PTX and QFX series running Junos Evolved

Junos Software Service Release version 20.4R1-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

Junos Software service Release version 20.4R1-S1-EVO is now available.

20.4R1-S1-EVO - List of Fixed issues

PR Number	Synopsis	Category: EVO interface software
1563684	ARP may not resolve and traffic may be dropped on Junos Evolved platforms Product-Group=evo	In scaled IFL scenario when interface connecting customer with many IFLs in same port is flapped, ARP may not resolve and traffic may be dropped.
PR Number	Synopsis	Category: EVO platform software
1534996	Interfaces may take longer to come up after loading baseline and rollback configuration Product-Group=evo	In scaled setup with a large number of routes and next-hops, and reloading the baseline config; interfaces and specifically AE interfaces take longer to come up. After loading baseline and rollback traffic loss is expected. The router comes back into service after 5 mins on evo whereas it takes 1.5 mins on Junos. This a router reconfig/upgrade scenario. /
PR Number	Synopsis	Category: Express BT PFE L3 Features
1551363	Traffic drops may be seen after egress FPC restarts and comes back online Product-Group=evo	On PTX platforms, the throughput might fall below the advertised line-rate if multiple traffic flows use the same egress interface. Hence traffic loss may be observed.
PR Number	Synopsis	Category: EVO Class of Services
1544531	The cosd may not come up after FPC restart Product-Group=evo	On Evo platforms, cosd process may not come up after FPC restart. Due to this, the classifier/rewrite/scheduler functionalities may stop working. This issue may be seen during validation check for CNP (Congestion Network Profile) and Drop profile configuration. This issue has a functional impact.
PR Number	Synopsis	Category: Issues related to evo operations - libevo infra, typeinfo ..
1564156	It might take a long time to create IFDs after restarting the FPC Product-Group=evo	In a Junos Evolved chassis with multiple FPCs scenario, while restarting an FPC, it might sync the configuration from other FPCs in the system. But if any of the FPC holds a scale of configuration, the evo-aftmand application might take a lot of time to reconcile the huge number of objects and will be stuck in an "online" state on this FPC, the IFDs might not be crated for a long time (above 1 hour).
PR Number	Synopsis	Category: Interface PRs defect & enhancement requests
1566752	PTX10001-36MR: Control IFL may not be present for ports et-0/0/11 and et-0/2/11 Product-Group=evo	PTX10001-36MR: Control IFL may not be present for ports et-0/0/11 and et-0/2/11
PR Number	Synopsis	Category: Express PFE MPLS Features
1562503	Slow memory leak Product-Group=evo	For topologies involving High Ingress and transit LSP scale, error messages can be seen in journalctl when tearing down the Ingress and Transit LSPs. This also leads to slow memory leak.
PR Number	Synopsis	Category: Interface PR for Lazurite LC
1562471	Interface loopback might not work if there is no optics connected to the port on PTX10000 Product-Group=evo	For retimer ports on PTX10000 platforms, loopback configuration might not work if there is no optics connected to the port.
PR Number	Synopsis	Category: OSPF routing protocol
1563350	The rpd might crash on Backup RE after the rpd restart is triggered on Master RE Product-Group=evo	On all Evo platforms, if the rpd restart is triggered on Master RE (using the following command : restart routing immediately) then rpd might crash on Backup RE.
PR Number	Synopsis	Category: UI Infrastructure - mgd, DAX API, DDL/ODL
1559786	It fails if the xml output from command "request vmhost mode test | display xml rpc" is picked and used in netconf Product-Group=evo	On vmhost platforms, if the xml output from command "request vmhost mode test | display xml rpc" is picked and used in netconf, it will fail. set vmhost mode custom test layer-3-infrastructure cpu count MIN set vmhost mode custom test layer-3-infrastructure memory size MIN set vmhost mode custom test nfv-back-plane cpu count MIN set vmhost mode custom test nfv-back-plane memory size MIN set vmhost mode custom test vnf cpu count MIN
PR Number	Synopsis	Category: Express ZX PFE L3 Features
1560901	[NGMVPN] [IR] : EVO-Brackla :: 20.4R1-EVO: Packets dropped with NORMAL DISCARD after PFE 0 power off/on done on the receiver PE with multicast receivers for groups 225.2.1.1 and 225.3.1.1 Product-Group=evo	Once a pfe restart is done, MVPN traffic coming from the core will not get forwarded . This is because before reset, FDB was having entries for forwarding the MVPN multicast traffic (classD) . Once the restart is done, these entries are deleted and not created again. This results in MVPN packets taking the wrong path (unicast forwarding path) and thereafter hits the RESOLVE route which is seen as policer drop/trapcode. In short MVPN functionality will not work if a pfe restart happens, as the MVPN handling FDB entries are not added back . To fix the issue, MVPN fdb enties should be added on per pfe basis and make sure the classD entries are getting added when a PFE is restarted
1562452	Complete ingress multicast traffic loss may be seen on interfaces that are flapped using PFE offine/online command Product-Group=evo	In a scenario where the upstream interface is an AE interface, if a PFE is made offline, one of the member link of the AE interface may also get removed. The ingress NH programming for the PFE instance will be removed from the hardware. If that PFE is brought online again, there is no NH rebake and the ingress NH words may be missing in hardware because of which there is multicast traffic drop.

