Alert Type

SRN - Software Release Notification
Low/NotificationSoftware Release Notification
Low/NotificationSoftware Release Notification

Product Affected

ACX, EX, MX, PTX, QFX, NFX, SRX, VRR, vMX, vSRX

Alert Description

Junos Software Service Release version 19.3R3-S2 is now available for download from the Junos software download site
Download Junos Software Service Release:
  1. Go to Junos Platforms - Download Software page
  2. Input your product in the "Find a Product" search box
  3. From the Type/OS drop-down menu, select Junos SR
  4. From the Version drop-down menu, select your version
  5. Click the Software tab
  6. Select the Install Package as need and follow the prompts

Solution

Junos Software service Release version 19.3R3-S2 is now available.
 

19.3R3-S2 - List of Fixed issues

PR NumberSynopsisCategory: EX4300 PFE
1545530Junos OS: EX4300: FPC crash upon receipt of specific frames on an interface without L2PT or dot1x configured (CVE-2021-0242)
Product-Group=junos		A vulnerability due to the improper handling of direct memory access (DMA) buffers on EX4300 switches on Juniper Networks Junos OS allows an attacker sending specific unicast frames to trigger a Denial of Service (DoS) condition by exhausting DMA buffers, causing the FPC to crash and the device to restart. The DMA buffer leak is seen when receiving these specific, valid unicast frames on an interface without Layer 2 Protocol Tunneling (L2PT) or dot1x configured. Interfaces with either L2PT or dot1x configured are not vulnerable to this issue. Please refer to https://kb.juniper.net/JSA11135 [juniper.net] for more information.
1548858The targeted-broadcast feature might not work after a reboot.
Product-Group=junos		On EX4300 Series platforms, the targeted-broadcast feature may not work after a reboot. It can be seen that no target-broadcast packets are received on the targeted-broadcast interface.
PR NumberSynopsisCategory: EX4300 Platform implementation
1502467The mge interface might still stay up while the far end of the link goes down.
Product-Group=junos		On EX4300-48MP platforms which support multi-rate gigabit ethernet (mge) interfaces, if a mge interface which is located within port range 24-47 is connected with some specific devices (e.g. a bypass module from DELL), when the far end of the link goes down, the mge interface might still stay up due to this issue. It will lead to traffic drop when sending traffic via the affected link.
1543890In every software upgrade host needs to get upgrade.
Product-Group=junosvae		Software upgrade will update host software also. Previously users needed to specified force-host option manually.
PR NumberSynopsisCategory: Marvell based EX PFE L3
1557229Traffic related to IRB interface might be dropped when mac-persistence-timer expires.
Product-Group=junos		On EX3400/EX4400/EX4300MP virtual chassis (VC) platforms, if the IRB interface is configured with members across master and backup VC, the new MAC address of the IRB interface might not be programmed in hardware after mac-persistence-timer expires. This might result in all traffic related to the IRB interface be dropped.
PR NumberSynopsisCategory: EX2300/3400 PFE
1543181The Slaac-Snoopd child process generates core file upon multiple switchovers on the Routing Engine.
Product-Group=junos		Slaac-Snoopd core in the child process of slaac-snoopd daemon is seen when Old Master transition to Master happens again. It means when RE has undergone 2 switchovers starting from Mastership role and again regaining the Mastership role after second switchover, slaac-snoopd core in the child process of slaac-snoopd daemon is observed. However it was observed that the core has no impact on base functionality of slaac-snoopd daemon.
1548159Classifier is not programmed in the hardware and error logs may be seen in syslog
Product-Group=junos		On EX platforms except EX4300, when configuring Q-in-Q with vlan-id-list, classifier is not programmed and error logs "Setting vlan id failed" may be seen. All the packet entering this interface will not classify the packets to designated egress queue, but to the default queue. If the default queue is in congestion, traffic with video or voice service may be impacted.
1556198Traffic might be dropped when a firewall filter rule uses the then VLAN action.
Product-Group=junos		If a firewall filter is configured with the action 'then vlan' on EX and QFX platforms, some of the traffic that matches the firewall filter might be dropped.
PR NumberSynopsisCategory: EX2300/3400 platform
1535106EX2300/EX3400 : RTC ERROR and SETTIME failed messages is seen
Product-Group=junos		On EX2300 and EX3400 series, you may observe RTC ERROR and SETTIME failed message sometimes without trigger.
PR NumberSynopsisCategory: QFX Multichassis Link Aggregrate
1562535MAC address entry issue might be observed after the MC-LAG interface.
Product-Group=junos		On all junos platforms with high scale setup (for ex: 40 mac per 3000 vlan), MAC address entry issue might be seen after MC-LAG interface failover/failback few times. Some MAC entry remains as DR after failover and these stale entries might cause service disruptions.
PR NumberSynopsisCategory: QFX PFE L2
1543169The dcpfe process might crash when IFD continuously attaching and detaching
Product-Group=junos		On QFX5K/QFX3600/EX4600 platforms, memory leak might happen when the IFD is continuously attaching and detaching. The dcpfe might crash if the device is running out of memory. Traffic loss might be seen during the dcpfe crash and restart.
1550918Traffic may be forwarded incorrectly on an interface having VXLAN enabled and "hold-time up xxx" statement configured
Product-Group=junos		If an interface is configured with "hold-time up xxx" statement and has VXLAN enabled, after interface flaps, traffic coming from this interface (such as ARP traffic) may be forwarded even it's not changed to the "up" state.
PR NumberSynopsisCategory: QFX L3 data-plane/forwarding
1568159The dcpfe process might crash when the size of the Local Bias Filter Bitmap string exceeds 256 characters.
Product-Group=junos		On QFX5K platforms with EVPN-VXLAN, the dcpfe process may crash when the size of the Local Bias Filter Bitmap string exceeds 256 characters.
1568533The untagged packets might not work on EX Series platforms.
Product-Group=junos		On EX/QFX5K platforms, if 'flexible-vlan-tagging', 'vlan-id-list' and 'native-vlan-id' are configured on the interface in SP style, the untagged packets that need to be egressed from this interface might be dropped by the peer device.
1573411The GRE egress traffic might not be forwarded between the different routing-instances
Product-Group=junos		In the GRE tunnel with the routing-instances scenario, if the next-hop of GRE tunnel destination is learned from the different routing-instance (e.g. the next-hop is learned via leaked route), it might be rejected to be installed into the routing table. Then, the egress GRE traffic will not be forwarded via the egress physical interface, the GRE traffic across the different routing-instances might not be forwarded.
PR NumberSynopsisCategory: Accounting Profile
1563641The configuration archive transfer-on-commit fails when running Junos OS Release 18.2R3-S6.5.
Product-Group=junos		RI name array is not initialised properly in PFED. Hence RI name is not populated properly in the cfg_file due to this the archival is not getting pushed during commit.
PR NumberSynopsisCategory: "agentd" software daemon
1455384Agentd memory might leak and crash when the RPD session closes without releasing memory.
Product-Group=junos		On PTX Series and MX Series devices, agentd memory might leak and crash because its memory leak happens when the internal communication is broken between agentd and rpd.
PR NumberSynopsisCategory: MX Layer 2 Forwarding Module
1548124The l2ald process might crash due to next-hop issue in the EVPN-MPLS.
Product-Group=junos		In the EVPN-MPLS scenario, if the flood next-hop info is updated with the stale Multicast Composite Netxthop (MCNH), the memory corruption might happen on l2ald (Layer 2 Address Learning Daemon) process, then l2ald might crash and the ingress L2-BUM-flooded traffic might be impacted.
PR NumberSynopsisCategory: Australia related infrastructure software
1501752Continuous l2ald and L2ALM log messages seen on nodes of chassis cluster of SRX5000
Product-Group=junos		On some JunOS SR releases continuous l2ald and L2ALM log messages are seen in chassis cluster setup of SRX5000 Series. The issue may cause a memory leak and drain the heap memory on card.
