processes

The following examples can help identify shells or download utilities launched by web services:

ps auxww | grep -E '[w]ww-data|[a]pache|[n]ginx|[p]hp-fpm'
pstree -ap 2>/dev/null | grep -E 'php|apache|nginx|sh|bash|curl|wget|python|perl'

For centralized telemetry, create detections for process trees such as:

php-fpm -> sh
apache2 -> bash
nginx -> curl
httpd -> wget
php -> python

Validate alerts against legitimate application behavior before containment.

5. Reinforce ClickFix defenses

User guidance should be direct:

  • Do not paste commands into PowerShell, Terminal, Command Prompt, or browser developer tools because a website requests it.
  • Do not approve unexpected security prompts or copy-and-paste “verification” instructions.
  • Report suspicious pages and preserve the URL for security review.
  • Use endpoint controls that log browser-to-shell process relationships.

6. Review third-party and managed-service dependencies

For the Onamae Desktop Cloud incident and other provider-side events:

  • Inventory systems hosted by the affected provider.
  • Confirm whether provider administrators accessed or modified customer systems.
  • Compare current files, packages, users, scheduled tasks, and network connections against a trusted baseline.
  • Rotate credentials, API keys, SSH keys, and tokens when exposure cannot be ruled out. A reputable password manager such as Try 1Password → can help teams manage unique replacement credentials securely.
  • Request incident timelines, indicators of compromise, and customer-specific impact details.
  • Document the provider’s notification and recovery commitments.

7. Strengthen business email compromise controls

Operation Serengeti III reinforces the need for resilient payment-fraud controls:

  • Require dual approval for payment and bank-account changes.
  • Verify requests through a known phone number or separate communication channel.
  • Alert on mailbox forwarding-rule changes.
  • Monitor suspicious OAuth grants and new inbox rules.
  • Protect privileged and finance accounts with phishing-resistant multifactor authentication where supported.

Technical Notes#

Identify internet-facing PHP services

Use asset inventory first. Where host access is available, identify the PHP runtime and linked service processes:

php -v
php --ini
ps -ef | grep -E '[p]hp-fpm|[a]pache2|[h]ttpd|[n]ginx'
ss -lntp

Record the PHP version, package source, service owner, application path, and exposure status. Do not rely solely on the version returned by a web application because containers, proxies, and multiple PHP runtimes may coexist.

Capture web requests around suspected exploitation

Preserve relevant logs before rotation:

mkdir -p /var/forensics/web-logs
cp -a /var/log/nginx /var/forensics/web-logs/ 2>/dev/null
cp -a /var/log/apache2 /var/forensics/web-logs/ 2>/dev/null
date -u > /var/forensics/web-logs/collection-time.txt

Search for requests associated with crashes, uploads, unusual methods, or command-like parameters:

grep -R -Ei \\
  'POST|multipart|upload|cmd=|exec|shell|base64|powershell|/bin/sh|/bin/bash' \\
  /var/log/nginx /var/log/apache2 2>/dev/null | tail -n 200

These strings are starting points only. Attackers can encode or otherwise modify payloads, and legitimate applications may use some of the same terms.

Contain a suspected web-shell compromise

If a server shows signs of exploitation:

  1. Remove it from public access while preserving necessary forensic connectivity.
  2. Capture volatile and persistent evidence according to incident-response procedures.
  3. Identify unauthorized files, accounts, processes, keys, and scheduled tasks.
  4. Rotate secrets from a trusted administrative system.
  5. Rebuild from known-good images or backups when integrity is uncertain.
  6. Patch before restoring service.
  7. Monitor closely for reinfection and credential abuse.

If a suspected exploit involves a race condition, the TOCTOU glossary explains the relevant security concept. After containment, document findings and lessons learned using this post-incident review guide.

Do not treat deleting a single suspicious PHP file as complete remediation. A successful web-shell deployment may have enabled credential theft, persistence, database access, or lateral movement.

CISA KEV additions this week#

CISA added 8 vulnerabilities to the Known Exploited Vulnerabilities catalog in the seven days to this digest. Source: CISA KEV catalog.

All KEV additions by date

Last verified: 2026-10-22

Disclaimer: This article may contain affiliate links. We earn a commission on qualifying purchases at no extra cost to you.