Alert Type

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

Product Affected

ACX, PTX, and QFX Series running Junos Evolved Software

Alert Description

Junos Software Service Release version 20.4R3-S2-EVO 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 20.4R3-S2-EVO is now available.

20.4R3-S2-EVO - List of Fixed issues

PR Number Synopsis Category: Yocto issues
1608006 Junos OS Evolved: Multiple vulnerabilities in cURL resolved
Product-Group=evo 		Multiple vulnerabilities have been resolved in Juniper Networks Junos OS Evolved by updating cURL third party software. Refer to https://kb.juniper.net/JSA11289 [juniper.net] for more information.
PR Number Synopsis Category: "agentd" software daemon
1600412 The gNMI Telemetry might stop working after RE switchover
Product-Group=evo 		In a dual RE (Routing Engine) system with GRES (Graceful Routing Engine Switchover) NSR (Nonstop Active Routing) enabled telemetry might stop working after RE switchover.
PR Number Synopsis Category: Border Gateway Protocol
1620463 The rpd may crash and restart when NSR is enabled
Product-Group=evo 		On all Junos with NSR (nonstop routing) enabled the rpd crash and restart may occur when RPKI (Resource Public Key Infrastructure) records are being replicated between the primary and backup RE (Routing Engine) and some of the records are withdrawn over the RPKI session.
1626367 Time delay to export prefixes to BGP neighbors might occur post applying peer-specific BGP export policies
Product-Group=evo 		On all Junos and EVO Platforms, when BGP export policies were changed from deny all to the peer-specific export policies, it might take several hours for the RPD/BGP to finish the export evaluation.
1626756 Multipath route with List-NH which has Indirect-NH as members fails into BGP-LU
Product-Group=evo 		"RT-multipath route with List-Next-hop which has Indirect-Next-hop as members" is unable to advertise into BGP-labeled-unicast address family. This scenario could happen in Carrier's Carrier (CsC) VRF which is also the reported case.
1630220 The BGP ECMP might not work and multipath route wont be created
Product-Group=evo 		On all platforms, if BGP multipath is configured, if an active path that is not eligible for multipath and later gets deleted, the device might not calculate multipath for the other routes. This causes the ECMP (Equal Cost Multi Path) feature to fail and thereby causes route learning to fail.
PR Number Synopsis Category: PFE L2 forwarding features on BT based platforms
1601915 MVRP enabled trunk ports might go into blocked/designated state on the PTX EVO platforms
Product-Group=evo 		On the PTX EVO platforms, MVRP enabled trunk ports with xSTP configured might go into blocked/designated state if the peer connects to the interface has no VLANs configured in its trunk port.
PR Number Synopsis Category: Express BT PFE L3 Features
1621037 Incorrect sensor modeling/mapping in the Telemetry streaming scenario
Product-Group=evo 		On PTX platforms running Junos OS Evolved, the interface sensor will export UDP data with either sensor ID 158 or 58 in the Telemetry streaming scenario. This is the incorrect sensor modeling/mapping.
1630215 Indirect next-hop (INH) Version ID higher than 255 might cause INH NH FRR Session moved down state and dropping transit traffic
Product-Group=evo 		This issue applies to a Junos Evolved platform configured with the BGP Multipath. Every indirect NH session state change with a Version ID of 256 or higher will be ignored at the PFE. If the Session is in DOWN state with Version ID 255, it will remain DOWN, if in UP state it will remain UP. Only if all Indirect NH sessions of a BGP Multipath ECMP path are in DOWN state, traffic for those prefixes will be dropped.
PR Number Synopsis Category: Firewall related development
1639391 PFE corruption might be seen due to the unsupported configuration of BGP flowspec
Product-Group=evo 		On EVO platforms for the 20.4R3 release, the unsupported configuration of BGP flow spec "interface-group exclude" might lead to some errors and PFE corruption which did not permit filter bind.
PR Number Synopsis Category: Issues related to evo operations - libevo infra, typeinfo ..
1616114 Master RE0 reloaded unexpectedly and new-master RE1 doesn't bring up ISIS/LDP adjacencies
Product-Group=evo 		On Junos Evolved platforms with dual RE, in very rare cases, while on bootup, the rpd on backup RE might not start up due to a software defect. If later any RE switchover happens, because of no rpd running on the new master RE, complete service impact will be seen.
PR Number Synopsis Category: software upgrade infra issues
1569294 zeroize doesn't delete snapshot images from the backup HDD
Product-Group=evo 		request system zeroize command wont cleanup file-system on backup disk.
