On QFX5100 platforms, LR4 QSFP can take up to 15 min to come up after VC reboot

On QFX5100 platforms, LR4 QSFPs might take take longer to come up than others (up to 15 minutes). This is a intermittent occurrence.

QFX5000 platforms may display fpc0 Error requesting CMTFPC SET INTEGER, illegal setting 37 observed after upgrade

QFX5000 series platforms may show the following ERROR MESSAGE: fpc0 Error requesting CMTFPC SET INTEGER, illegal setting 37 observed after software upgrade.

The rpd might crash when doing RE switchover with NSR and logical-system configurations

When doing RE switchover with NSR (nonstop-routing) and logical-system configurations, rpd core might happen. This issue is platform independent. And it would cause traffic or service impact.

The rpd might crash during the policy configuration changes

The rpd might crash during the policy configuration changes

L2ALD restarts when changing "protocols" related configuration.

protocol configuration changes might trigger l2ald to restart unexpectedly.

Type 2 EVPN routes are missing after deactivating/activating protocol evpn

Type 2 EVPN routes are missing after deactivating/activating protocol EVPN.

[ACX] Commit error seen when configuring "mac-table-size" under bridge domain after upgrade to 15.1R7

On ACX1x00/ACX2x00/ACX4x00 running in 15-releases previous to 15.1R8, when configuring "mac-table-size" under bridge-domain, a wrong commit error appear not allowing the commit to pass.

The kernel crash might be observed when there is a firewall filter modification

In firewall scenario, when the apply-path statement is used to expand a prefix-list pointed to a defined path, if any configuration modification causes the prefix-list changes, in a rare condition, the kernel crash might happen. Traffic disruption might be seen if NSR (Nonstop active routing) is not used during the crash.

VPLS with "vlan-id-list" is not working properly in some releases when the link between a PE device and a CE device is an aggregated Ethernet interface with a single member link and child physical interface flap.

On a VPLS scenario, on which the CE-PE facing interface is an aggregate (ae) interface with a single child link, and VPLS VLAN-ID-list is configured: If the child link flaps, then the VLAN-ID-List does not get properly programmed in Hardware and the VPLS does not work properly.

Constant dcpfe process crash might be seen if using an unsupported GRE interface configuration

On QFX platform, if configuring GRE interface and its underlying tunnel source interface in two different routing instances, although this is a limitation, but the configuration can be committed and will cause constant dcpfe process crash.

MS-MPC might have performance degradation under scaled fragmented packets

On MX Series platforms with MS-MPC, it might have performance degradation if the MS-MPC receives scaled fragmented packets.

IPv6 ping might fail for spine node in EVPN scenario

In EVPN-IPv6 scenario, if Layer 2 ifl is brought down or flap under irb, the IPv6 ping might fail for spine node.

PFE on QFX5k may have DISCARD next-hop for overlay-bgp-lo0-ip in the VxLAN scenario

On QFX5000 Series platforms, the packet forwarding engine (PFE) may show DISCARD next-hop for overlay-bgp-lo0-ip.

The log of "RPD_KRT_Q_RETRIES: list nexthop ADD: No such file or directory" might be continuously shown after the rpd restart

When reading back next-hops from the kernel, the rpd could set an incorrect flag on the next-hop, which could potentially affect next-hop installation for composite next-hops.

The jpppd process might crash if the EPD value contains a format specifier

If the trace options for ppp-service are set to level all, the jpppd might crash when the EPD (Endpoint Discriminator) value which is used for LCP (Link Control Protocol) has a format specifier('%s').

ARP and ethernet-table entry in pointing to ae interface which state is down

In EVPN/VXLAN scenario, after changing the mtu of an ae interface which state is down, ARP and ethernet-table entry might point to ae interface.

The dcpfe core might be observed when doing "restart routing" or BGP neighbors flaps when EVPN-TYPE 5 routes are present

On QFX5000 Series platforms, if EVPN-TYPE 5 routes are present, when doing "restart routing" or a BGP session to a neighbor device flaps, the dcpfe coredump might be seen. It might cause service impact and traffic impact.

