[rancid] Timeout Issue with RANCID Against Junos

[Mark Tinka]

Hi all.

I seem to be hitting a new issue that just came out of the
blue:

Trying to get all of the configs.
hostname: End of run not found
Error: TIMEOUT reached
hostname: End of run not found
Error: TIMEOUT reached
=====================================
Getting missed routers: round 1.
hostname: End of run not found
Error: TIMEOUT reached
hostname: End of run not found
Error: TIMEOUT reached
=====================================

RANCID version is 2.3.8 running on FreeBSD 9.2.

RANCID was working well all along and just started spewing
this (I did run a system update to patch the recent NTP
vulnerability, and that's about it).

It affects only Junos systems (IOS systems have no issue).

I've also noticed that it affects systems that have a higher
latency. Other Junos systems which are physically closer
seem to have no issue getting checked by RANCID. That said,
IOS systems that have a higher latency and much busier CPU
than the Junos one are all working fine.

At one point, the failing Junos router was successful, but
after that, all attempts for the same and other routers have
been unsuccessful.

Any hints would be most appreciated.

Cheers,

Mark.

[Mark Tinka]

Hi folks.

For the archives, after spending the night running tests,
this turned out to be an IPv6 MTU issue on one of the
transit nodes.

Fixed and everything is looking good again (still trying to
figure out why the issue was hidden for so long, given the
transit node had the wrong MTU for a while now).

Cheers,

Mark.

Post by Mark Tinka
Hi all.
I seem to be hitting a new issue that just came out of
Trying to get all of the configs.
hostname: End of run not found
Error: TIMEOUT reached
hostname: End of run not found
Error: TIMEOUT reached
=====================================
Getting missed routers: round 1.
hostname: End of run not found
Error: TIMEOUT reached
hostname: End of run not found
Error: TIMEOUT reached
=====================================
RANCID version is 2.3.8 running on FreeBSD 9.2.
RANCID was working well all along and just started
spewing this (I did run a system update to patch the
recent NTP vulnerability, and that's about it).
It affects only Junos systems (IOS systems have no
issue).
I've also noticed that it affects systems that have a
higher latency. Other Junos systems which are physically
closer seem to have no issue getting checked by RANCID.
That said, IOS systems that have a higher latency and
much busier CPU than the Junos one are all working fine.
At one point, the failing Junos router was successful,
but after that, all attempts for the same and other
routers have been unsuccessful.
Any hints would be most appreciated.
Cheers,
Mark.