PR Number Synopsis Category: System Management daemon and related issues
1606839 Around 500ms to 800ms of traffic loss might be seen with one of AE member links of p2mp LSP branches is down on Junos Evolved PTX10008 platform
Product-Group=evo 		On Junos Evolved PTX10008 platform in FMBB (Fast Make-Before-Break) scenario, around 500ms to 800ms of traffic loss might be seen if the member link of AE bundle of the p2mp LSP branches is down.
PR Number Synopsis Category: Lacp related problems and issues.
1636093 Some daemons might get stuck when snmpd is at 100% CPU utilization
Product-Group=evo 		On all Junos and EVO platforms, when snmpd is at 100% CPU utilization and some daemons are trying to raise an SNMP trap, these daemons might get stuck which results in unbaling to handle any incoming corresponding packets and unbaling to response to any CLI commands.
PR Number Synopsis Category: EVPN control plane issues
1629953 Removing knob "es-label-oldstyle" does not take effect if it is the only knob configured under the protocol EVPN
Product-Group=evo 		On MX, QFX, and EX series platforms with EVPN enabled, removing knob "es-label-oldstyle" does not take effect if it is the only knob configured under the protocol EVPN.
PR Number Synopsis Category: EVPN Layer-2 Forwarding
1629426 The l2ald crash might be seen after performing restart routing on EVPN PE
Product-Group=evo 		On all Junos and Evo platforms, when restart routing performed on EVPN PE in the network, l2ald crash might be seen on EVPN PE where restart routing was done. Furthermore, its peering PEs may see l2ald crash due to protocol flap following to restart routing. In both case, l2ald crash would be seen in a few minutes (~ 10 min) after restart routing was done.
PR Number Synopsis Category: ISIS routing protocol
1631738 The rpd might crash after clearing isis database
Product-Group=evo 		When ISIS database is cleaned, rpd crash might be observed.
1633858 IS-IS database may not be synchronized in some multiple areas scenario
Product-Group=evo 		On all platforms with IS-IS multiple areas scenario, if the knob "flood-group" is enabled, IS-IS Databases may not get synchronized between areas after clearing the IS-IS DB or making the DB change in any other way. This is because when link-state packet (LSP) is fragmented, only the first packet has the area ID list (for flood-group matching), while the rest of the fragmented LSPs do not have that list, which will result in these packets not being flooded, so that ISIS will not work properly. Note: Flood Group is a feature of IS-IS, used to limit link-state packet data unit (PDU) flooding over IS-IS interfaces. When a link-state packet (LSP) that is not self-originated will be flooded only through the interface belonging to the flood group that has the configured area ID in the LSP. This helps minimize the routes and topology information, thus ensuring optimal convergence.
PR Number Synopsis Category: Layer 2 Control Module
1602588 Junos OS and Junos OS Evolved: An l2cpd memory leak can occur when specific LLDP packets are received leading to a DoS (CVE-2022-22172)
Product-Group=evo 		A Missing Release of Memory after Effective Lifetime vulnerability in the Layer-2 control protocols daemon (l2cpd) of Juniper Networks Junos OS and Junos OS Evolved allows an unauthenticated adjacent attacker to cause a memory leak. Continued exploitation can lead to memory exhaustion and thereby a Denial of Service (DoS). Refer to https://kb.juniper.net/JSA11278 [juniper.net] for more information.
PR Number Synopsis Category: Layer2 forwarding on EX/NTF/PTX/QFX
1626714 Broadcast traffic might not be forwarded to LT interface in VPLS routing instance after LT interface is deleted then added back
Product-Group=evo 		On MX platforms, when Logical Tunnels (LT) interface is used for VPLS, VPLS broadcast traffic might not be forwarded to LT interface properly after deleting the LT interface adding it back, which might cause LT interface missing from VPLS flood topology and eventually affects VPLS communication.
PR Number Synopsis Category: Issues related to Junos licensing infrastructure
1595409 [PTX10008 EVO] Application error alarms and trace-writer core are generated due to defunct rcp zombie
Product-Group=evo 		Application error alarms and trace-writer core are generated on PTX10008 EVO due to defunct rcp zombie.
1628733 PTX10008 EVO : license installation fails with "validation hook evaluation failed" commit error
Product-Group=evo 		On PTX10008 EVO, license installation succeeds with "request system license add terminal" CLI command, but installing the same license fails with "validation hook evaluation failed" error while performing commit.
