An Incorrect Synchronization vulnerability in the management daemon (mgd) of Juniper Networks Junos OS and Junos OS Evolved allows a network-based attacker with low privileges to cause a complete Denial-of-Service (DoS) of the management plane.
When NETCONF sessions are quickly established and disconnected, a locking issue causes mgd processes to hang in an unusable state. When the maximum number of mgd processes has been reached, no new logins are possible. This leads to the inability to manage the device and requires a power-cycle to recover.
This issue can be monitored by checking for mgd processes in lockf state in the output of 'show system processes extensive':
user@host> show system processes extensive | match mgd
<pid> root 20 0 501M 4640K lockf 1 0:01 0.00% mgd
If the system still can be accessed (either via the CLI or as root, which might still be possible as last resort as this won't invoke mgd), mgd processes in this state can be killed with 'request system process terminate <PID>' from the CLI or with 'kill -9 <PID>' from the shell.
This issue affects:
Junos OS:
* 23.4 versions before 23.4R2-S4,
* 24.2 versions before 24.2R2-S1,
* 24.4 versions before 24.4R1-S3, 24.4R2;
This issue does not affect Junos OS versions before 23.4R1;
Junos OS Evolved:
* 23.4 versions before 23.4R2-S5-EVO,
* 24.2 versions before 24.2R2-S1-EVO,
* 24.4 versions before 24.4R1-S3-EVO, 24.4R2-EVO.
This issue does not affect Junos OS Evolved versions before 23.4R1-EVO;
CVE-2026-21919 is a medium-severity Denial-of-Service vulnerability in Juniper Junos OS management daemon (mgd) affecting versions 23.4 through 24.4. Rapid NETCONF session establishment and disconnection trigger a locking issue causing mgd processes to hang, eventually blocking all management access and requiring device power-cycle recovery. While no public exploit exists, the vulnerability is easily triggered through normal network operations and impacts device manageability.
IMMEDIATE ACTIONS:
1. Identify all Juniper Junos OS devices running affected versions (23.4 before 23.4R2-S4, 24.2 before 24.2R2-S1, 24.4 before 24.4R1-S3/24.4R2)
2. Monitor mgd process status using 'show system processes extensive | match mgd' for lockf state processes
3. Implement NETCONF session rate limiting and connection pooling to reduce rapid session cycling
PATCHING GUIDANCE:
1. Upgrade to patched versions: 23.4R2-S4 or later, 24.2R2-S1 or later, 24.4R1-S3/24.4R2 or later
2. For Junos OS Evolved: upgrade to 23.4R2-S5-EVO, 24.2R2-S1-EVO, or 24.4R1-S3-EVO/24.4R2-EVO
3. Schedule maintenance windows for phased upgrades across network infrastructure
COMPENSATING CONTROLS (if patching delayed):
1. Implement connection rate limiting on NETCONF clients
2. Configure session timeout parameters to prevent connection accumulation
3. Deploy out-of-band management access (serial console, IPMI) for emergency recovery
4. Establish automated monitoring for mgd lockf states with alerting
5. Document power-cycle recovery procedures for emergency response
DETECTION RULES:
1. Monitor for repeated NETCONF connection/disconnection patterns from single source
2. Alert on mgd process count approaching system limits
3. Track mgd process state transitions to lockf state
4. Monitor failed login attempts coinciding with high mgd process counts
الإجراءات الفورية:
1. تحديد جميع أجهزة Juniper Junos OS التي تعمل بالإصدارات المتأثرة (23.4 قبل 23.4R2-S4، 24.2 قبل 24.2R2-S1، 24.4 قبل 24.4R1-S3/24.4R2)
2. مراقبة حالة عملية mgd باستخدام 'show system processes extensive | match mgd' للعمليات في حالة lockf
3. تنفيذ تحديد معدل جلسة NETCONF وتجميع الاتصالات لتقليل دورات الجلسات السريعة
إرشادات التصحيح:
1. الترقية إلى الإصدارات المصححة: 23.4R2-S4 أو أحدث، 24.2R2-S1 أو أحدث، 24.4R1-S3/24.4R2 أو أحدث
2. لـ Junos OS Evolved: الترقية إلى 23.4R2-S5-EVO أو 24.2R2-S1-EVO أو 24.4R1-S3-EVO/24.4R2-EVO
3. جدولة نوافذ الصيانة للترقيات المرحلية عبر البنية التحتية للشبكة
الضوابط البديلة (إذا تأخر التصحيح):
1. تنفيذ تحديد معدل الاتصال على عملاء NETCONF
2. تكوين معاملات انتهاء الجلسة لمنع تراكم الاتصالات
3. نشر وصول الإدارة خارج النطاق (وحدة التحكم التسلسلية، IPMI) لاسترجاع الطوارئ
4. إنشاء مراقبة آلية لحالات mgd lockf مع التنبيهات
5. توثيق إجراءات استرجاع إعادة التشغيل للاستجابة للطوارئ
قواعد الكشف:
1. مراقبة أنماط اتصال/قطع NETCONF المتكررة من مصدر واحد
2. التنبيه عند اقتراب عدد عمليات mgd من حدود النظام
3. تتبع انتقالات حالة عملية mgd إلى حالة lockf
4. مراقبة محاولات تسجيل الدخول الفاشلة المتزامنة مع أعداد عمليات mgd العالية