PR NumberSynopsisCategory: BBE Autoconfigured DVLAN related issues
1541796Subscriber might not come up on some dynamic VLAN ranges in a subscriber management environment.
Product-Group=junos		On the MX series platforms, if dynamic VLAN ranges are configured more than 32 on an interface, subscriber may only come up on the first 32 dynamic VLAN ranges of that interface.
PR NumberSynopsisCategory: BBE database related issues
1554539The subscriber sessions might be missed but stay in the authd after performing ISSU
Product-Group=junos		In the subscriber management environment, the subscriber session might be missed in bbe-smgd (BBE system management daemon) but stay in authd (authentication daemon) after ISSU (In-Service Software Upgrade) or VC (Virtual Chassis) global switchover. Because there is no accounting session stop sent from bbe-smgd, the radius server does not clean up these forgotten sessions and this leads to radius free pool limits which have an impact on the subscribers.
PR NumberSynopsisCategory: BBE interface related issues
1518543On the MX960 routers, the show interfaces redundancy rlt0 statement shows current status as primary down as FPC is still in the Ready state after rlt failover (restart FPC).
Product-Group=junos		Issue happens only with specific RLT interface configuration followed by FPC restart. In this PR, GRES was done when FPC was doing the interface cleanup after FPC restart. Amount of time taken for interface cleanup on FPC depends on scale. If GRES is done before FPC comes to clean state, backup RE may not be in sync with master. Which may cause interface states inconsistent after GRES. This is a negative scenario of testing.
PR NumberSynopsisCategory: BBE Layer-2 Bitstream Access
1551207The PPPoE subscribers might fail to login.
Product-Group=junos		In Broadband Network Gateway (BNG) scenario where Layer-2 Bitstream Access (L2BSA) and PPPoE subscribers are working on the same interface/VLAN, PPPoE subscribers login may get stuck in a pending state, if the IN-FLIGHT flag of L2BSA is not cleared after its Port-Up packet is rejected by RADIUS, and during this process, another Port-Up for the same access-loop arrives.
PR NumberSynopsisCategory: BBE routing
1556980The framed route installed for a demux Interface has no MAC address.
Product-Group=junos		On MX platforms with Broadband Edge(BBE) scenario, traffic sent to/transit via Framed-Route might be dropped, as there is no MAC associated with Framed-Route on the Demux Interface if "qualified-next-hop" is configured in dynamic-profile access route.
PR NumberSynopsisCategory: MIBs related to BBE
1535754Snmp mib walk for jnxSubscriber OIDs returns a general error.
Product-Group=junos		Snmp mib walk for jnxSubscriber OIDs returns General error
PR NumberSynopsisCategory: Border Gateway Protocol
1532414Transit labels for Layer 3 VPN routes are pushed momentarily to the MPLS.0 table.
Product-Group=junos		In L3VPN scenario with in-line RR (Route Reflector) which has an export policy with next-hop self to advertise all transit L3VPN routes to its clients and also uses route-target family, the undesired L3VPN routes may be pushed to FIB (Forwarding Information Base) after enabling/disabling route-target family on the RR or a new BGP peer joins a group with 'next-hop self'.
1541768The rpd process generates the core file at gp_rtarget_tsi_update,bgp_rtarget_flash_rt,bgp_rtarget_flash.
Product-Group=junos		If RTarget module tries to take an access of the active route which does not exist (since NextHop is not resolved), a reference is taken on the non-existent active route, and rpd crashes.
1545837If output-queue-priority expedited update-tokens is configured, rpd might crash might upon BGP flapping.
Product-Group=junos		On all Junos platforms with 'output-queue-priority expedited update-tokens' configured, rpd crash might be seen upon BGP flap.
1560827All the Layer 3 VPN route resets when a VRF is added or removed.
Product-Group=junos		After configuring a new VRF or performing GRES, VPN routes in bgp.l3vpn.0 table might be refreshed. There is unexpected packet loss when this issue happens.
PR NumberSynopsisCategory: Track PRs in BGP Flow Spec area & is part of BGP inside RPD.
1537085Junos OS: A specific BGP VPNv6 flowspec message causes routing protocol daemon (rpd) process to crash with a core (CVE-2021-0236)
Product-Group=junos		Due to an improper check for unusual or exceptional conditions in Juniper Networks Junos OS and Junos OS Evolved the Routing Protocol Daemon (RPD) service, upon receipt of a specific matching BGP packet meeting a specific term in the flowspec configuration, crashes and restarts causing a Denial of Service (DoS). Continued receipt and processing of this packet will create a sustained Denial of Service (DoS) condition. Refer to https://kb.juniper.net/JSA11131 [juniper.net] for more information.
PR NumberSynopsisCategory: bras licensing prs
1563975The enforce-strict-scale-limit-license configuration enforces subscriber license incorrectly in the ESSM subscriber scenario.
Product-Group=junos		In Extensible Subscriber Services Manager (ESSM) subscriber scenario, the "enforce-strict-scale-limit-license" configuration enforces subscriber license incorrectly, after high churn of ESSM subscribers login/logout, the subscribers could not able to login.
PR NumberSynopsisCategory: PFE SW evo-pfemand,packet-io on BRCM platforms running EVO
1545455The chip on FPC linecard might crash when the system reboots.
Product-Group=junos		On the FPCs with Broadcom chip, if the jinsightD (health-mon) is not disabled ("set system processes health-mon disable"), the FPC might crash during the system booting. Traffic loss is seen during the FPC crash and restart.
PR NumberSynopsisCategory: MX Platform SW - Environment Monitoring
1551760The LCM Peer Absent message might be seen.
Product-Group=junosvae		On all TVP platforms, a major alarm of "LCM Peer Absent" might be seen.
PR NumberSynopsisCategory: Virtual-chassis platform/chassisd infrastructure PRs for MX
1552588The VCP port is marked as administratively down on the wrong MX-VC member.
Product-Group=junos		On MX/EX/QFX platforms with Virtual Chassis (VC) scenario, some interfaces might be shutdown unexpectedly, which might cause traffic to be interrupted if there is an error generated on an FPC. The reason is that after an ASIC error, the IFD down messages is not sent to the local chassis master where the error was reported, instead, it will be sent to the master of the Virtual Chassis, so that another interface with the same Slot/PIC/Port number will be shutdown as well.
PR NumberSynopsisCategory: MX Platform SW - UI management
1537194The chassisd memory leak might cause traffic loss.
Product-Group=junos		On MX/PTX platforms with 18.1 or higher release, chassisd memory leak may be caused by configuration commit. When chassisd consumes ~3.4GB of memory it may crash, chassisd crash may cause GRES or/and FPC restart. If GRES is enabled, commits are being synchronized between REs, so backup RE chassisd may suffer from memory leak too.
PR NumberSynopsisCategory: Class of Service
1556103The explicit classifier or rewrite-rule might not work as expected for a logical interface if the wildcard configuration is also applied
Product-Group=junos		On all Junos platforms enabled with class of service (CoS), if there is wildcard classifier or rewrite-rule applied to the logical interfaces, the explicit classifier or rewrite-rule which is present for one of the logical interfaces might not work as expected due to this issue.
PR NumberSynopsisCategory: L2NG Access Security feature
1546166Junos OS: Receipt of specific DHCPv6 packet may cause jdhcpd to crash and restart (CVE-2021-0241)
Product-Group=junos		On Juniper Networks Junos OS platforms configured as DHCPv6 local server or DHCPv6 Relay Agent, Juniper Networks Dynamic Host Configuration Protocol Daemon (JDHCPD) process might crash with a core dump if a specific DHCPv6 packet is received, resulting in a restart of the daemon. Please refer to https://kb.juniper.net/JSA11168 [juniper.net] for more information.
PR NumberSynopsisCategory: MX-ELM l2ng stormcontrol
1552815The action-shutdown command of storm control does not work for the ARP broadcast packets.
Product-Group=junos		With knob 'action-shutdown' configured in storm control scenario, the interface will not go to shutdown state if ARP storm exceeds the configured limit. The excess packets will be dropped normally.
PR NumberSynopsisCategory: Device Configuration Daemon