PR Number Synopsis Category: lldp sw on MX platform
1617151 L2 cpd Memory leak may lead to l2cpd process crash
Product-Group=evo 		On all Junos and Evo platforms, there is a one-shot timer created for LLDP (Link Layer Discovery Protocol) regardless of whether LLDP is configured or not, which may not get freed before creating the new one-shot timer because of which there is 160 bytes of leak every minute. This gradual memory leak in l2cpd may lead to l2cpd process crash. This may impact traffic only if protocols other than LLDP (example xSTP) are running.
PR Number Synopsis Category: Port-based link layer security services and protocols that a
1624524 MACsec session might flap if multiple logical interfaces are created on single physical interface
Product-Group=evo 		On all Junos and Evo platforms with MACsec enabled, when multiple ifls (logical interfaces) are created on single ifd (physical interface), deleting one/few ifls might flap MACsec session and traffic drop might be seen during the issue.
PR Number Synopsis Category: Multiprotocol Label Switching
1615001 The RPD crash may happen due to refcount leak in routing table metrics
Product-Group=evo 		On all Junos and Junos OS Evolved platforms with BGP-LS configured, addition of routes via TED_LS module might cause refcount leak and could lead to RPD crash on the device.
PR Number Synopsis Category: Protocol Independant Multicast
1630144 The multicast forwarding cache might not get updated after deactivating the scope-policy configuration
Product-Group=evo 		On all Junos and EVO platforms with multicast setup, the multicast forwarding cache might not get updated after deactivating the scope-policy configuration. This could result in the PIM register process to be incomplete and further multicast traffic to be dropped.
PR Number Synopsis Category: RPD policy options
1565629 The rpd might crash when the deletion of routing table occurs
Product-Group=evo 		The rpd might crash when the deletion of routing table occurs.
PR Number Synopsis Category: RPD route tables, resolver, routing instances, static routes
1635009 Multipath route getting formed for a VPN prefix due to incorrect BGP route selection logic
Product-Group=evo 		On all Junos and EVO platforms running BGP, when a specific route is received from multiple places under a VRF, multipath route is getting formed even though the BGP route selection algorithm has the active route with higher local preference. Once multipath is formed, the traffic forwarding is happening based on that, and it may result in some traffic going to an unwanted path. Please refer to KB37775 [juniper.net] for more details.
PR Number Synopsis Category: PTX10K specific platform PRs
1586728 PTX10008 EVO supports multiple LED entries in snmp mib walk
Product-Group=evo 		On PTX10008 EVO, multiple LED indexes and color values are added for each FRU on jnxLEDState MIB.
PR Number Synopsis Category: PTX10K RE EVO Issues
1589433 Scapa:RCB: Error/Warning message to be corrected, when we try to do "request chassis cb slot 1 offline", before node offline
Product-Group=evo 		Error message "Node is online", emitted while performing steps described in the PR, is modified as shown below: root@re0-re0> request chassis cb slot 1 offline Error: CB offline is not allowed, when node is in online state. Try to offline the node using "request node offline [node]"
PR Number Synopsis Category: SRX branch platforms
1630886 LLDP packets may be sent with incorrect source MAC for RETH/LAG child members
Product-Group=evo 		On all platforms, when LLDP is run on child members of LAG/Redundant Ethernet (RETH) interfaces, LLDP packets may not be sent out with interface hardware address as source MAC. Instead, packets will be sent out using LAG/RETH interfaces MAC address. So, in case if the RETH/LAG interface MAC address changes, the source MAC address of LLDP packets will also change dynamically. Which will affect any service that relays on LLDP.
PR Number Synopsis Category: Trio pfe l3 forwarding issues
1568944 Traffic might be dropped when the default route is changed in inet.0 table
Product-Group=evo 		Traffic might be dropped on MX Series routers or the EX9200 line of switches when the default route is changed in the inet.0 table. It might take 2 to 3 seconds to be updated in the Packet Forwarding Engine. This issue can be recovered automatically.
PR Number Synopsis Category: Issues related to Logging/Tracing, errmsg, eventd infrastruc
1612535 Syslog messages may be lost partially in case of lots of messages generated to eventd
Product-Group=evo 		On all Junos/EVO platforms, with 100 or more IFL's up/down, a lot of messages will be generated within few ms and eventd may not be able to process all of messages and may be collected/lost partially.
 

