Alert Type

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

Product Affected

ACX PTX QFX Running Junos Evolved Software

Alert Description

Junos Software Service Release version 21.4R3-S1-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 21.4R3-S1-EVO is now available.

21.4R3-S1-EVO - List of Fixed issues

PR NumberSynopsisCategory: Border Gateway Protocol
1670673Source/Destination AS fields shows up as 0 in the flow record
Product-Group=evo		On all Junos and Junos Evolved platforms, Inline Jflow may export flow data incorrectly setting source/destination AS to 0 or setting unrelated AS number, when a change happens to the BGP route in a routing table. A change need not make route selection happen among BGP routes. Increasing and decreasing the number of hidden routes, which does not affect to existing active route, could cause wrong AS information in flow data.
PR NumberSynopsisCategory: PFE COS features on BT based platforms
1673738CoS drops seen for priority traffic on some PTX platforms
Product-Group=evo		On certain EVO-PTX platforms, traffic impact on high-priority queues is seen. This is due to inaccurate priority scheduling.
PR NumberSynopsisCategory: Express BT PFE L3 Features
1638904Traffic does not load-balanced
Product-Group=evo		On EVO-based PTX platforms, traffic is not load balanced if performing deactivate and activating routing-options static.
1671065Fragment frames errors will be seen on the 400G interface
Product-Group=evo		On the PTX10003 platform, fragment frames error will be seen on the 400G optics interface when the optics cable interoperates with third-party equipment. Due to this, some packets would be lost.
PR NumberSynopsisCategory: Firewall related development
1683361The aftmand process crash might be observed
Product-Group=evo		On all Junos OS Evolved platforms configured with Border Gateway Router (BGP) hierarchy in prefix-list apply-path, then changing the BGP configuration associated with apply-path will cause the firewall filter to be updated and re-program within the Packet Forwarding Engine (PFE). When filter change is processed, aftmand process crash would be observed leading to system crash.
PR NumberSynopsisCategory: Distributor related issues
1658426Routes in RIB and FIB table may go out of sync on all Junos Evolved platforms and causes a traffic impact
Product-Group=evo		On all Junos evolved platforms, routes in RIB (Routing Information Base) and FIB (Forwarding Information Base) are not in sync when it is running over long period of the time (200+ days) with a constant route churn. It causes a traffic impact due to stale entries.
PR NumberSynopsisCategory: management ethernet related issues - mgmt-ethd daemon
1648371IPv6 master-only IP address does not move to the new master RE after a switchover.
Product-Group=evo		IPv6 master-only IP address does not move to the new master RE after a switchover. As a workaround please deactivate and activate the master-only IPv6 address on the new master.
PR NumberSynopsisCategory: System Management High availability
1673306The new master RE could self-reboot after the kernel crashes on an old master
Product-Group=evo		On all Junos Evolved chassis platforms, after an old master kernel crashes then the new master RE (routing engine) could self-reboot after taking over the old RE (routing engine) master role due to a flaw in mastership arbitration logic. There might be a service impact because the old master is having issues and the new master will be rebooted which could blackhole the traffic. This is a very rare corner case.
PR NumberSynopsisCategory: Zookeeper related issues
1678880RE reboot, rpd crash, etc. can be seen if the volume of zookeeper logs is high on Junos Evolved platforms
Product-Group=evo		Undesirable and impactful behavior such as kernel crash, RE (Routing-Engine) reboot, rpd crash, etc. can be seen if the volume of recorded zookeeper logs is high in the /run/zookeeper/conf/default/version-2 directory.
PR NumberSynopsisCategory: Layer2 forwarding on EX/NTF/PTX/QFX
1675861Traffic flow will be affected as interfaces will be removed from VLAN
Product-Group=evo		On all Junos and Junos-Evolved platforms, when both vlan name (vlan name having vlan-id-list under "edit vlans" hierarchy) and vlan-id-list having overlapping vlan IDs are added to the interface, traffic will be affected when the vlan name is deleted from the interface.
PR NumberSynopsisCategory: Issues related to Junos licensing infrastructure
1658869The license might get out of sync between master and backup RE
Product-Group=evo		On all Junos OS Evolved platforms, the license might get out of sync between master and backup RE after the addition/deletion of licenses.
PR NumberSynopsisCategory: OSPF routing protocol
1634162OSPF adjacency may take longer to converge when the neighbor restarts non-gracefully
Product-Group=evo		On possibly a different vendor device or a Junos device running Junos OS Release 15.1 or later, when an OSPF neighbor is restarted non-gracefully, there might be a delay in convergence. An eventual traffic loss might appear as the Junos device marks the immediately originated LSA from the restarting neighbor as stale and waits for the next Type-1 LSA to arrive.
PR NumberSynopsisCategory: Path computation client daemon
1678874The rpd process crashes when a delegated LSP with IPv6 install prefix is configured
Product-Group=evo		On all Junos and Junos Evolved platforms, when a LSP (Label Switched Path) is configured with IPv6 install prefix and the management of the LSP is delegated to a PCE (Path Computational Element) controller, rpd process crashes.
PR NumberSynopsisCategory: SRX branch platforms
1594014During reboot, "warning: requires 'idp-sig' license" can be seen on the screen even when the device has valid license
Product-Group=evo		If a device is rebooted manually or reboots for any other reason, The following messages can be seen on the boot up screen even when the device has valid license and proper configuration to use the features like IDP/UTM
PR NumberSynopsisCategory: Express ZX PFE L3 Features
1652623After Configure rpf-check on lag interface , lag interface is going down
Product-Group=evo		After Configure rpf-check on lag interface , lag interface is going down
 