1530935Backup Routing Engine or backup node may get stuck in bad status with improper backup-router configuration
Product-Group=junos		Redundant group 1+ may report Interface Monitor failure if backup router destination prefix is configured same as interface IP address.
1537491Junos OS: SRX1500, SRX4100, SRX4200, SRX4600, SRX5000 Series with SPC2/SPC3, vSRX Series: In a multi-tenant environment, a tenant host administrator may configure logical firewall isolation affecting other tenant networks (CVE-2021-0235)
Product-Group=junos		On SRX1500, SRX4100, SRX4200, SRX4600, SRX5000 Series with SPC2/SPC3, vSRX Series devices using tenant services on Juniper Networks Junos OS, due to incorrect permission scheme assigned to tenant system administrators, a tenant system administrator may inadvertently send their network traffic to one or more tenants while concurrently modifying the overall device system traffic management, affecting all tenants and the service provider. Further, a tenant may inadvertently receive traffic from another tenant. Refer to https://kb.juniper.net/JSA11130 [juniper.net] for more information.
1539991The logical interface might flap after the addition or deletion of the native VLAN configuration.
Product-Group=junos		On EX/QFX platforms, the unrelated logical interface on a physical interface would flap when adding or deleting native VLAN configuration on the physical interface.
PR NumberSynopsisCategory: dhcpd daemon
1542400DHCP discover packet might be dropped if the DHCP inform packet is received first.
Product-Group=junos		On all Junos and EVO platforms, when devices are configured as DHCP relay agent, if DHCP discover packet is received immediately after DHCP inform packet in the same session, the DHCP discover packet might be dropped. This issue will impact subscriber login and it can be recovered automatically.
PR NumberSynopsisCategory: Layer 3 forwarding, both v4+v6
1495563PFE_ERROR_FAIL_OPERATION: Failed to install in h/w, LOG: Err] dnx_nh_unilist_install: BCM L3 Egress create object failed for:Unilist nh 2097369 (0:Ok) nh 0
Product-Group=junos		During PFE initialization the unicast NH which are part of the unilist are miss programmed. In that condition if core link flap happens the error messages are observed as ASIC will not find the relevant unicast NH in the hardware as part of unilist.
PR NumberSynopsisCategory: Ethernet OAM (LFM)
1529209Junos OS: ethtraceroute Local Privilege Escalation vulnerabilities in SUID binaries (CVE-2021-0255)
Product-Group=junos		A local privilege escalation vulnerability in ethtraceroute Ethernet OAM utility of Juniper Networks Junos OS may allow a locally authenticated user with shell access to escalate privileges and write to the local filesystem as root. Please refer to https://kb.juniper.net/JSA11175 [juniper.net] for more information.
PR NumberSynopsisCategory: EVPN control plane issues
1513759With dynamic list next hop configured, a forwarding problem occurs after performing graceful switchover.
Product-Group=junos		On all Junos platforms with EVPN scenario, if the knob 'dynamic-list-nexthop' is configured, traffic loss might be observed in some EVPN instances after performing GRES. This is a timing issue and seen in a scaled setup.
1547275VLAN ID information is missed while installing the EVPN route from the BGP Type 2 Route after modifying a routing-instance from instance-type EVPN to instance-type virtual-switch.
Product-Group=junos		VLAN ID information might be missing while installing the EVPN route from the BGP Type 2 Route after modifying a routing-instance from "instance-type evpn" to "instance-type virtual-switch". As a result, the data traffic sent via these EVPN routes doesn't push vlan-id in the inner Ethernet Header. This might result in traffic getting discarded on the remote PE.
PR NumberSynopsisCategory: EVPN Layer-2 Forwarding
1512253Traffic might not get load balanced for multiple ESI/VTEP pairs with underlay AE between leaf and spine
Product-Group=junos		On all Junos platforms, under the EVPN-VxLAN scenario, inter-VLAN traffic load balancing might not work for multiple ESI/VTEP pairs with underlay AE between leaf and spines. This happens during certain overlay load balancing configuration changes or reboot conditions where remote VxLAN tunnels (RVTEPs) are not getting added to local VLANs.
1535515All the ARP reply packets toward some address are flooded across the entire fabric.
Product-Group=junos		In the EVPN-VXLAN scenario, if the spine has irb and the leaves don't have irb, and the leaves have multi-home interfaces, the ARP reply packets flooding across the entire fabric might be seen.
PR NumberSynopsisCategory: EX4400 PFE software
1552623"show pfe route summary hw" shows random high free and 'Used' column for 'IPv6 LPM(< 64)' routes
Product-Group=junos		For routes <= or > 64 bit mask, after route delete from hardware or during overflow handling of routes from host to LPM table, there was an issue with accounting whereby an incorrect large value showed up in the route summary calculation. Affected platform: QFX5200-32C-32Q and EX4400-48F.
PR NumberSynopsisCategory: EX4400 platform
1573889EX4400: Mgmt LEDs are not working as per expectations
Product-Group=junos		Below is the status of management LED when speed is set to: 1. 10m - Activity LED is not blinking when ping/traffic is runnig (instead it is remaining steady GREEN) 2. 100m - Activity LED is blinking without any ping/traffic.
PR NumberSynopsisCategory: Express PFE L2 fwding Features
1534340The dcpfe process might crash and cause FPC to restart due to the traffic burst.
Product-Group=junos		The dcpfe process might crash on the QFX10002-60C/PTX10002-60C platform and might lead to FPC restart causing traffic loss. This issue is seen during traffic bursts on the device.
1561084When configuring the static MAC and static ARP on the EVPN core aggregate interface, the underlay next-hop programming might not be updated in the Packet Forwarding Engine.
Product-Group=junos		After installing static MAC/ARP into the core underlay link and reverting the configuration, the Next Hop (NH) in PFE might still point to the configured MAC address. It might cause traffic blackholing towards the CE.
PR NumberSynopsisCategory: SRX4100/SRX4200 platform software
1547953On vSRX2.0, vSRX3.0, SRX1500, SRX4100, SRX4200, SRX4600 running chassis cluster in Junos OS Release 18.3 or later, multiple messages of "LCC: ch_cluster_lcc_set_context:564: failed to lock chassis_vmx mutex 11" are generated in the chassisd log file. These messages may recur after every few seconds and they do not have any impact on system operation.
Product-Group=junos		On vSRX2.0, vSRX3.0, SRX1500, SRX4100, SRX4200, SRX4600 running Chassis Cluster in Junos 18.3 or later, multiple messages of "LCC: ch_cluster_lcc_set_context:564: failed to lock chassis_vmx mutex 11" are generated in the chassisd log file. These messages may reoccur after every few seconds and they do not have any impact on system operation.
PR NumberSynopsisCategory: IDP on logical system
1561298The idpd process might stop when committing IDP configuration under logical systems and tenant systems during RGs failover.
Product-Group=junos		On SRX Series devices, if there are a considerable number of Logical-systems/Tenants configured. The idpd process might crash if the IDP-related configuration under Logical-systems/Tenants is changed and committed repeatedly during Redundancy Groups(RGs) failover. It is suggested not to modify and commit the IDP-related configuration in that situation.
PR NumberSynopsisCategory: jdhcpd daemon
1564434Junos OS: Receipt of malformed DHCPv6 packets causes jdhcpd to crash and restart. (CVE-2021-0240)
Product-Group=junos		On Juniper Networks Junos OS platforms configured as DHCPv6 local server or DHCPv6 Relay Agent, the Juniper Networks Dynamic Host Configuration Protocol Daemon (JDHCPD) process might crash if a malformed DHCPv6 packet is received, resulting in a restart of the daemon. The daemon automatically restarts without intervention, but continued receipt and processing of this packet will create a sustained Denial of Service (DoS) condition. Please refer to https://kb.juniper.net/JSA11168 [juniper.net] for more information.
PR NumberSynopsisCategory: jl2tpd daemon
1537772Junos OS: MX Series: Executing CLI command repetitively may cause the system to run out of disk space (CVE-2021-0238)