Link problems might occur with 100G-AOC on QFX series

On QFX Series switches except for QFX10000, the interfaces with 100G-AOC (Active Optic Cable with embedded transceivers) might be down. The description of affected AOC is QSFP28-100G-AOC (e.g., QSFP28-100G-AOC-1M, QSFP28-100G-AOC-3M).

Traffic being dropped when passing through MS-DPC to MPC

On MX series platform, when traffic passes through MS-DPC service card and then egresses the router through an AE interface on MPC, partial traffic loss might be seen due to a memory initializing issue.

The vmcore might be seen when routing changes are made on the peer spine in an EVPN VXLAN scenario

On all QFX-Series platforms with a standard EVPN VXLAN scenario, when the eBGP is used for underlay and iBGP for the overlay, the OSPF neighborship is established between two IRBs on the spines, upon exporting OSPF routes on any of the spines, as a result, a routing loop might occur on the spine once it receives the OSPF route update and causes the Routing Engine to have a kernel crash with a vmcore generated on the peer spine.

lcmd core and FPC restarted

In corner cases lcmd may crash

QFX : error message 'Failed with error (-7) while deleting the trunk 1 on the device 0'

On QFX series switch, error message below would be seen when adding or removing local-bias setting on SP style LAG interface. %PFE-3: fpc0 Failed with error (-7) while deleting the trunk 1 on the device 0

A few VPN tunnels do not forward traffic after RG1 failover.

A few VPN tunnels do not forward traffic after RG1 failover when traffic-selector is configured in the AutoVPN.

Junos OS: Firewall filter terms named "internal-1" and "internal-2" being ignored (CVE-2019-0036)

Junos OS: Firewall filter terms named "internal-1" and "internal-2" being ignored (CVE-2019-0036); Refer to https://kb.juniper.net/JSA10925 for more information.

IPv6 link-local address for virtual-gateway address is marked as duplicate in EVPN

In EVPN A-A mode, if the same link-local IPv6 address is configured on both IRB interfaces on the two PEs, DAD will mark one of the IRB interfaces as the duplicate.

CLI command "show system firmware" might provide unexpected output on some MX platform such as MX104

CLI command "show system firmware" might provide unexpected output on some MX platform such as MX104. The "current version" might be shown as ?? instead of the correct version number.

All FPC cards might restart after L3VPN routes churn

In L3VPN network with large-scale prefixes, if the peer PE is other vendor's router (e.g. Cisco) configured with "per-prefix label", all FPC cards might restart after L3VPN routes churn multiple times.

The process rpd may crash in BGP setup with NSR enabled.

In BGP setup where "prefix-limit" is configured with teardown parameter and damping is configured, if a flapping route is received, the related figure of merit doesn't reach the "suppress-limit" but it causes the device to teardown the respective peer since the "prefix-limit" to be exceeded, then the process rpd may crash because of the corruption on the NSR process to resync the database entries.

IPSEC tunnel can not be established because that the tunnel SA and rule are not installed in the PIC

On MX-Series platforms, when IPSEC is used in an interoperability scenario with other verndor`s devices (such as CISCO/HUAWEI) and peer device sends IPSEC tunnel establishment request using the port and protocol as Traffic/Flow distinguisher, the SA for the tunnel is not installed in the PIC, namely the impacted tunnels are up on the RE but these are not programmed in the PFE. It would cause that IPSEC tunnel can not be established and traffic failure.

The rpd might crash when LDP route with indirect next-hop is deleted

If Label Distribution Protocol (LDP) route with indirect next-hop exists (e.g. LDP egress-policy is used to advertise BGP route into LDP), the rpd might crash when the LDP route is deleted.

Junos OS: BGP packets can trigger rpd crash when BGP tracing is enabled. (CVE-2019-0019)

Junos OS: BGP packets can trigger rpd crash when BGP tracing is enabled. (CVE-2019-0019); Refer to https://kb.juniper.net/JSA10931 for more information.

Command "show | compare" output on global group changes lose the diff context after a rollback or 'load update' is performed

Command "show | compare" output displays the output in patch format. Changes in the global groups loses the context in the patch if a rollback or 'load update' is performed. The context loses until the commit is performed. This issue can be resolved by using fast-diff option.

