Open Issues
Learn about open issues in this release for EX Series switches.
For the most complete and latest information about known Junos OS defects, use the Juniper Networks online Junos Problem Report Search application.
General Routing
-
misleading syslog message "L2CKT/L2VPN acquiring mastership for primary" though no VPN/L2CKT configured on the routerPR1105459
-
runt, fragment and jabber counters are not incrementing on EX4300-MPsPR1492605
-
On all EX platforms, whenever beacon LED functionality is enabled, there is a mismatch between the physical LED status and the output of the CLI command ?show chassis led? showing incorrect port LED status for interfaces as LED up instead of off.PR1697678
-
The interface of ge-x/0/1 port might go down after virtual-chassis split and merge on EX4300-VCPR1745855
-
EX4300MP: VC member status toggling between "Inactive" and "NotPrsnt" state after member downgradePR1751871
-
on EX2300-48 MP without VC pre-provisioned configuration, If master's member-id and master member's interface config are changed then VC is taking more time to get stabilized.PR1764542
-
After rebooting a mixed Virtual Chassis (VC) of EX4300-xxP and EX4300-MP switches or rebooting a EX4300-xxP member, interfaces with Power over Ethernet (PoE) configured will not come up on EX4300-xxP members.PR1782445
-
Ex-Hardening:Local/Remote fault insertion from TG is failingPR1789999
-
On EX4300-MP platforms, ge (Gigabit ethernet) interfaces configured for 10 or 100Mbps (Mega bits per second), jumbo frame packets will be dropped.PR1812891
-
For Junos OS platforms, in a specific configuration change after NSSU (Nonstop Software Upgrade), i.e. delete and add sequence of LAG (Link Aggregation Group) bundles performed via load baseline configuration and re-apply original configuration, OSPF (Open Shortest Path First) session might get stuck in EXSTART state. This issue will impact the traffic.PR1817034
-
On Junos platforms, if disk is full, when scaling configuration is moved to baseline configuration and move back to scaling configuration, l2ald try to create a source VTEP, but old VTEP is still in the middle of deletion, so kernel return EBUSY and l2ald will retry. If retry too many times, l2ald insist and core.PR1817705
-
Traffic loss will be seen on 1G-SFP-T if speed is configured to 100m. 1G SFP-T has the AN feature enabled but the PHY we have b/w SFP-T and switch ie., PHY82756 doesn't support AN and this mismatch is causing the traffic loss. This needs feature enhancement PR1817992
-
Time Domain Reflectometry (TDR) support for detecting cable breaks and shorts aborts intermittently on some random ports.PR1820086
-
On Junos QFX and EX platforms in an EVPN-VXLAN (Extended Virtual Private Network- Virtual Extensible LAN) CRB (Centrally-Routed Bridging) scenario where the ingress leaf switch is configured with ESI (Ethernet Segment Identifier) lags (i.e. the server is multihomed), if there is an overlap between ESI lag(s) trunk ID with physical port number(s) and overlap of DMAC (destination MAC) between VGA (Virtual Gateway Address) MAC address 00:00:5e:00:01:01 (CRB setup with VGA / GW is on spine) with VRRP (Virtual Router Redundancy Protocol) MAC (specifically for the VRRP group 1 MAC address 00:00:5e:00:01:01) on the physical ports of the Leaf switches, then traffic loss will be observed for the inter-VLAN traffic.PR1820830
-
On all Junos QFX5K platforms, with ECMP (Equal Cost Multi Path) configured, when there is any routing protocol change (like ISIS cost metric change), the protocol traffic on the network is dropped.PR1823601
-
On a working VC system, if there happens a dc-pfe process restart for any reasons, then there is a possibility of some interfaces not getting created after the dc-pfe restart.PR1823688
-
When a poe bounce command is issued in quick succession for multiple ports, the 'poe enabled' logs may not be printed for some of the poe ports. This is a cosmetic issue and functionality works as expected.PR1845161
-
For EX4100 platforms, with default UFT profile configured and high scale of IPv6 host routes, hash collisions may prevent routes from being installed in the ipv6 host table. Additionally, these routes might not get installed in the LPM table as well because the size limit for ipv6 prefix > 64 is restricted to 1k, resulting in table full errors. This can result in traffic loss. When the issue occurs, below error message could be observed: fpc0 brcm_rt_ip_uc_lpm_install:1583(LPM route change failed) Reason : Table full unit 0 fpc0 brcm_rt_ip_uc_host_install:2355(ip host add into LPM table failed) Reason :Operation failed fpc0 brcm_rt_ip_uc_entry_install:1321brcm_rt_ip_uc_entry_install Error: host(/32) ip route install failed vrf 9 ip 100:27::52:102 nh-swidx 6381 nh-hwidx 400215PR1849471
High Availability (HA) and Resiliency
-
Graceful Routing Engine Switchover (GRES) not supporting the configuration of a private route, such as fxp0 , when imported into a non-default instance or logical system. Please see KB https://kb.juniper.net/InfoCenter/index?page=content=KB26616 resolution rib policy is required to apply as a work-aroundPR1754351
Platform and Infrastructure
-
An Improper Check for Unusual or Exceptional Conditions vulnerability in the Packet Forwarding Engine (PFE) of Juniper Networks Junos OS on EX4300 Series allows a locally authenticated attacker with low privileges to cause a Denial-of-Service (Dos). Please refer to https://supportportal.juniper.net/JSA79186 for more information.PR1774634
-
On EX4300 or EX4300-VC, removal of a Physical Interface Card (PIC), or if the software fails to detect a PIC that is installed, it can cause a crash in the pfex process. This crash can lead to high CPU usage and potentially disrupt network traffic.PR1779410
-
On EX4300 Platforms, Packet Forwarding Engine (PFE) crash will be seen due to an unexpected switchover after committing interface configuration .PR1785058
-
VSTP BPDU are not properly processed by EX4300 when the interface is configured in SP style causing VSTP not to convergePR1849492
Routing Protocols
-
On all Junos and Junos OS Evolved platforms, multiple simultaneous Command Line Interface (CLI) sessions will lead to high Management Daemon (mgd) CPU utilization, impacting the device's reachability over the loopback interface from IS-IS nodes.PR1749850