Open Issues

Back port issue due to different size. PR1633344

The timingd application cannot be successfully restarted on ACX7100-48L:ACX7100-32C with G.8275.1 profile configuration. If there are PTP streams configured and there is a timingd restart due to timingd core file or requested application restart, the PTP streams might not be restored correctly, and/or changes to the PTP stream configuration after the restart might result in PTP streams that do not forward data packets correctly. It is very likely that a restart always affect the streams.

PR1597120

With IGMPv3 reports received at a higher rate more than 1600 pps, packets drop due to control plane rate limit. Hence, it is not possible to form 256,000 IGMP groups. You must change the DDoS configuration to reach 256,000 IGMP groups. PR1599998

Given the higher number of internal ports, two vendor's Q2C ASICs and eight FPCs supporting 20 ports (maximum), the statistics thread is expected to consume system resources. The current statistics collection, which is done in the evo-pfemand application needs to be refactored, so that the statistics thread does not end up consuming the system. PR1603899

PTP to PTP noise transfer fails for frequencies 1. 0.03125 HZ and 2. 0.123125 HZ. PR1608786

syncE to PTP and syncE to 1pps transient response marginally fails. This happens when the servo gets the initial 100ns jump in one measurement window and the next 100ns in the next measurement window adjusting less initially. PR1608934

PTP to PTP noise transfer fails for frequency 0.03125 HZ. PR1611838

syncE to PTP and syncE to 1pps noise transfer tests fails for frequencies 1. 0.00781 HZ, 2. 0.01563 HZ, 3. 0.03125 HZ, 4. 0.06156 HZ, and 5. 0.12313 HZ. PR1611911

MAC statistics for lacp on lag interfaces do not display the multicast packets. As a workaround, instead of querying the MAC statistics under the aggregated Ethernet interface, query for the actual physical interface created under the aggregated Ethernet interface. PR1612105

Journalctl logs do not print for evo-pfemand and packetio-brcm. You must use journalctl -t appname instead of journalctl -u appname. PR1621009

No details of auto negotiation settings are available on 1GE interfaces. PR1621991

DHCP transit unicast packets get dropped if you do not configure the interface in the DHCP relay stanza. PR1626759

On ACX7100-32C router, the 400G-ZR optic transceiver firmware upgrade failure occurs. This is a generic issue faced across all single RU platforms supporting QDD-400G-ZR. PR1626882

After the picd or rpdagent application restarts multiple object-info anomalies for the evo-pfemand process, the following anomalies get generated:

While sending traffic from both cores/units, all (16) VOQs/connectors are asking credits from same port. As a result credits are distributed as per weight ratio and flows asking for higher credits get higher bandwidth than others. PR1611028

In certain segment routing topologies having routers with varying CPU capabilities/load, higher convergence time might occur for a switchover event due to micro-loops. This is a known behavior when the one node converges faster than the neighboring node. The best solution for this is to fine tune the network topology using the micro-loop avoidance statement. PR1621263

With combination of triggers like restart rpd and fpc, route object leaks in the Packet Forwarding Engine. PR1641947

On ACX7509, PCI device misses alarm raises for all FPCs from the backup Routing Engine. There are no functional impact. PR1627348

If the system gets fully scaled across features and firewall also gets scaled. The CPU consumptionmight be more for a small window of around 5 seconds after every 18 seconds. The evo-pfemand process might become busy collecting the scaled firewall stats for that 5 seconds window and any other applications like pfe-cli might fail to execute commands. PR1629342

On ACX7100-32C and ACX7509 routers, 4x100g channels do not come up after multiple iterations Tx lase disabled alarm on. PR1631193

In a working and non-working logs, l2d index is different for vrrp group number 187. This is the same group for which packet gets dropped out of 400 groups. Other groups works as expected. There is some fix which went between working and NKWR related to l2dId which has exposed VRRP issue. Both VRRP MAC and interface MAC gets stored in SLU my_mac_hash table. For finding hash index for vrrp MAC, use l2dId, protocol type and vrrp group number as a key. In a non-working scenario, there is a collision between interface MAC and vrrp MAC on same hash index. Ideally hash movement should have happened to address collision but somehow it is not properly done. PR1633986

On ACX7509 routers and in a scaled setup, if you restart FPC, FPC becomes nonresponsive in the Online after ungraceful FPC OIR. PR1633117

On ACX7100 and ACX7509 routers, few control packets gets forwarded on ERPS discarding port and causes traffic loop. You must avoid the port blocking on local node. PR1641454

On ACX7100 and ACX7509 routers, OAM link fault management (LFM) Discovery state is not correct for some interfaces and discovery state is either Active Send Local or Fault. PR1651580

On ACX7100-48L devices, the G.8275.1- G.8273.2 1PPS cTE performance test might be marginally outside class-C for PTP BC, especially for the mixed speed port testing with combinations of 10G or 25G channelized ports and 100G ports. On each reboot, the 1PPS cTE measurement might be within the class-C measurement threshold or randomly be out of threshold by a few nanoseconds. PR1607381

On ACX7100-48 and ACX7024 devices, enabling or disabling of PTP TC or BC causes all interfaces to flap one time. PR1609927

PTP performance might randomly get wrose for around 30 minutes. The IDT DPLL firmware must be updated to version 4.8.15 to solve the issue. PR1614309

On ACX7509 devices, after multiple FPC online/offline, FPCs goes in to the Fault state. PR1616227

On ACX7100-32C devices, the G.8275.1- G.8273.2 1PPS cTE performance test might be outside class-C when using channelized 10G ports for PTP BC . On each reboot, the 1PPS cTE measurement might be within the class-C measurement threshold or randomly be out of it by a few nanoseconds. PR1629819

On ACX7509 devices, during ungraceful removal of FEB, the RCB crash or vmcore might occur. RCB comes back up after reboot. PR1636780

Due to software design limitation, support is available for only 64 member links in one aggregated Ethernet interface. From ASIC, there is no limitation. PR1627951

On ACX7509 routers, IS-IS, LPD, and BFD protocols flaps during graceful switchover while testing ldp oam. PR1638882

On ACX7100-32C routers, traffic does not flow after deleting or adding the VLAN configuration with load override. PR1647853