Reset Search
 

 

Article

System memory depleted for process "dot1ag" caused unexpected reboot

« Go Back

Information

 
TitleSystem memory depleted for process "dot1ag" caused unexpected reboot
Symptoms
  • System memory depleted for process "dot1ag" caused unexpected reboot.
  • Configuring only CFM related configurations and verified the memory leak by using memory leak tool and found the switch memory is getting depleted slowly.
    If we disable the CFM domain, memory leak is not occurring. 

-> show debug system-dump   
===============================================
            Switch system dump information
===============================================
core_dump_info storage: 8/5120 used [EMPTY]
--------------------------------
System memory depleted.
time: Sat Oct 22 00:49:22 2016
Global Memory Stats:
                   MemTotal:       992380 kB
                    MemFree:         4016 kB
                    Buffers:          128 kB
                     Cached:         8112 kB
               Active(anon):       404616 kB
             Inactive(anon):       410132 kB
               Active(file):         3388 kB
             Inactive(file):         4608 kB
                  HighTotal:            0 kB
                   HighFree:            0 kB
                   LowTotal        992380 kB
                    LowFree:         4016 kB
               Committed_AS:      1244620 kB
Top Processes by RSS:
                    dot1ag:       743808 kB
                  cliMaster:        15476 kB
                        hal:         7456 kB

 

Switch logs:
-----------------
10/22/2016 00:52:40.27 <Info:HAL.Sys.Info> Starting hal initialization ....
10/22/2016 00:52:40.27 <Noti:DM.Notice> DM started
10/22/2016 00:52:40.00 <Noti:NM.StrtProc> The Node Manager (NM) has started processing.
10/22/2016 00:52:39.68 <Noti:EPM.start> EPM Started
10/22/2016 00:52:39.67 <Warn:EPM.UnexpctRebootDtect> Booting after System Failure.
10/22/2016 00:52:39.52 <Noti:EPM.wd_warm_reset> Changing to watchdog warm reset mode

10/12/2016 18:37:56.11 <Warn:BFD.RxPktInv> Received BFD an invalid packet on VLAN TCG51TCG61 Error:Invalid ttl value for single Hop session



 
Environment
  • Summit X460G2-24x-10G4
  • EXOS 15.6.2.12
  • EXOS 16.1.4.2
Cause
  • Configuring CFM domain causing Slow memory leak on process "dot1ag" leading to an unexpected reboot.
Resolution

Workaround:

  • We can disable the CFM domain to avoid memory leak/unexpected reboot.

Solution:

  • We have created a CR# xos0066156 to address/track the updates for this issue.

The CR fix will be scoped in following EXOS versions:

  1. EXOS 16.1.4 (Build) expected release date end of December 2016 
  2. EXOS 15.6.5 (Build) expected release date end of January 2017
  3. EXOS 15.7.5 (Build) expected release date not yet scoped (check with GTAC)
  4. EXOS 16.2.2 (Build) expected release date not yet scoped (check with GTAC)
  5. EXOS 22.2.1 (Build) expected release date not yet scoped (check with GTAC)
  6. EXOS 21.1.3 (Build) expected release date not yet scoped (check with GTAC)
  • Regarding timeline/target releases or more information please contact GTAC for assistance.
Additional notes

Feedback

 

Was this article helpful?


   

Feedback

Please tell us how we can make this article more useful.

Characters Remaining: 255