20.4R3-S2-EVO - List of Known issues

PR Number Synopsis Category: EVO platform software
1597999 PTX10001-36MR: Inconsistency in the platform name used in multiple places, version, snmp mibs, etc.
Product-Group=evo 		"show snmp mib walk sysDescr" will show ptx10001-36mr
PR Number Synopsis Category: PFE COS features on BT based platforms
1616772 The Strict-Priority-Scheduler (SPS) might not work accurately across port queues
Product-Group=evo 		On EVO PTX platforms, an inaccurate traffic distribution across port queues is observed which have strict priority scheduler enabled on 10G interface.
PR Number Synopsis Category: Express BX PFE L3 Features
1648156 EVO adding config "hash-key family inet layer-4" disables inet Hash-key Protocol.
Product-Group=evo 		EVO adding config "hash-key family inet layer-4" disables inet Hash-key Protocol.
PR Number Synopsis Category: EVO L2 Control Plane PRs
1598217 arpd and ndp daemon crash in scale setups
Product-Group=evo 		In scaled scenarios (4k BDs, IRBs), with restart of l2ald and pfe daemons, arpd and ndp daemon crash is observed and it recover by itself and no functionality impact expected
PR Number Synopsis Category: Express PFE CoS Features
1580795 In certain scenarios, shapers applied on a 10g interface may drop the traffic more than the configured max-rate.
Product-Group=evo 		In certain scenarios, shapers applied on a 10g interface may drop the traffic more than the configured max-rate.
PR Number Synopsis Category: Express PFE L3 Multicast
1634982 MTS-MCAST: [Brackla] Auto RP base verification failed with multiple RPs with same group range
Product-Group=evo 		We have found the workaround which is to deactivate/activate auto-rp config on brackla. This is PR is opened since 20.4 and there is no LKWR .We will be fixing it in DCB.
PR Number Synopsis Category: Port Mirroring feature on express PFE
1618589 PFE restart is causing port-mirroring to stop working on the respective pfe
Product-Group=evo 		Port mirroring configuration stops working, when a given PFE restarts.
PR Number Synopsis Category: jdhcpd daemon
1618977 Junos OS: The jdhcpd crashes upon receiving a specific DHCP packet (CVE-2022-22179)
Product-Group=evo 		An Improper Validation of Specified Index, Position, or Offset in Input vulnerability in the Juniper DHCP daemon (jdhcpd) of Juniper Networks Junos OS allows an adjacent unauthenticated attacker to cause a crash of the jdhcpd and thereby a Denial of Service (DoS). Refer to https://kb.juniper.net/JSA11285 [juniper.net] for more information.
PR Number Synopsis Category: PFE infra to support jvision for EVO
1640442 [jvision] [jvisiontag] Verification of DB data collection is failing after executing the jvision decoder
Product-Group=evo 		The system ID that is exported should be different for UDP and GRPC/GNMI. In the case of UDP, the system name should be appended with the local IP address.
PR Number Synopsis Category: RPD Next-hop issues including indirect, CNH, and MCNH
1619229 Traffic loss observed on BGP-LU paths after restoring primary paths in setup with IGP/LDP RLFA enabled
Product-Group=evo 		Traffic loss observed on BGP-LU paths after restoring primary paths in setup with IGP/LDP RLFA enabled
PR Number Synopsis Category: PTX10K specific platform PRs
1478951 PTX10008 graceful OIR leads to missing of sensor data for power and temperature. Follow the optimized step given as a workaround to have OIR working.
Product-Group=evo 		PTX10008 graceful OIR leads to missing sensor data for Power and Temperature. Please follow the optimized step given as a workaround to have OIR working.
1552955 Junos Evolved Ethernet Mac Address allocation is incorrectly distributed for FPCs
Product-Group=evo 		When upgrading the Junos Evolved software to versions with this change, the Ethernet MAC addresses on PTX interfaces will be changed. If your deployment depends on static MAC address configurations, you should make a plan to accommodate this.
PR Number Synopsis Category: UI Infrastructure - mgd, DAX API, DDL/ODL
1608718 In an SRX cluster with VPN configuration, primary node in cluster may generate kmd core files in a loop when a commit fails with "lock can not be taken on other node" followed by another commit.
Product-Group=evo 		When a commit is failed due to "lock can not be taken on other node", "/var/etc/vpn_tunnel.id+" and other ffp(foreign file propagation) files are not getting cleaned up on srx cluster. In next commit, these stale ffp files are activated for use, it is resulting in discrepancy and resulting in assert failure in applications/KMD daemons. The problem moves to the secondary node if the failover is executed.
PR Number Synopsis Category: VNID L2-forwarding on Trio
1630163 Scenario where within a VRF multiple VXLAN type-5 tunnels with same decap prefix are not handled on MPC10/11
Product-Group=evo 		On MPC10/11 line cards, EVPN/VXLAN scenario where within a VRF multiple VXLAN type-5 tunnels with same decap prefix are created is not well handled, which might lead to traffic drop.
PR Number Synopsis Category: Express PFE L2 fwding Features on ZX platforms
1643308 The addition of new member to LAG might result in FPC crash
Product-Group=evo 		On all PTX platforms running EVO with LAG (Link Aggregation Group) enabled, when a new member is added to an existing AE (Aggregate Ethernet) interface which has L2 (Layer 2) configuration and has MAC (Media Access Control) learning happening around the same time, it might cause memory corruption. This would result in FPC crash with core. The issue is rare.

Modification History

First publication 2022-03-09