Product-Group=junos		When a MX Series is configured as a Broadband Network Gateway (BNG) based on Layer 2 Tunneling Protocol (L2TP), executing certain CLI command may cause the system to run out of disk space, excessive disk usage may cause other complications. Please refer to https://kb.juniper.net/JSA11133 [juniper.net] for more information
PR NumberSynopsisCategory: Flow Module
1467654TCP session cannot time out properly upon receiving the TCP RESET packet, and the session timeout does not change to two seconds.
Product-Group=junos		The TCP session cannot time out properly upon receiving the TCP RESET packet, and the session timeout value does not change to 2 seconds.
1541954The rst-invalidate-session command does not work if configured together with the no-sequence-check command.
Product-Group=junos		On SRX Series platforms, the "rst-invalidate-session" does not work if configured together with no-sequence-check. It might result in a TCP connection unestablished.
PR NumberSynopsisCategory: JSR Infrastructure
1484872JFlow/IPFix - tracking pr for fixing crash when sampling is more than 65535
Product-Group=junos		There was a crash when setting the sampling rate more than 65535, this is since fixed. set forwarding-options sampling instance s1 input rate 70000
PR NumberSynopsisCategory: Firewall Policy
1546120Traffic might be dropped unexpectedly when the url-category match condition is used on a security policy
Product-Group=junos		An issue was discovered within Unified Policies that affected the url-category match condition that can cause it to "over-match" and apply to more traffic than it should. This can result in the SRX dropping traffic that would otherwise be permitted
1558382On the SRX5000 line of devices, the secondary node might get stuck in performing ColdSync after a reboot, upgrade, or if ISSU is performed.
Product-Group=junos		On SRX5000 line of devices, the secondary node might get stuck in performing ColdSync after a reboot.
1576038Traffic loss might be seen when a big number of applications or addresses is referenced by one policy.
Product-Group=junos		On all SRX platforms, when a big number of applications or addresses is referenced by one policy (e.g. 3k applications) that causes IPC (Inter Process Communications between RE and PFE) fragmentation, policy out-of-sync might be seen. The issue results in the policy work incorrectly and traffic loss might be seen.
PR NumberSynopsisCategory: IPSEC/IKE VPN
1530684On all SRX Series devices using IPsec with NAT traversal, MTU size for the external interface might be changed after IPsec SA is re-established.
Product-Group=junos		On all SRX series devices using IPsec with NAT Traversal, MTU size might be changed to a lower value for the ike external interface after IPsec SA is re-established.
1545916The flowd process might stop during IPsec SA renegotiation on SRX5000 line of devices.
Product-Group=junos		On SRX5000 Series devices with IPsec VPN configured running on Junos OS 18.2R1 or above, during IPsec Security Association (SA) renegotiation, a timing issue that the VPN packets refer to an invalid key might occur, which results in the flowd process crash.
1550232Traffic that goes through policy-based IPsec tunnel might be dropped after RG0 failover.
Product-Group=junos		On branch SRX series devices in a chassis cluster, when policy-based IPSec VPN is configured and the IPSec SA's lifetime is about to expire in a few minutes, the traffic might be dropped in the VPN tunnel after an RG0 failover.
1564444A session might be closed when the session is created during the IPsec rekey.
Product-Group=junos		A session might be closed when the session is generated during IPsec rekey. It might cause the traffic drop on SRX platforms.
PR NumberSynopsisCategory: PFE infra to support jvision
1547698The SENSOR APP DWORD leak is observed during the period of churn for routes bound to the sensor group.
Product-Group=junos		SENSOR APP DWORD leak observed during the period of churn for routes bound to Sensor group. Sensor types that are affected are Segment Routing, Segment Routing-TE, LDP and RSVP LSPs. https://kb.juniper.net/TSB17912 [juniper.net]
PR NumberSynopsisCategory: Layer 2 Control Module
1561235The l2cpd process might generate a core file on reboot.
Product-Group=junos		When xSTP is used, the l2cpd core might be seen on reboot. This will be a one-time core and will not impact on functionality.
PR NumberSynopsisCategory: Layer2 forwarding on EX/NTF/PTX/QFX
1534796High rate of ARP or NS packets might be observed between a device that runs Junos OS and host when the device that runs Junos OS receives an ARP or NS packet on an interface in transition.
Product-Group=junos		On Junos device in EVPN scenario, if an interface is in transition status and an ARP request is received on the interface from a host, the Junos device may send out re-arp out towards the host, and the host responds to this re-arp and the Junos device sends another re-arp in response to this arp reply from host. This goes forever causing high rate of arp packets until the interface comes up. This issue is also applicable to ND/NS in IPv6 environment.
1551025The l2alm processes high CPU utilization might be observed in the EVPN-VXLAN environment.
Product-Group=junos		In the EVPN-VxLAN scenario, as part of fixing PR1535515, the mac-ip entry's aging timer is adjusted by plus 30 seconds. These changes expose an issue in ARP expiry handling and result in the l2alm process high CPU utilization. This issue may cause MAC learning issue even traffic loss.
1551631The IRB interface might not work after chassisd and l2ald reboot in EVPN scenario
Product-Group=junos		In EVPN scenario, if NSR is not used, the IRB interface might not work after chassisd and l2ald reboot, there is traffic impact due to this issue.
PR NumberSynopsisCategory: lldp sw on MX platform
1528856The l2cpd process might crash when removing LLDP on an aggregated Ethernet interface.
Product-Group=junos		On all Junos platforms, if Link Layer Discovery Protocol(LLDP) is enabled on 'interface all' and some AE interface at the same time, the Layer 2 Control Protocol process (l2cpd) might crash when lldp is removed from the AE interface. The l2cpd crash might affect all the protocols running under it (such as X-STP, LLDP, ERP, MVRP, etc.).
1538482DUT did not receive the LLDP packet from phone.
Product-Group=junos		On EX4300 platforms, the LLDP (Link Layer Discovery Protocol) neighborship with the VoIP (Voice over Internet Protocol) phones can't be established when LLDP is configured on the PoE (Power over Ethernet) enabled port on EX4300 and connects to the VoIP Phone.
PR NumberSynopsisCategory: MQTT protocol, Mosquitto Broker and Client API
1522265Junos OS: Receipt of specific packets could lead to Denial of Service in MQTT Server (CVE-2021-0229)
Product-Group=junos		An uncontrolled resource consumption vulnerability in Message Queue Telemetry Transport (MQTT) server of Juniper Networks Junos OS allows an attacker to cause MQTT server to crash and restart leading to a Denial of Service (DoS) by sending a stream of specific packets. Please refer to https://kb.juniper.net/JSA11124 [juniper.net] for more information.
PR NumberSynopsisCategory: Multicast Routing
1555518Sending multicast traffic to downstream receiver on Trio based Virtual Chassis platforms might fail.
Product-Group=junos		On Trio based Virtual Chassis (VC) platform, when there are multicast tunneled packets being received, which come into the Virtual Chassis Ports (VCP) and then pop out of the tunnel, if the VCP ports and the interfaces where multicast packets enter/leave the router are located on the same Packet Forwarding Engine (PFE), it might fail in sending multicast traffic to downstream receiver due to this issue.
PR NumberSynopsisCategory: Fabric Manager for MX
1535787All SFBs might go offline due to fabric failure and fabric self-ping probes performing the disable-pfe action.
Product-Group=junos		When an MPC is removed without being taken offline, and the chassisd process is not able to process this event on the primary Routing Engine due to additional primary-role switch, and later the MPC that is pulled out of the slot is re-inserted, many Switch Fabric Boards (SFBs) might be offline due to max_total_cell_usage overflow condition on the xfchip. The MX2020 platform is not exposed to such an event if it has SFB2 or if "set chassis fabric disable-grant-bypass" is configured.
PR NumberSynopsisCategory: OS IPv4/ARP/ICMPv4
1544398The RP expired timer on the backup Routing Engine is not the same as the primary Routing Engine if the aging-timer is configured.
