Bug 1965 – restrictive assert in ECMP code

Bug 1965 - restrictive assert in ECMP code


Summary:	restrictive assert in ECMP code

Status:	CONFIRMED

Product:	ns-3
Classification:	Unclassified
Component:	global-routing
Version:	pre-release
Hardware:	PC Linux

Importance:	P5 enhancement
Assigned To:	Tom Henderson

URL:
Keywords:

Depends on:
Blocks:
	Show dependency tree / graph

Reported:	2014-08-26 14:09 EDT by Tom Henderson
Modified:	2014-08-26 14:12 EDT (History)
CC List:	1 user (show)

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Tom Henderson 2014-08-26 14:09:54 EDT

Tommaso found that in the ipv4 global routing test suite, replacing the link n0->n2 with anything other than point-to-point causes this assert to fire:

SPFVertex::NodeExit_t
SPFVertex::GetRootExitDirection () const
{
  NS_LOG_FUNCTION (this);

  NS_ASSERT_MSG (m_ecmpRootExits.size () <= 1, "Assumed there is at most one exit from the root to this vertex");
  return GetRootExitDirection (0);
}

This was added with the bug 667 ECMP patch many years ago.  The assert may be too restrictive.

I'm filing this as a reminder to myself to investigate and fix.