 

---

20.4R1-S1-EVO - List of Known issues

PR Number	Synopsis	Category: Border Gateway Protocol
1555288	The BGP session might not come up if extended-nexthop is enabled by default on the other vendor remote peer. Product-Group=evo	On all Junos platforms, if the BGP remote peer runs other vendor software and enables "extended-nexthop" by default, the BGP session might not come up due to the software defect if the knob is not configured on Juniper device.
PR Number	Synopsis	Category: EVO build infrastructure related issues.
1508626	Add Python 3.x for EVO - module list Product-Group=evo	There are Python 3.x modules that are missing from the Junos Evolved library. The equivalent of these modules is already installed under Python 2.7 module library.
PR Number	Synopsis	Category: EVO Class of Services
1566161	cosd crash may be observed with classifier & rewrite bindings Product-Group=evo	cosd crash may be observed in the following scenarios configs : cos features (classifier & rewrite and corresponding IFL bindings) 1) with deactivate & activate cos configs along with deactivate & activate of interface 2) Daemon restart - class-of-service 3) Router reboot
PR Number	Synopsis	Category: Ztp related issues
1561142	When user aborts ztp using config commit, config committed by user should persist. Product-Group=evo	When user aborts ztp by applying configuration, ztp will abort. This fix addresses cases where the user configuration is not retained when ztp exits.
PR Number	Synopsis	Category: Lacp related problems and issues.
1553196	[lacp] [lacptag] PTX10004 lacpd core is obserevd after router reboot Product-Group=evo	The "lacpd" process may generate a core dump file while a router is shutting down in the following scenarios: a. router reboot b. router upgrade Since the router is shutting down, this won't have an impact on forwarding traffic.
PR Number	Synopsis	Category: EVO MACSEC Platform Independent Implementation
1566665	BT:MACsec-"Encrypted packets counter displays 0 under "Secure Association transmitted" of MACsec statistics output when AN rollovers with sak-rekey-interval configured" Product-Group=evo	Macsec counters are being read only for AN0, AN1. hence, AN2 and AN3 counters are always zero. The fix takes care of copying SA statistics from AN0 to AN2 and AN1 to AN3
PR Number	Synopsis	Category: Express PFE CoS Features
1553943	[aft] [generic_evo] ptx10004 : PTX10004 :: - seeing cmerror dp_1_viqm_intr_req_enq_err, not cleared after loading latest 20.4 EVO Product-Group=evo	In scenario of queue congestion, interface enable/disable or in PFC mode this interrupt can come on this release.
1554220	"CoS WRED Curve: Create Expr Curve: No curve data points!!" errors are seen when interpolate is configured under drop profile. Product-Group=evo	"CoS WRED Curve: Create Expr Curve: No curve data points!!" errors are seen when interpolate is configured under drop profile. Interpolate option under drop profiles is not supported in ptx-10008.
PR Number	Synopsis	Category: KRT Queue issues within RPD
1561812	EVO System - after recovering from restart routing immediately, object-info anomalies is observed on rpdagent Product-Group=evo	This issue is applicable to Junos Evolved platform which the RPD process restarts either by mastership switched over, or restarts after a crash, or via the "restart routing immediately" command. The ISIS routing protocol may not work correctly after the restart event.
PR Number	Synopsis	Category: PTX10K platform specific fabric PRs
1497212	Ungraceful SIB failures result in transient loss of traffic Product-Group=evo	On a PTX10008, six SIBs are required to carry line-rate traffic, with no fabric redundancy. Even when ingress traffic rate is such that five SIBs are sufficient to carry ingress traffic (for example, traffic is less than 1280 Gbps), ungraceful SIB failures result in transient loss of traffic, till system failure handling is triggered. In Junos OS Release 20.1R2, failure handling may result in about 4-6 seconds of traffic loss. We recommend that you take the fabric cards offline by using the request chassis sib offline command before removing the SIBs for maintenance.
1539685	PFEs on FPC stuck in "READY" state on New Master RE with GRES switch while FPC is restarting. Product-Group=evo	The software doesn't support GRES when FPCs are triggered for restart. GRES with FPC triggered for offline or online is supported.
PR Number	Synopsis	Category: Express PFE L2 fwding Features on ZX platforms
1568566	Routes learned via IRB interface might not be reachable in IBGP setup Product-Group=evo	On Junos EVO-based PTX10003/QFX10003 platforms with IBGP setup, the route is not reachable if it is learned via indirect Next Hop, which further points to unicast Next hop and getting resolved through IRB interface. There will be complete traffic for the affected route.

First publication 2021-03-16