Reset Search
 

 

Article

MLAG: Traffic loss temporarily seen when a module having ISC ports reboots later

« Go Back

Information

 
TitleMLAG: Traffic loss temporarily seen when a module having ISC ports reboots later
Symptoms
In the case that a BDX8 acts as an MLAG peer switch, the ISC ports and MLAG ports could be present on different slots.
When an MLAG switch reboots and the module having the ISC link boots up later than other modules having MLAG ports, undesirable traffic is temporarily seen until the ISC link completely comes up.

At L2 layer, since FDB tables of MLAG peers are out of sync, temporary flooding would occur and take up unnecessary bandwidth of ports down to hosts, eventually causing service impact until it stops. 
At L3 layer, ARP entries would not get resolved since some of ARP replies go to the other peer and stop there until the ISC link comes back up, resulting in traffic blackholing at the MLAG switch exhibiting high CPU utilization. 


User-added image
Environment
  • EXOS All
  • BlackDiamond X8
  • BlackDiamond 8K
Cause
Current MLAG implementation doesn’t take into account of I/O module boot-up timings in packet forwarding decisions.
Resolution
CR# xos0063961 has been created for this issue. Fix schedule is TBD. 
Additional notes

Feedback

 

Was this article helpful?


   

Feedback

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

Characters Remaining: 255