Product-Group=junos		If aging-timer is configured on master RE for an IRB interface, the ARP timer configuration is not synced properly to backup RE for the IRB interface. It might cause ARP storm after RE switchover.
1547583An internal timer on the backup Routing Engine might cause an ARP storm upon GRES switchover on the new primary (old backup) Routing Engine.
Product-Group=junos		On all MX platforms with BNG (Broadband Network Gateway) scenario, an internal timer (re-ARP timer) on backup RE could cause an ARP storm upon GRES switchover since there are lots of arp timeout on the new master RE in 2 minutes. The re-ARP timer is one-tenth of the ARP aging timer (default ARP aging timer is 20 minutes, so 1/10 of 20 minutes is 2 minutes). The fix will automatically adjust the timer based on the scale and the configured aging time avoiding ARP storm on new master.
PR NumberSynopsisCategory: Kernel Composite Next Hop (composite / l3vpn) Infrastructure
1548545The kernel crash with core file might be seen if churn happens for a flood composite next hop.
Product-Group=junos		After continuous churn happens for a flood composite next hop, the kernel crash might be seen.
PR NumberSynopsisCategory: FreeBSD Kernel Infrastructure
1563647Memory corruption of any binary in /usr/bin/ or /usr/sbin/ may be triggered when a recovery snapshot is being copied to the OAM volume
Product-Group=junos		Memory corruption of a binary from /usr/bin/ or /usr/sbin/ directory can occur if such binary is invoked when a recovery snapshot creation is in progress. The exact symptoms will be different depending on the exact binary and JUNOS version - some programs will show an error, and some programs will crash every time it is executed. Such memory corruption will be persistent until the affected Routing Engine is restarted. Please refer to TSB17954 [juniper.net] (https://kb.juniper.net/TSB17954 [juniper.net]) for further details.
PR NumberSynopsisCategory: "ifstate" infrastructure
1484322The SNMP index in the Packet Forwarding Engine reports as 0, causing sFlow to report either IIF or OIF (not both) as 0 in the sFlow record data at the collector.
Product-Group=junos		The SNMP index for bundle interface might become zero in PFE after restarting the FPC. This could cause the sflow records to have either "input interface value" (IIF) or "output interface value" (OIF) as 0 value.
1545463Continuous rpd errors might be seen and new routes fails to be programmed by the rpd process.
Product-Group=junos		In case of high route churn in the network, all kuackmem (Kernel ACK mechanism) entries may be exhausted as a result of a rare timing issue and any new routes will fail to get programmed by rpd.
PR NumberSynopsisCategory: Kernel Multicast Infrastructure
1569957FPC might crash in a multicast scenario
Product-Group=junos		On EX/PTX platforms with AE bundle interface(s), FPC crash might be seen after GRES. It might be seen when the device is configured with a multicast scenario (composite Next-Hops for multicast routes are present over AE child links. After child member(s) of AE interface flapped, there is a sync issue between master Routing-Engine and backup Routing-Engine, which caused the FPC crash.
PR NumberSynopsisCategory: TCP/UDP transport layer
1552603The BGP session replication might fail to start after the session crashes on the backup Routing Engine.
Product-Group=junos		On certain Junos platforms with Dual-REs (platforms capable of installing Junos packages with name format as "junos*install"), BGP replication may fail to start under GRES/NSR setup after a crash on backup Routing Engine. NSR starts un-replicating the socket since backup Routing Engine is no longer present. Massive unreplicated request leads to memory buffer getting full with multiple BGP sessions (e.g., 20 BGP peers). Hence BGP unreplicated request returned with an error. Besides, the kernel is left with stale data. It does not allow the JSR (Juniper Socket Replication, BGP in this case) when backup RE comes up due to the stale data. BGP-NSR (Nonstop Routing) is broke under the conditions. Traffic outage will be observed after performing GRES.
PR NumberSynopsisCategory: OSPF routing protocol
1561207Duplicate LSP nexthop is shown on inet.0, inet.3 and mpls.0 route table when ospf traffic-engineering shortcuts and mpls bgp-igp-both-ribs are enabled.
Product-Group=junos		mpls.0 and inet.3 LDP routes showed duplicate RSVP LSP nexthops when "protocols mpls traffic-engineering bgp-igp-both-ribs" and "protocols ospf traffic-engineering shortcuts" were configured.
1561414Wrong SPF calculation might be observed for OSPF with ldp-synchronization hold-time configured after the interface flaps.
Product-Group=junos		On all Junos platforms with LDP protocol configured on an interface, set the interface type p2p in OSPF and configure ldp-synchronization with hold-time for the same interface, after flapping the interface, the wrong SPF calculation due to pointing to the old link might happen and this might cause a routing loop and traffic outage.
PR NumberSynopsisCategory: Express Chip L3 software
1533814On the QFX10002, the firewall logs are incorrectly populating from the Packet Forwarding Engine.
Product-Group=junos		When a multicast feed is received with TTL 1 on a QFX. There will be 2 copies of the packet sent to the host - one from the normal flow and another from the multicast module. This packet being sent from the multicast module had a sample class of 0 because of which it was getting logged in the firewall log. In order to resolve the issue engineering has modified the sample class of the multicast packet with ttl=1 so that it doesnt reflect in the firewall logs.
1540793The PFE might crash ans MPLS IPv6-tunneling scenario when the next hop changes.
Product-Group=junos		On PTX or QFX10K platforms, when the type of the next hop for a IPv6 route learned via MPLS tunneling is unicast, and if the next hop changes, memory is not freed. Once memory is consumed the PFE might crash.
1550632The Neighbor Solicitation might be dropped from the peer device.
Product-Group=junos		The Neighbor Solicitation (NS) might be dropped after the IPv6 binding is flushed from the peer side. The ping to IPv6 peer fails due to the NS message not reach RE. Since ping doesn?t work between the connected interface, any kind of traffic sent towards QFX might also not work. It has a traffic impact.
1569120QFX10K: Firewall log incorrectly populating from PFE for IPv6 traffic
Product-Group=junos		IPv6 packets with TTL 1 was getting logged in firewall without any firewall configuration. There will be no impact on the traffic but the logs will be generated in "show firewall log" for the same traffic.
PR NumberSynopsisCategory: vMX Platform Infrastructure related issue tracking
1526855On the MX150 router, IFDs stay up during vmhost halt or power-off.
Product-Group=junosvae		MX150: IFDs stay up during vmhost halt or power-off
1548422Traffic with jumbo frame may be discarded on the vMX platforms
Product-Group=junos		On the VMX platforms which are installed on ESXI 6.7 with vmxnet3 driver, traffic with jumbo frame (Packets with MTU more than 1500) may be discarded upon receiving.
PR NumberSynopsisCategory: PTP related issues.
1557758Packets corruption on 100G or 40G interface are configured with protocol PTP.
Product-Group=junos		On MX Platform with any of these linecards -MPC9E/JNP10K-LC2101/JNP10003-LC2103/MX204-MPC, Packets corruption might occur with enabling PTP(Protocol Time protocol) on 100G/40G interfaces mapped to Channelized MAC.
1561372PTP lock status gets stuck at the Acquiring state instead of the Phase Aligned state.
Product-Group=junos		In some cases the PTP slave port will stay in acquiring mode indefinitely, because the QFX is starting with EPOCH time Jan 1, 1970, and the time difference to the GM is too large for the servo algorithm. Fix is to start QFX at a time closer to the current NTP time.
PR NumberSynopsisCategory: Chassis mgmt for all QFX systems - chassis MIB, alarms, CLI
1481143On the QFX10000 device, the chassisd process might generate core files on the backup Routing Engine after commit for 200 seconds due to the following error message: CHASSISD_MAIN_THREAD_STALLED.
Product-Group=junos		Chassisd cores might happen on the backup RE after commit on QFX10K8/16s due to CHASSISD_MAIN_THREAD_STALLED for 200 seconds.
PR NumberSynopsisCategory: Interface related issues. Port up/down, stats, CMLC , serdes
1538340On the QFX5100-48T, interfaces are not created after 10g channel-speed is applied across the 48 to 53 ports.
