- Introduction
- play_arrow Junos OS Evolved Release Notes for PTX10001-36MR, PTX10003, PTX10004, and PTX10008 Devices
- play_arrow What's New
- play_arrow What's Changed
- What’s Changed in Release 21.2R1
- Known Limitations
- Open Issues
- play_arrow Resolved Issues
-
- play_arrow What's New
- play_arrow What's Changed
- Known Limitations
- Open Issues
- play_arrow Resolved Issues
- Upgrade Your Junos OS Evolved Software
- Licensing
- Finding More Information
- Documentation Feedback
- Requesting Technical Support
- Revision History
Open Issues
Learn about open issues in this release for the ACX7100-32C and ACX7100-48L.
For the most complete and latest information about known Junos OS Evolved defects, use the Juniper Networks online Junos Problem Report Search application.
General Routing
Monitor interface traffic statistics is not flushed out after interface goes down. This is a common behavior across Junos OS Evolved platforms. Even on re-enabling it is still not flushed.
PR1564672Errors related to pfe_portE, loopback on system reboot. [Interface: can't get pfe_portE for ifd:ae0, Interface: PIL set loopback failed for IFD:ae0 error:-1] seen. There is no functional impact. PR1565831
If you configure unsupported actions in egress firewall filter, commit errors are thrown. If the unsupported actions are removed from the configuration and committed, the filter might not work in the Packet Forwarding Engine even then. This behavior depends on the order the BIND and MODIFY of filter comes to the Packet Forwarding Engine from firewalld. Syslog errors and pfe-cli commands can be used to check the filter install status. PR1568525
PFC feature is not supported on ACX platforms. Therefore
show interfaces interface-name extensive
command always displays the MAC Priority Flow Control Statistics table with 0 value. PR1576317Each routing instance has an associated {vrf, jtdv} device pair, and these devices are managed as a service by systemd. As such, whenever a routing instance is created or deleted there is work that systemd needs to do, and when using scaled numbers of routing instances the
The amount of CPU consumed by systemd can be high whilst these devices are being created or deleted.
The time to take down or boot the system is longer. PR1581283
If evo-pfemand process is restarted and at the same time if ethernet-connectivity-fault-management has also restarted, there is a possibility that a ppmd core occurs. As evo-pfemand is restarted, the services are also interrupted. Once the process is back online, all the CCM sessions recover post the mentioned triggers. PR1585178
In scaled cfm config scenarios, ppmd or evo-pfemand process core file might be hit after a router reboot. PR/1591534
After a router reboot, in scaled cfm config scenario, the sessions (some or all) might not come up. Deactivating and reactivating cfm config must resolve the issue. PR/1591978
On ACX7100 evo-pfemand core file can be seen rarely with link flaps when there is VPLS configuration with high number of MAC entries learnt (> 10,000). PR/1596587
Few LSI MACs are not properly learnt in the software with 8000 vpls instance scale: The issue is seen above 4000 VPLS instances, hardware LSI MAC learning is fine. However, we observe some difference in LSI MAC learned in the software. PR1597125
No MAC address present in the Ethernet table or Junos OS Evolved Packet Forwarding Engine, however arp is present: The issue is seen only while upgrading or downgrading an image on the router. The issue is only applicable to aggregated Ethernet interfaces, physical interfaces are fine. The hardware learns the MACs properly, only from the software point of view MACs are missing. PR1597277
In scaled scenarios (4k BDs, IRBs), with restart of l2ald and pfe daemons, arpd and ndp daemon crash is observed and it recovers by itself and no functionality impact expected. PR1598217
Egress firewall ACL rules with count or policer actions are skipped for BUM traffic and does not hit. This is because of ASIC limitation for BUM traffic and applies to all services like Layer 2, Layer 3, EVPN etc. PR1598489
In ACX7100-32C, it causes the traffic to drop while sending continuous traffic against the ACX7100-32C 400G ports. PR1601151
For ACX7100-32C and ACX7100-48L, the voltage thresholds are not correct which might cause the Voltage Threshold Crossed alarm sometimes, however it does not impact the functionality. PR1601493
On ACX7100-32C or ACX7100-48L,
show system firmware
some times might show Current Firmware version for FPC 0 as blank. This is only show CLI issue due to random read failure issue of FPC 0's firmware version and this does not impact any normal operations of FPC 0. PR1618949- When IPv6 NDP is configured on IRB interface with one of the l2-interfaces from BD, it programs the NDP correctly. However, when the Layer 2 interface is changed for that NDP with other Layer 2 interface from same BD, the configuration does not taking effect. NDP still points to the old Layer 2 interface. PR1602894