The MPC might crash due to the CPU hogging by dfw thread

When a large amount of packets hit the firewall filter term action 'syslog' and a thread hogs CPU for more than 4 minutes, the MPC might crash.

The rpd crash due to memory corruption in EVPN

In Ethernet VPN (EVPN) active/active multi-homing scenario with MPLS encapsulation, toggling of multi-homed interface might cause memory corruption leading to rpd crash.

12th and 13th SFP-T ports on the NFX250 device are going down with the 18.4R1.3 image installation

On an NFX250, an SFP-T interface will not become active (UP) when it is plugged into either a ge-12/0/0 or a ge-13/0/0 interface.

The rpd might crash on a leaf node when handling the withdrawal of remote or local MAC address in an EVPN-VXLAN scenario

On all Junos OS platforms that are running Ethernet VPN (EVPN) with Virtual Extensible LAN (VXLAN) on the device, when handling the withdrawal of remote or local MAC address, it may cause stack corruption and may subsequently result in rpd crash on the leaf node.

Fabric performance drop on MPC7/8/9E and SFB2 based MX2000 platform

On MPC7/8/9E and SFB2 based MX2000 Series platforms, code change done by PR 1336446 fixing MPC7/8/9E fabric re-ordering issue with SFB causes fabric performance drop. The throughput might not reach the expected value in high volume traffic scenario.

The rpd may crash when RSVP bypass path flaps

In the MPLS-TE with RSVP scenario, when bypass path of RSVP goes down or comes up, CSPF re-calculation is triggered by this update event of state changing and rpd needs to try to remove/add this path from the TED (Traffic Engineering Database), and that may cause the rpd crash.

Ports 4, 5, 14, 15, 24 and 25 on PTX10008 and PTX10016 FPC LC1101 may fail to come up after a link flap or FPC reboot

Due to a regression in Junos 17.4R2-S3, the following ports on JNP10K-LC1101 on PTX10008/10016 might not come up after FPC reboot or port flap: 4, 5, 14, 15, 24, 25. JNP10K-LC1102 does not expose this behaviour.

Traffic over the AE getting policed to the policer applied on one of the IFLs when the vlan-id is changed

Sometimes in AE case, the CVLAN information is not available during filter bind. Subsequently, when child IFL is added multiple times, AE IFL finally gets the CVLAN information and filter entries were updated. But the AE IFL still does not have CVLAN info and entire IFD traffic is matchedin filter entry. This causes disruption to IFLs which does not have any filter. For solution, a new table is maintained to keep track of AE IFLs which does not have CVLAN info. When the tirgger comes from child IFL add, the table is checked & if present allowed to proceed for tcam reinstall for that IFL.

The PFE might get disabled unexpectedly due to a auto correctable non-fatal hardware error on PTX or QFX10002/QFX10008/QFX10016

On PTX or QFX10002/QFX10008/QFX10016, a auto correctable non-fatal hardware error on PE chip (which is ASIC on PTX1000, PTX10002, QFX10002, the third-generation FPC on PTX3000/PTX5000, and the Line card on PTX10008/PTX10016/QFX10008/QFX10016) is reported as 'FATAL' error and hence the related Packet Forwarding Engine (PFE) will get disabled. The code changes have been made to change the error category from 'FATAL' to 'INFO' to avoid the PFE to be disabled unexpectedly.

Traffic drop occurs when deleting MPLS family or disabling interface which has non-default EXP rewrite-rules

The non-VPN packets might be dropped when deleting family MPLS or disabling interface which has non-default EXP rewrite-rules. This is due to a cos-rewrite mask programming issue in Packet Forwarding Engine (PFE).

The rpd process might crash when "routing-options flow" configuration is removed

In BGP FlowSpec scenario, when configuration hierarchy "routing-options flow" is removed, the rpd process might crash due to a deleted data structure being called in code.

LLDP memory leak when ieee dcbx packet is received in auto-neg mode followed by another dcbx packet with none of ieee_dcbx tlvs present.

LLDP memory leak when ieee dcbx packet is received in auto-neg mode followed by another dcbx packet within same second with none of ieee_dcbx tlvs present which is leading to this second packet not being classified as ieee_dcbx.