Product-Group=junos		After channelizing port 48 through 53 and channel speed, the interfaces are down on QFX5100-48T platform. This issue causes interfaces are deleted and traffic might be dropped.
PR NumberSynopsisCategory: QFX platform optics related issues
1561181The tunable optics SFP+-10G-T-DWDM-ZR doesn't work on EX/QFX devices
Product-Group=junos		On EX4600/EX4650/QFX5110 devices with tunable optics SFP+-10G-T-DWDM-ZR used, the configured wavelength value does not take effect when connecting two EX/QFX across a mux (multiplexer) using tunable optics SFP+-10G-T-DWDM-ZR.
PR NumberSynopsisCategory: analyzer on QFX 5100,5200, 5110
1557274Traffic storm might be caused by the analyzer due to the flapping of the link.
Product-Group=junos		On all Junos platforms with port mirroring analyzers configured, if multiple paths for the Analyzer IP configured and default route flaps then a traffic storm might be observed due to mirroring of traffic on the wrong port and analyzer might not work as expected.
PR NumberSynopsisCategory: Filters
1558320On the QFX5000 devices, the firewall filter might fail.
Product-Group=junos		On QFX5K platforms, if per ifl-filtering on regular VLAN is configured with no match conditions then destination port matching condition may fail to match intended packets.
PR NumberSynopsisCategory: QFX L2 PFE
1535555The following Packet Forwarding Engine error message is observed in the BRCM-VIRTUAL: brcm_virtual_tunnel_port_create() ,489: Failed NW vxlan port token(45) hw-id(7026) status(Entry not found).
Product-Group=junos		On a QFX5110 or QFX5120, when the Type 5 tunnels are destroyed, sometime we can see error messages "brcm_virtual_tunnel_port_create() ,489:Failed NW vxlan port token(45) hw-id(7026) status(Entry not found)". There is no functionality impact due to this.
PR NumberSynopsisCategory: QFX L3 data-plane/forwarding
1484440IRB MAC is not programmed in hardware when the MAC persistence timer expires.
Product-Group=junos		On QFX5XXX/EX46XX virtual chassis platforms with GRES, if an IRB interface is configured with members across master and backup REs, and when mac-persistence-timer expires, the new MAC address of the IRB interface might not be programmed in hardware, which might result in failure on protocols and traffic.
1512175The DHCP traffic might not be forwarded correctly when DHCP sends unicast packets.
Product-Group=junos		On EX4600/QFX5K platforms, DHCP unicast packets are getting dropped in the device due to DHCP relay filters which are getting installed during the init time without any DHCP configuration.
1560161Few IPv6 ARP ND fails after loading the base configurations
Product-Group=junos		On QFX5k platforms, when configuring a VLAN ID for a VxLAN, recommendation is to use VLAN ID of 3 or higher. If VLAN ID of 2 is used, replicated broadcast, multicast, and unknown unicast (BUM) packets for these VxLANs might be untagged, which in turn might result in the packets being dropped by a device that receives the packets.
PR NumberSynopsisCategory: QFX EVPN / VxLAN
1524955Traffic loss might be observed on interfaces in a VXLAN environment.
Product-Group=junos		On the QFX5K/EX4600 series platforms with VXLAN setup, if changing the VLAN (VXLAN enabled) configuration under an interface stanza from service provider style to enterprise style in a single commit without deactivating/activating the corresponding VLAN configuration under "vlans" stanza, traffic loss may be observed on the interface after the change.
1555835Traffic might not passed due to the addition of the VLAN tag 2 while passing through the Virtual Chassis port.
Product-Group=junos		When ingress and egress interfaces are in different FPC on QFX5120VC with OVSDB vxlan, the VLAN tag 2 might be added automatically and the peer device drops it.
PR NumberSynopsisCategory: uboot & loader for DCG TOR and CB
1536799Software recovery or installation using the Bootable USB Flash Drive option might fail
Product-Group=junosvae		Using "Bootable USB Flash Drive" to recover/install software may fail on the platforms with releases starting from 19.3, after power cycle (off/on).
PR NumberSynopsisCategory: QFX VC Infrastructure
1548079On the QFX5100 Virtual Chassis, the backup Routing Engines clear the reporting alarm for a PEM failure intermittently for a missing power source.
Product-Group=junos		The PEM failure alarm for a missing power source on a QFX5100 VC is incorrectly being toggled on the Backup RE
PR NumberSynopsisCategory: RPD Interfaces related issues
1526481The following error message is observed during GRES if an IRB interface is configured without a profile: RPD_DYN_CFG_GET_PROF_NAME_FAILED.
Product-Group=junos		In DHCP subscriber scenario, if IRB interface is configured under dhcp-local-server without dynamic-profile, the DHCP process might be abnormal after GRES or restart rpd and cause DHCP subscribers unable to login.
PR NumberSynopsisCategory: KRT Queue issues within RPD
1539601The rpd memory leak might be observed on the backup Routing Engine due to the flapping of the link.
Product-Group=junos		On all Junos platforms with dual REs, rpd memory leak may be seen when an AE member interface flaps or immediate restart of master RE. The memory leak was observed be around 32 bytes per session, the leak is only seen when AE have more than 8 legs.
PR NumberSynopsisCategory: RPD policy options
1523891The policy configuration might be mismatched between the rpd and mgd process when deactivate policy-options prefix-list is involved in the configuration sequence.
Product-Group=junos		If "deactivate policy-options prefix-list" is involved in configuration sequence along with other policy configurations, the mgd process might not notify the rpd process about the policy configuration changes after committing the configuration. This can cause the policy configuration to be out-of-synchronization between policy database used by rpd and configuration database used by mgd. Due to the missing policy entries in policy database, later the rpd might crash when accessing these entries via CLI command like "show policy".
PR NumberSynopsisCategory: RPD route tables, resolver, routing instances, static routes
1482112The rpd process might crash when deactivating logical systems.
Product-Group=junos		On all Junos platforms running with logical systems, if the logical systems get deactivated either by manually restarting the rpd process or by the deletion of the logical system configurations, the rpd process might crash in a race condition. It is a timing issue.
1525363Traffic loss might occur during VRF route resolution over indirect next hop.
Product-Group=junos		On all Junos platforms with Layer 3 Virtual Private Network (L3VPN) and "chained-composite-nexthop" enabled (the knob is enabled by default on PTX platform and MX platform only supporting MPCs), if a route resolution of indirect nexthop happens, where the route with indirect nexthop has a label, for example, resolution happens for VRF routes using the same indirect nexthop and forwarding nexthop after committing VRF related configuration change, traffic loss might happen due to this issue as part of an optimization process.
PR NumberSynopsisCategory: SW PRs for SCBE3 related kernel drivers
1564539MX platforms with MX-SCBE3 may reboot continuously.
Product-Group=junos		A recent change in the kernel boot loader causes a system with MX-SCBE3 to experience the Routing Enginne's kernel memory corruption which causes the system to reboot continuously.
PR NumberSynopsisCategory: SW PRs for SCBE3 fabric
1553641The fabric errors are observed and the FPC processes might get offline with the SCBE3, MPC3E-NG, or MPC3E and MPC7 or MPC10 line card in the increased-bandwidth fabric mode.
Product-Group=junos		On MX240/MX480/MX960 platforms, with default "increased-bandwidth" fabric mode and SCBE3, if we have MPC3 or MPC3-NG exist on the system along with high bandwidth MPC, during high traffic situation or bursty traffic through the fabric towards MPC3/MPC3-NG. MX fabric might report unreachable destination condition and causes fabric healing to trigger in. This issue is exacerbated when having MPC7 or MPC10 line cards installed due to high fabric bandwidth that can be generated.
PR NumberSynopsisCategory: IPSEC functionality on M/MX/T ser
1540538The mspmand process leaks memory in relation to the MX Series telemetry reporting the following error message: RLIMIT_DATA exceed.
