Description

This article explains the meaning of the following syslog message and indicates the steps to overcome it.

fpcx INTR: throttle 60sec TLCHIP[1]:tl.edf.intr.status:page_mem_prog_err(0)

Symptoms

Syslog Message:

user@host> show log messages | match INTR:
May 30 06:33:16  fpc3  INTR: throttle 30sec TLCHIP[1]:tl.edf.intr.status:page_mem_prog_err(0): (Count:40)
Jun  1 02:04:33  fpc0  INTR: throttle 30sec TLCHIP[2]:tl.edf.intr.status:page_mem_prog_err(0): (Count:78)

Detailed Syslogs triggered on an FPC:

user@host> show log messages |match fpc5
Sep  1 20:58:56  fpc5  INTR: throttle 60sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:1) <-- Check the TLCHIP Is on which error is seen. Here, it is TLCHIP 0
Sep  2 08:15:01  fpc5  INTR: throttle 90sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:3)
Sep  2 09:57:24  fpc5  INTR: throttle 120sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:5)
Sep  2 11:45:20  fpc5  INTR: throttle 90sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:6)
Sep  3 05:30:21  fpc5  INTR: throttle 60sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:7)
Sep  3 08:47:02  fpc5  INTR: throttle 90sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:9)
Sep  3 10:28:11  fpc5  INTR: throttle 120sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:11)
Sep  3 12:07:31  fpc5  INTR: throttle 90sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:12)
Sep  3 12:38:17  fpc5  INTR: throttle 60sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:13)
Sep  3 13:04:46  fpc5  INTR: throttle 30sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:14)
Sep  3 15:18:07  fpc5  INTR: throttle 60sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:16)
Sep  3 22:58:11  fpc5  INTR: throttle 30sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:17)
Sep  3 23:28:35  fpc5  INTR: throttle 60sec  TLCHIP[0]:tl.edf.intr.status:page_mem_prog_err(0): (Count:18) <-- Count keeps increasing when each error is triggered

Solution

Egress Descriptor Fetch, EDF accesses descriptor based on next-hop provided by the ingress, calculates L2 and L3 length needed by counters and policers' L3 MTU check. The logs are cosmetic in nature and does not impact any services.

Check if there is any impact/active alarm matching the timestamp of this error. In case there is any impact/ Active alarm, contact JTAC for detailed analysis.

If the frequency of the alarm is low and does not log continuously, you can choose to ignore the message. 

If you want to get rid of the error, restart the FPC.

user@host> request chassis fpc slot 5 offline

<wait for 2 minutes>

[MASTER]
user@host> request chassis fpc slot 5 online

OR

[MASTER]
@pl-ktw01a-rc1> request chassis fpc slot 5 restart

 

For JTAC analysis, please collect the logs below from FPC 5:

  • user@host> request execute pfe command "show nvram" target fpc5
  • user@host> request execute pfe command "show syslog messages" target fpc5
  • user@host> request execute pfe command "show tlchip 0 intr_counts all" target fpc5 <-- For TLCHIP 0

Example:

tl.edf.intr.status
page_mem_prog_err(0) : 18    <-- Gives the count of error

user@host> request execute pfe command "show tlchip 1 intr_counts all" target fpc5  <-- For TLCHIP 1

user@host> request execute pfe command "show tlchip 0 edf config trapcodes" target fpc5
user@host> request execute pfe command "show tlchip 1 edf config trapcodes" target fpc5
user@host> request execute pfe command "show tlchip 0 edf stats periodic" target fpc5
user@host> request execute pfe command "show tlchip 1 edf stats periodic" target fpc5
user@host> request execute pfe command "bringup jspec read tlchip[0] register edf intr status" target fpc5
user@host> request execute pfe command "bringup jspec read tlchip[1] register edf intr status" target fpc5