Table of Contents
Advertisement
JUNOS OS 10.4 Release Notes
58
configuration when the configuration is actively being used by subscribers can result
in an unpredictable behavior. [PR/512902]
An NTP server might not reply to clients with a source address that is explicitly
configured. [PR/540430]
The IPv6 BGP neighbors might not come back to the up state when an FPC associated
with that session is manually taken offline, removed, and re-inserted. [PR/552376]
No ICMP host redirect messages are generated when there are multiple VLANs
configured on an interface (multiple logical interfaces on a single physical interface).
[PR/559317]
When the same local link address is configured on two interfaces, the message "/kernel:
ip6_getpmtu: Invalid Stored MTU" is displayed continuously. [PR/560079]
When IPv6 packets have a size greater than 1232 bytes, the packets get fragmented.
[PR/571596]
After a few graceful Routing Engine switchover, the firewall filter applied on the
loopback interface might affect the internal control packets from the PICs to the
Routing Engine. The PICs might fail to come back online if the packets are blocked.
[PR/578049]
Routing Protocols
When aggregate interfaces are used for VPN applications, load balancing may not
occur with a Layer 2 circuit configuration. [PR/471935]
Under certain circumstances, the BGP path selection does not follow the local
preference. This might lead to incorrect BGP path selections. [PR/513233]
When an interface is added to a routing instance with rpf-check enabled, the routing
protocol process might crash if a route-distinguisher is also changed at the same time.
[PR/539321]
In Junos OS Release 10.0 and later, a direct route to a VRF with a rib-group is not
advertised as an inet-vpn route to the IBGP neighbor due to the error "BGP label
allocation failure: Need a nexthop address on LAN." [PR/552377]
In some cases, MX Series routers might not send the Link Layer Discovery Protocol
(LLDP) notification trap when the LLDP is disabled on the remote neighbor.
[PR/560855]
When a routing protocol process is restarted after a crash or a mastership switch, the
kernel and the reference counters for the routing protocol process flood branch next
hop might not be in sync anymore. The exposure is high in NGEN-MVPN with many
local receivers and constant churn of joins and prunes of multicast groups. The routing
protocol process might assert and restart while deleting a flooded next hop. As a
workaround, restart the system, or deactivate all MVPN instances to get the kernel
and the routing protocol process to be in sync upon a routing protocol process restart.
[PR/561127]
The 3D Packet Forwarding Engines might experience a rare transient error that
temporarily corrupts one of the lookup engines, resulting in packet loss. A set of
messages similar to the following is displayed:
Copyright © 2011, Juniper Networks, Inc.
Advertisement
Table of Contents