 

21.4R3-S1-EVO - List of Known issues

PR NumberSynopsisCategory: Express BT PFE L3 Features
1664600PDT: Invalid massive count seen on firewall filter counters
Product-Group=evo		On adding filter to loopback interface, at times wrong count seen for firewall policer. workaround is to clear firewall all once.
PR NumberSynopsisCategory: Alias for DHCP issue on DNX based platform.
1689194Xmen: DHCP Server not honoring range config in IP Pool
Product-Group=evo		Current DHCP server implementation on XMEN uses "set access address-assignment pool family inet network" configuration to allocate all possible IP's from the subnet. One possible workaround is to adjust the netmask according to either increase or decrease the pool size. Full functionality of authd is not available on XMEN as it is a small-size platform.
PR NumberSynopsisCategory: Layer 3 forwarding, both v4+v6
1677123ECMP Next hop installation failure
Product-Group=evo		For ECMP nexthop installation in 21.4R3, there have been some timing issues which have been fixed in 22.2 as a result of design change. In certain scenarios, if there are multiple scale configurations seen with Unilist next hop indices with AE IFLs, then for some of the next hops, HW programming can fail. Therefore, programming of all routes pointing to the Unilist next hop will also fail. Similar issue has been identified and fixed in the PR:1637301 with certain design changes
1677855ECMP next hop installation failure.
Product-Group=evo		For ECMP nexthop installation in 21.4R3, there have been some timing issues which have been fixed in 22.2 as a result of design change. In certain scenarios, if there are multiple scale configurations seen with Unilist next hop indices with AE IFLs, then for some of the next hops, HW programming can fail. Therefore, programming of all routes pointing to the Unilist next hop will also fail. Similar issue has been identified and fixed in the PR:1637301 with certain design changes
PR NumberSynopsisCategory: Ztp related issues
1683964PTX EVO : Major alarm "Application ztp fail on node Re0" appears about one day later post system zeroize
Product-Group=evo		On PTX EVO, application ztp fails on node Re0 about one day later post system zeroize.
PR NumberSynopsisCategory: EVO RPD agent PRs
1665362Sometimes BGP and RSVP sessions remain down after quick arpd process disable and enable.
Product-Group=evo		Sometimes BGP and RSVP sessions remain down after quick arpd process disable and enable. Whenever customer encounter such scenario, system can recovered from erroneous state by executing "restart routing gracefully" in cli.
PR NumberSynopsisCategory: cmdd, crd issues
1655441agentd core - seen upon FPC ungraceful restart with cda-bt process killed
Product-Group=evo		when evo-cda-bt is killed, fpc is restarted. We might see agentd crash due to ungraceful FPC restarts. This issue is not seen in a normal working scenario and is seen only upon ungraceful fpc restarts.
PR NumberSynopsisCategory: Configd, ffp issues
1647853The traffic might not flow after deleting/adding VLAN config with load override
Product-Group=evo		On Junos Evolved platforms, when vlan-bridge and corresponding logical interface is configured, a link is created between them. When the logical interface is deleted using the load-override method, the link information is not backed up in the system. When the same logical interface is re-created, the link creation might not happen. Delete the logical interface and re-create the same using load-update to recover from the issue.
1680192MFT: AE interface remain down as physical interface didn't attach to AE when GRES just after MFT config.
