topic RE: Warn:BGP.RIBMgr.ProcStubMsgDataLost in ExtremeSwitching (EXOS/Switch Engine)

Warn:BGP.RIBMgr.ProcStubMsgDataLost

Sumanta_Ghosh — Thu, 03 Nov 2016 22:57:00 GMT

Hi Experts

What does the given error message mean? It shows up in logs.

******************************************************

Warn:BGP.RIBMgr.ProcStubMsgDataLost

An attempt to process an ATG_I3_INET_ADDR_IND has failed due to local
resource shortages or indication that the I3 join has lost some
information for the request
VR = 0x00000005
I3 join index = 1
Data lost flag value = 1
Local resource ret code = 0

RE: Warn:BGP.RIBMgr.ProcStubMsgDataLost

Ron_Huygens — Fri, 04 Nov 2016 15:27:00 GMT

Hi Sumanta,

What EXOS version are you running?
This can caused by a buffer issue when you have lots of iBGP peers that send updates and exceed the maximum number of routes per message.
As far as I can see that should not really impact the normal traffic.

RE: Warn:BGP.RIBMgr.ProcStubMsgDataLost

Sumanta_Ghosh — Fri, 04 Nov 2016 19:03:00 GMT

Hardware is X770 and OS is 21.1.2.14.

There are hardly 30 BGP peers, spread cross in three VRs.

RE: Warn:BGP.RIBMgr.ProcStubMsgDataLost

Ron_Huygens — Fri, 04 Nov 2016 19:16:00 GMT

Then I suggest that you open a case for this. The issue reported should be investigated more thoroughly than what is possible on this forum.

RE: Warn:BGP.RIBMgr.ProcStubMsgDataLost

Sumanta_Ghosh — Fri, 04 Nov 2016 21:30:00 GMT

Hi Ron

Thanks, I have also noticed the below behavior during redundancy testing.

When Switch A reboots, traffic fails over to switch B. When Switch A recovers, after the reboot, all the i-BGP sessions drop and then re-establish on Switch B, causing communication loss for 2-5 minutes. When Switch B reboots, traffic fails over to switch A. When Switch B recovers after reboot, all the i-BGP sessions on switch A drop then re-establish on Switch A.

This happens every single time.