Description

We have suffer logs that indicates the evo-pfemand process unresponsive.

We have restarted the evo-pfemand process and the device appears to be operating normally now.

The device functions as a PE in a service provider environment, and committing configuration changes is a routine part of customer operations.

Symptoms

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : BrcmPlusCosRateClassContainer:  func = int BrcmPlusCosRateClassContainer::deleteProfile(rateclassProfileId_t), Failed to get key from ProfileId: = 41

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : Internal:  int BrcmPlusSchedMapProfile::freeRateClassProfileId(int) = Failed to delete rate class , profile id: = 41

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : unbindRateClassProfile:  int BrcmPlusSchedMapProfile::unbindRateClassProfile(int, BrcmPlusCosIfdPtr, BrcmPlusCosIflPtr, BrcmPlusCosIflSetPtr, int, uint8_t) = Failed to free rate class on , Unit : = 0, Port : = 17, voqBaseId : = 2096, queue : =

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : resetBufferAndDropProfile:  int BrcmPlusSchedMapProfile::resetBufferAndDropProfile(BrcmPlusCosIfdPtr, BrcmPlusCosIflPtr, BrcmPlusCosIflSetPtr, int, uint8_t) = Failed to resetBufferAndDropProfile on , Unit : = 0, Port : = 17, queueNum: =

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : Unbind:  int BrcmPlusSchedMapProfile::hqosUnbind(uint32_t, uint32_t, uint32_t, int, uint64_t, bool) = Failed to resetBufferAndDropProfile on , unit : = 0, port: = 17, ifd Index: = 1339, ifl Index: = 0, iflset index : = 1003, smapId: = 493921241282, queueNum: =

Jan 28 15:20:03.602 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : hqosUnbindIflSet:  int BrcmPlusSchedMapProfile::hqosUnbindIflSet(uint32_t, uint32_t, uint64_t) = hqosUnbind Failed for iflSet , index: = 1003

Jan 28 15:20:03.603 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Cos : BrcmPlusCosIflSet:  func = int BrcmPlusTrafficControlProfile::unbindIflSet(uint32_t, uint64_t), failed to unbind smap: = 493921241282, for ifd index: = 1339, iflset: = 1003

Jan 28 15:20:03.603 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] BrcmPfe : BrcmPlusCos: unbindbindiflset failed forindex:1003 profileId:493921242261

Jan 28 15:20:03.603 2026 JTAC evo-pfemand[31323]: %USER-3-EVO_PFEMAND_ASIC_PROGRAM_FAILED: Failed to program the hardware with an error - entry doesn't exist in hardware

Jan 28 15:20:03.605 2026 JTAC evo-pfemand[31323]: %USER-3-EVO_PFEMAND_ASIC_PROGRAM_FAILED: Failed to program the hardware with an error - entry exist in hardware

Jan 28 15:20:03.605 2026 JTAC evo-pfemand[31323]: %USER-3: [t:31323] [Error] Pfe: Failed to configure ifl ae4014.2300

Jan 28 15:20:03.606 2026 JTAC evo-pfemand[31323]: %USER-4-EVO_PFEMAND_IFL_PROCESSING_FAILED: Failed to process installing for the logical interface ae4014.2300

 

show platform binding-queue incomplete

error: communication failure with /re0/evo-pfemand/

1. The producer app is either down or unresponsive

Run "show system processes node | grep " to check if app is running

Run "show system application app " to check the state of the app

2. If there is a change in mastership recently, please wait for the switchover to complete for the app to be online

Run "request chassis routing-engine master switch check" to check the status of switchover

error: communication failure with /re0/evo-pfemand/local-ps-stack

1. The producer app is either down or unresponsive

Run "show system processes node | grep " to check if app is running

Run "show system application app " to check the state of the app

2. If there is a change in mastership recently, please wait for the switchover to complete for the app to be online

Run "request chassis routing-engine master switch check" to check the status of switchover

Solution

Problem Description:

Port forwarding is not happening properly after interface activate/deactivate or flaps, for multiple times


Root Cause:

IFD VLAN domain or BRCM port class ID is colliding among physical IFDs and AEs with different pool

This is affecting forwarding in those ports


Fix Details:

Using allocator for IFD vlan domain for different kids of IFDs


This issue was analysed in internal PR1911281 & was identified to be fixed via PR1875544.

https://prsearch.juniper.net/problemreport/PR1875544

Modification History

2026-02-09 : Article Created