Product-Group=evo		In EVO, when any configuration is committed and 'commit complete' is seen on the cli, the system operational state will take a little extra time to reflect the committed configuration. This extra time is taken by EVO configd daemon to process the commit. The extra time taken is proportional to the size of the committed config. In the PR scenario, the RE switchover is triggered immediately after the commit complete. The immediate RE switchover interrupts the commit processing, which is still under progress by configd to publish the objects to EVO DDS(Distributed Data Store). This results in an inconsistent EVO operational state which is out-of-sync with the committed config.
PR NumberSynopsisCategory: mgd, ddl, odl infra issues
1674890[manageability_mgd_infra] [generic_evo] Scapa : [PRedator] Seen configd-streamer core in 22.3R1EVO evf_walk_and_diff cfg_print_reorder
Product-Group=evo		Configd-streamer cores during commit of wild-carded groups related config. The core is only seen with the wild-carded config which is used in the reported fusion test case.
PR NumberSynopsisCategory: All Guardian (ACX7509) Linecard (FPC) related issues
1592388ACX7509: some of the interfaces from 16x100G and 20XSFP56 will not go down after evo-pfemand restart
Product-Group=evo		ACX7509: some of the interfaces from 16x100G and 20XSFP56 will not go down after evo-pfemand restart.
PR NumberSynopsisCategory: OSPF routing protocol
1689057PDT - SCAPA - with CFM scale, under NSR, RE switchover, seeing ospf neighbor bouncing on scaled CFM interfaces
Product-Group=evo		We should have this cli configured "set routing-options ppm redistribution-timer 120? to ensure seamless operation of protocols during NSR.
1690036ospf session events need not be present in syslogs always
Product-Group=evo		When there is system wide churn in the system (like post switch-over etc) ospf events can be dropped from syslog.
PR NumberSynopsisCategory: Path computation client daemon
1675816In a rare case, 'pccd' will crash when the PCEP connection is down
Product-Group=evo		On all Junos and Junos Evolved platforms, on the intermediate device when connection towards Path Computation Element (PCE) is down, Path Computation Element Protocol (PCEP) session will be down resulting in pccd crash. Routing protocols are not impacted but pccd crashes.
PR NumberSynopsisCategory: Category for QFX5K EVO Platform Software PRs related to Inte
1618488On QFX5700m, 400G DAC flap may be seen after OIR, FPC restart, device reboot enable/disable interface
Product-Group=evo		On QFX5700m, 400G DAC flap may be seen after OIR, FPC restart, device reboot enable/disable interface
PR NumberSynopsisCategory: KRT Queue issues within RPD
1612487On backup RE during GRES, you may see "RPD_KRT_KERNEL_BAD_ROUTE: krt unsolic client.128.0.0.5+62000: lost ifl 0 for route" warning messages
Product-Group=evo		Several warning messages show up while the RPD process restarts during performing GRES on a system running Junos EVO.
PR NumberSynopsisCategory: PTX10K RE EVO Issues
1681609PTX10008 EVO : STS(Status) LED on backup RCB goes off and never turns on when backup RE is rebooted
Product-Group=evo		On PTX10008 EVO, STS(Status) LED on backup RCB goes off and never turns on when backup RE is rebooted.
PR NumberSynopsisCategory: Junos Automation, Commit/Op/Event and SLAX
1656436A large configuration might not get committed at boot
Product-Group=evo		On all Junos and Junos Evolved, the mgd process might fail to commit a scaled configuration (~350K set statement) during boot due to insufficient memory.

Modification History

First Publication 2022-11-08