Extended ports in JFE do not adjust MTU when VoIP is enabled

In Junos Fusion Enterprise (JFE) setups, Voice over IP (VoIP) enabled extended ports on satellite devices (SD) are set to the default Maximum Transmission Unit (MTU) of 1514 bytes. Due to this, the maximum data size is limited to 1468 bytes beyond which packets are dropped with MTU errors (when DF bit is set).

The unexpected AS prepending action for AS path might be seen after the no-attrset knob is configured or deleted with vrf-import/vrf-export configuration

If the independent AS domain (It is enabled with independent-domain knob, and attribute set messages are enabled by default) is configured for the virtual routing and forwarding (VRF) instance, the global autonomous system (AS) number in the master routing instance should be prepended to the AS path when the route prefix is imported into the VRF instance. And with no-attrset configured (which disable the attribute set messages), the global AS number in the master routing instance should not be prepended to the AS path. But the current implementation violate the above behavior when vrf-import/vrf-export policy is used in the VRF routing-instance and the no-attrset knob is configured or deleted.

VC Ports using DAC may not establish link on QFX5200

On QFX5200, when virtual-chassis is configured, if the QSFP configured as VCP is removed and then inserted, VC Ports using direct attach copper (DAC) may not establish link.

LDP route is not present in inet6.3 if IPv6 interface address is not configured

LDP (Label Distribution Protocol) checks for configured IPv6 interface address before it brings up the LDP IPv6 interface. If the interface is not configured with IPv6 interface address, LDP will not bring up the LDP IPv6 interface, hence LDP fails to install the route in inet6.3.

Dynamic routing protocol flapped with vmhost RE switchover on NG-RE

In NG-RE dual RE platform, some commands for RE switchover might lend process rpd to down/up, due to a delay that made the process chassisd failed to update its status of mastership promptly. As the mastership status the chassisd governed determines which action the process rpd need to take the next, so if RE is rebooted and the chassisd mastership state is RE-Master then, in that case, rpd clears all the kernel states; and if chassisd mastership state for that RE is RE-Backup, then rpd just quits silently and restarts again in backup mode without any kernel states being cleaned. So that rpd cleanup kernel states caused this issue.

The bbe-smgd process might have memory leak while running "show system subscriber-management route route-type <> routing-instance <>"

On MX platforms enabled with enhanced subscriber management, if the route-type and the routing-instance are used at the same time, there might be memory leak in the bbe-smgd process while running the command "show system subscriber-management route route-type <> routing-instance <>".

L2TP LAC may not tunnel static PPP subscriber when add/change interface events for related PPP logical interface comes in a short time interval

In some rare situations due to interface event message compressing a static pp0 logical interface may not be added in the jl2tpd database causing inability to tunnel a subscriber over that interface to L2TP LNS. In that case logs: Feb 15 12:55:44.508952 processIflChangeEvent: entry does not exist for pp0.1680 ifl index = 2561 Feb 15 12:56:02.629603 sendTunnelResp: NACK for sdbId = 4567611, iflIndex = 2561, sessionStartTime = 0, tunnelUid = UID null will be seen in the jl2tpd traceoptions.

The traffic might get blackholed in Junos Fusion Enterprise scenario with dual-AD

In Junos Fusion Enterprise scenario with dual-aggregate device (dual-AD), if the inter-chassis link (ICL) is over FPC EX9200-12QS , "show chassis satellite redundancy-group" might show "session pending". ICL link may not come up, so that redundancy with dual-AD cannot be achieved and the cascade port on a AD goes down, traffic diversion over ICL cannot occur and hence traffic will be blackholed.

Complete device outage might be seen when SPU VM core happens

On SRX high-end platforms, when SPU (Services Processing Unit) VM core happens on one node, this triggers bad kernel state on this node and complete device outage could be seen, which means all IGP and BGP adjacencies would be affected. The reason is that the SPU VM core causes primary PE to dump live VM core, which blocks jsrpd from committing RG (Redundancy Group) state updates to kernel to set PFE to primary state. And if no PFE is in the primary state traffic would be lost because the original primary SPU is reset in the process of booting up. It is a very rare timing issue.