Product-Group=junos		On MX platforms with MS-MPC/MS-MIC service card installed, the card might run out of memory due to process mspmand memory leak, which may cause traffic interruption if adding and/or deleting of telemetry sensor. This is because these operations will trigger the memory allocation for decoding configuration change messages and will not release the memory at the end of processing.
PR NumberSynopsisCategory: SNMP Infrastructure (snmpd, mib2d)
1557384The mib2d process crashes and generates a core dump on backup RE
Product-Group=junos		The mibd process might crash on backup RE and generate core dumps. No major impact due to this issue.
PR NumberSynopsisCategory: MPC7/8/9 chassis issues
1481879Delay in disabling PFE might be seen on MX platforms with MPC7/8/9 and PTX series with PECHIP equipped FPCs inserted
Product-Group=junos		On MX240, MX480, MX960, MX2010, MX2020 platforms with MPC7/8/9 inserted, and PTX series with PECHIP equipped FPCs, if recovery code is triggered due to 'fatal' hardware error on the HMC, the HMC Fatal Error registers are dumped as part of the recovery code. This PR could cause delay in disabling PFE, which in turns delays traffic switch over to the redundant network interface.
PR NumberSynopsisCategory: Trio pfe qos software
1538960The following major error message might cause the Packet Forwarding Engine(s) to disable: XQ_CMERROR_SCHED_L3_PERR_ERR.
Product-Group=junos		On EX9200 platforms with EX9200-6QS, MX Series platforms with MPC2E/3E/5E(Q), and SRX5000 Series platforms with MPC3, the Packet Forwarding Engine on that FPC might be disabled due to a major alarm: "XQ_CMERROR_SCHED_L3_PERR_ERR". Without the fix, this major alarm triggered "disable-pfe" action. This defect could be seen if there is a parity error in the Layer 3 node static memory.
PR NumberSynopsisCategory: Trio pfe bridging, learning, stp, oam, irb software
1542211Trio-based FPC might crash when the underlying Layer 2 interface for ARP over IRB interface is changed from the physical interface to the LSI interface.
Product-Group=junos		This issue applies to a Trio-based FPC using the integrated routing and bridging (IRB) interface with a VPLS instance using the label-switched interface (LSI) to provide VPLS functionality. The FPC may restart unexpectedly. The restart is triggered when the underlying layer-2 interface for ARP over IRB interface changes from the physical interface to the LSI interface.
1560788The BUM frame might be duplicated on an aggregate device if the extended-port on the satellite device is an aggregated Ethernet interface.
Product-Group=junos		On the Fusion AD (Aggregate Device), the BUM frame might be duplicated if the Extended-port on the SD (Satellite Device) is an aggregate ethernet.
1564667[EVPN-MPLS] - NH DWORD LEAK observed for MAC+IP route churn in the case of EVPN-MPLS having IRB_ARP_NDP NH type
Product-Group=junos		NH DWORD LEAK observed for MAC+IP route churn in the case of EVPN-MPLS having IRB_ARP_NDP NH type
PR NumberSynopsisCategory: Trio pfe l3 forwarding issues
1562120The interface statistics might be reported incorrectly if a large scale of interfaces is configured
Product-Group=junos		On the MX/EX92xx/SRX5x00 Trio based platforms, the interface statistics might be reported incorrectly if a large scale of interfaces is configured. In the adaptive load balance (ALB) use case, the ALB uses these statistics to determine the load-balancing between the member links of the AE bundle, if wrong interface statistics are used, the traffic imbalance on the egress interface will be observed.
1569047Traffic loss might be observed when SCU accounting is configured and logical-systems is enabled
Product-Group=junos		On all Junos platforms with logical-systems enabled, when SCU (Source Class Usage) policy is configured on the main system while not on the logical-system, and if the logical-system comes up, the associated destination route in SCU policy might not be installed. As a result, traffic destined to or passing through this IP address might get dropped. Below is the example configuration for SCU accounting. set interfaces x/x/x unit 0 family inet accounting source-class-usage input set interfaces x/x/x unit 0 family inet accounting source-class-usage output
1569715The MPLS traffic passed through the back-to-back PE topology might match the wrong CoS queue.
Product-Group=junos		In a scenario involving back-to-back PE routers with CoS configurations where the LDP or RSVP LSP will be single hop LSP due to penultimate hop popping (PHP) and a real outer label is not imposed. In such a scenario, the EXP bits in the inner label (the label corresponding to L2circuit, L2VPN, L3VPN etc) may not be propagated based on the configured EXP rewrite rule to the downstream router. This will result in traffic being classified incorrectly on the egress PE and the forwarding of traffic might occur in an incorrect queue.
PR NumberSynopsisCategory: DDos Support on MX
1549656The kernel might crash if GRES is performed on either new iteration or after swapping the Routing Engine and restoring the HA configuration.
Product-Group=junos		On dual Routing Engine (RE) platforms with only 2 slots, the kernel crash may be seen on the backup RE after performing multiple Graceful Routing Engine Switchover (GRES).
1562474The DHCPv4 request packets might be wrongly dropped when DDoS attack occurs.
Product-Group=junos		On MX platform, T4000 platform and EX9200 platform, end-users or end-hosts might not get an IPv4 address from Dynamic Host Configuration Protocol (DHCP) server when Distributed Denial-of-Service (DDOS) attack is happened on DHCP rebind packets or renew packets. In the end, end-users or end-hosts could not access into network after lease time of the IPv4 address expired.
PR NumberSynopsisCategory: UI Infrastructure - mgd, DAX API, DDL/ODL
1553577The request system software validate on host command does not validate the correct configuration file.
Product-Group=junos		When using the "request system software validate on host username ", please use the latest os-package on remote host for it to properly use the configuration file sent from the host whose configuration file is being validated.
PR NumberSynopsisCategory: Antivirus UTM issue
1557278Stream buffer memory leak might happen when UTM is configured under unified policies.
Product-Group=junos		On all Junos platforms that support unified policies (layer 7 application), stream buffer memory leak might happen when Unified Threat Management (UTM) Antivirus Protection (AV)/ Antispam Filtering (AS)/Content Filtering (CF) is configured under unified policies. If the stream buffer memory is exhausted, traffic related to UTM will be affected.
PR NumberSynopsisCategory: For GPRS security features on highend SRX series
1559802SPU pause might be seen under GPRS tunneling protocol scenario.
Product-Group=junos		In SRX series devices, if mobile handover between SGSN/SGW more than once, and the last handover is GTPv1 to GTPv2 (3G -> LTE), then both cluster nodes may crash and cause a disruption in the traffic.
PR NumberSynopsisCategory: Virtual Router Redundancy Protocol
1526851When SRX Series devices receive proxy ARP requests on VRRP interfaces, SRX Series devices send ARP replies with the underlying interface MAC address.
Product-Group=junos		On SRX Series devices with VRRP scenario, the proxy ARP reply uses interface MAC address instead of VRRP MAC address if the VRRP is configured on some IFL and the proxy-arp is configured on same IFL, traffic loss may occur.
1550993An irb interface which has large unit value over 32767 cannot be active group for inheriting VRRP.
Product-Group=junos		irb (integrated routing and bridging) interface can be set large unit value. The unit number can be set until 65534 for an irb interface. However if an irb interface which has large unit value over 32767 the irb cannot be active group for inheriting VRRP.
1558560Junos device might send VRRP advertisement packets in the VRRP Init or Idle state before startup-silent-period timer expiry on the VRRP primary device with NSR disabled after GRES.
Product-Group=junos		If VRRP master device has dual Routing Engines (REs) and GRES enabled but nonstop-routing (NSR) disabled, after performing GRES, both REs will move to VRRP init then idle state and the new master RE will send VRRP advertisement packets in this stage before startup-silent-period timer expiry. Since the VRRP backup device can still receive the VRRP advertisement packets with higher priority, it will not transition to VRRP master state and hence cause the longer traffic downtime until the VRRP master device re-take the VRRP mastership after startup-silent-period timer expiry.
 