CGNAT with MS-MPC card doesn't account for AP-P out of port errors or generate a syslog message when this condition is met.

Subscriber is requesting another block while there is no more ports available in that NAT address, you see APP errors increasing and this condition should be logged

there is no SNMP Trap message generated for jnxHardDiskMissing/jnxHardDiskFailed on Summit MX

There is no Trap message generated in case of SSD missing or Disk failed on Summit platform (MX10003 and PTX10002 .etc.), with the fix of the PR, JUNOS will generate OID as below accordingly. OID: 1.3.6.1.4.1.2636.4.1.15 jnxHardDiskMissing OID: 1.3.6.1.4.1.2636.4.1.14 jnxHardDiskFailed

COS table error can sometimes cause traffic outages and SNMP timeouts if the optic is plugged out and inserted back

COS table error can sometimes cause traffic outages and SNMP timeouts if the optic is plugged out and inserted back

Rebooting QFX5200-48Y using "request system reboot" doesn't take physical links offline immediately

After rebooting QFX5200-48Y using "request system reboot", the physical link doesn't become offline immediately, which might result in traffic loss.

A PPP session under negotiation might be terminated if another PPPoE client bearing the same session ID

In PPPoE (Point-to-Point Protocol over Ethernet) subscriber scenario, if a PPPoE client is under PPP (Point-to-Point Protocol) session negotiation while another PPPoE client bearing the same session ID sends LCP (Link Control Protocol) terminate request to the MX, the MX might terminate this PPP session. The issue results in failure of PPPoE client negotiation.

High CPU usage on fxpc process might be seen on ACX5K platform

On ACX5K platform, the fxpc process high CPU usage might be seen under rare condition if parity errors are detected in devices. It has no direct service/traffic impact. However since CPU utilization is high during this issue, there are some side-effects. Eg, it could impact time-sensitive features like BFD.

A memory leak in rpd might be seen if source packet routing is enabled for IS-IS protocol

If source packet routing or segment routing is enabled for IS-IS protocol, a memory leak might happen in the routing protocol process (rpd). The rpd will crash and restart once the rpd runs out of memory.

In a rare scenario with multicast extranet vpn, rpd can crash due to reference count of next hop becomes 0

In a rare scenario with multicast extranet vpn, rpd can crash due to reference count of next hop becomes 0

In the scenario where the MX and the peer device both try to bring an IPsec tunnel up, so both sides are acting as an initiator, if the peer side does not answer the MX ISAKMP requests the MX can bring the peer initiated tunnel down.

In the scenario where the MX and the peer device both try to bring an IPsec tunnel up, so both sides are acting as an initiator for separate tunnels, if the peer side does not answer the MX ISAKMP requests the MX can bring the peer initiated tunnel down.

jdhcpd daemon might crash during continuous stress test

jdhcpd daemon might crash during continuous stress test

While commiting huge configuration customer is seeing the error "error: mustd trace init failed"

"error: mustd trace init failed" during configuration commit

Traffic is dropped after FPC reboot with AE member links deactivated by remote device

On PTX series platforms with Link Aggregation Control Protocol (LACP) enabled, deactivating a remote Aggregate Ethernet (AE) member link will make the local member link move to LACP Detached state. The Detached link will be invalidated from the PFE AE-Forwarding table as expected. However, if the device is rebooted with this state, all the member links will be enabled in PFE AE-Forwarding table irrespective of LACP states and result in traffic drop.

Traffic not flowing through macsec interfaces when configured with an unknown cipher algorithm and change back

When an unsupported cipher was configured, macsec_cap_is_not_capable flag in pic structure will be set to TRUE. However, the issue was that the flag remained TRUE and was not being set to FALSE even though the cipher suite was reverted back from unsupported one to supported. As a result the check always returned TRUE. The fix now corrects this and updates the macsec_cap_is_not_capable flag to FALSE when it is a supported cipher suite.

When MBB for P2MP LSP fails, it is stuck in old path.

When MBB for P2MP LSP fails due to one or more sub LSPs did not come UP, the MBB is stopped and it is not re-tried until there is a change in the TED, during the subsequent re-optimization.