 

19.3R3-S2 - List of Known issues

PR NumberSynopsisCategory: QFX PFE L2
1574435On the QFX5000, software-forwarded VXLAN de-encapsulated packets have illegal length.
Product-Group=junos		On QFX5K switches, software forwarded VXLAN decapsulated packets (i.e. STP, DHCP etc) received on a VTEP interface might be forwarded with illegal length. During decapsulation the packet length might not be adjusted to the length on the inner payload and packet would get forwarded by adding trailer for the remaining length.
PR NumberSynopsisCategory: ACX L2 related features
1565642ACX5048: Entry for mac address from which no traffic is seen for mac age timer does not age out if there is active traffic destined for this mac
Product-Group=junos		As per the current code, ACX would not delete a mac address from the mac table there is- (a) traffic destined to the mac address or (b) traffic sourced from the mac address or (c) both Fix of this PR will allow ACX to only look at (b) traffic sourced from mac address before deleting the mac address entry from mac table. So, if there is no traffic sourced from the mac for an interval of mac aging timer, the mac would be deleted from the mac table at the end of mac aging timer with out taking into account the traffic destined to the mac address.
PR NumberSynopsisCategory: Control Plane and Infrastructure for the Junos Fusion Enterprise
1188254Junos Fusion Enterprise: LLDP might stop working if manually deactivated and reactivated
Product-Group=junos		On a Junos Fusion Enterprise, LLDP might stop working if it is reenabled after being manually disabled.
PR NumberSynopsisCategory: Bi Directional Forwarding Detection (BFD)
1474521The BFD session might get stuck in the Init or Down state after the BFD session flaps.
Product-Group=junos		On all Junos OS platforms, a BFD session configured with authentication may get stuck in init or down state after BFD session flap. This issue happens due to internal software logic error.
1518106The BFD sessions might flap continuously after disruptive switchover followed by GRES.
Product-Group=junos		Disruptive switchover (no GRES or NSR configured) can lead to stale PPM entries programmed on the new master Routing Engine. If both GRES and NSR are activated after disruptive switchover and then a GRES switchover is performed, BFD sessions might flap continuously.
PR NumberSynopsisCategory: Border Gateway Protocol
1456260Packet drop and CPU spike on Routing Engine might be seen in certain conditions if labeled-unicast protection is enabled for a CsC-VRF peer.
Product-Group=junos		On all Junos OS platforms in a carrier supporting carrier (CsC) scenario, when PE link protection (labeled-unicast protection) is enabled, after one of the redundant links between CsC-CE flap or RSVP-TE LSP re-route from the primary path happens, it might result in a slow convergence issue. Packet drops and CPU spike on the Routing Engine might be seen during this period.
1554569The BGP session neighbor shutdown configuration does not effect the non-established peer.
Product-Group=junos		BGP neighbor shutdown configuration "set protocols bgp group <*> neighbor xx.xx.xx.xx shutdown" does not take effect on non-established peer.
PR NumberSynopsisCategory: Layer 3 forwarding, both v4+v6
1380145On the ACX5448 router, latency is observed for the host-generated ICMP traffic.
Product-Group=junos		This ping latency behavior is expected for host generated ICMP traffic due to the design of PFE queue polling the packets from ASIC. lab@jtac-acx5448> ping 10.0.0.4 PING 10.0.0.4 (10.0.0.4): 56 data bytes 64 bytes from 10.0.0.4: icmp_seq=0 ttl=63 time=8.994 ms 64 bytes from 10.0.0.4: icmp_seq=1 ttl=63 time=49.370 ms 64 bytes from 10.0.0.4: icmp_seq=2 ttl=63 time=47.348 ms 64 bytes from 10.0.0.4: icmp_seq=3 ttl=63 time=45.411 ms <<< 64 bytes from 10.0.0.4: icmp_seq=4 ttl=63 time=106.449 ms <<< 64 bytes from 10.0.0.4: icmp_seq=5 ttl=63 time=79.697 ms <<< 64 bytes from 10.0.0.4: icmp_seq=6 ttl=63 time=37.489 ms <<< 64 bytes from 10.0.0.4: icmp_seq=7 ttl=63 time=31.436 ms << 64 bytes from 10.0.0.4: icmp_seq=8 ttl=63 time=35.460 ms << 64 bytes from 10.0.0.4: icmp_seq=9 ttl=63 time=77.198 ms << ^C --- 10.0.0.4 ping statistics --- 10 packets transmitted, 10 packets received, 0% packet loss round-trip min/avg/max/stddev = 8.994/51.885/106.449/26.824 ms
1502523dnx_bcm_mpls_label_stat_get: pkt stat counter get failed for Label 48033 messages whenever mpls stats cli is checked
Product-Group=junos		"show mpls lsp statistics" command is not supported in ACX710/ACX5448 routers.
PR NumberSynopsisCategory: Express pfe ddos protection feature
1547032OSPFv3 session might keep flapping and OSPFv3 hellos might be dropped in the host path.
Product-Group=junos		On QFX10008/QFX10016/QFX10002-36Q/QFX10002-72Q platforms, the OSPFv3 sessions might keep flapping and the hello packets maybe dropped in the host path. This might happen with high amount of control traffic with OSPFv3 protocol configured. This is because OSPFv3 hello packets are not proper classified going to the unclassified DDOS queue.
PR NumberSynopsisCategory: Express PFE FW Features
1420560On the PTXplatform, using firewall "family any" on lo0 interface, firewall counters do not increment.
Product-Group=junos		On PTX platforms, firewall counters shown a '0' count when using firewall "family any" on the lo0 interface. Please use "family inet" or "family inet6" instead.
PR NumberSynopsisCategory: GMIC2 platform driver issues
1382024The 10g-capable ports of the 2X10GE/20x1GE MACsec MIC may show the link status as up while the peer side is down
Product-Group=junos		The 10G speed-capable ports of the MIC-MACSEC-20G MIC may show the link status as up while the peer side may remain down
PR NumberSynopsisCategory: Firewall Policy
1454907Traffic might be dropped when policies are changed in SRX Series devices
Product-Group=junos		If a huge number of policies are configured on SRX Series devices and some policies are changed, the traffic that matches the changed policies might be dropped.
PR NumberSynopsisCategory: Layer 2 Control Module
1532992On the EX4300 device, complete traffic drop is observed when the MSTP edge port is configured over the access and QinQ ports.
Product-Group=junos		On all Junos and EVO platforms, in a QinQ environment, if xSTP is enabled on interface having logical interface with vlan-id-list configured then it will only run on those logical interfaces whose vlan-id range includes native-vlan-id configured and all others will in discarding state. This might lead to traffic drop.
PR NumberSynopsisCategory: vMX Data Plane Issues
1546803VMX/vFPC may reboot when firewall terms contain syslog/log.
Product-Group=junos		VMX /vfpc reboots may when firewall terms contain syslog/log. It might result in riot and vmxt coredumps.
PR NumberSynopsisCategory: Interface related issues. Port up/down, stats, CMLC , serdes
1560086PRBS (Pseudo Random Binary Sequence) test on the QFX5200 devices fails for 100GbE interfaces with the default settings.
Product-Group=junos		PRBS (Pseudo Random Binary Sequence) test on QFX5200 platform fails for 100G interfaces with default settings.
PR NumberSynopsisCategory: show route table commands, tracing, and syslog facilities
1574497PIM rib-group fails to be added in VRF.
Product-Group=junos		PIM rib-group failure to add in vrf - PIM: ribgroup vrf not usable in this context; all RIBs are not in instance
PR NumberSynopsisCategory: Resource Reservation Protocol
1576979With the local reversion on, there is a possibility of the transit router not informing the headend of RSVP disabled link when the link flaps more than once.
Product-Group=junos		With local reversion ON, there is a possibility of transit router not informing headend of RSVP disabled link when link is flapped more than once. Work around is to remove local-reversion configuration.
PR NumberSynopsisCategory: UI Infrastructure - mgd, DAX API, DDL/ODL
991081The configuration change on Backup RE cannot be synchronized while deleting inactivated interfaces/routing-instances from master
Product-Group=junos		In the dual Routing Engine scenario, the backup Routing Engine does not sync up the configuration change while deleting an inactivated interface/routing-instances from the master. So after the operation, the inactivated interface/routing-instances still exists on the backup Routing Engine. This issue has been addressed via PR/1586229.
1452136The mgd might crash when you use the replace pattern command.
Product-Group=junos		When you use the "replace pattern" command to replace the name in the apply-group, the mgd crashes.
 

Modification History

First publication date 2021-05-27

01/04/2022: Non-technical changes; formatting modified