On 7/19/2012 5:56 PM, Morgan McLean wrote: > So I actually don't use the FXP interface. I basically have four OSPF > connections coming into my edge firewall srx cluster, and I use the > loopback address advertised over OSPF to manage all of my devices. The > MX80's are the only ones that seem to have a problem...am I S.O.L if I'm > not using the FXP interface? > > Morgan
Not at all. Management and monitoring over the loopback (in-band) is a perfectly valid and workable configuration. Knowing nothing at all about the config on your gear, I would think that the first places to look for the source of intermittent failures would be route stability and RE firewalls/policers. You said that you get intermittent ping failures to the box. Can you ssh into the box reliably? Can you ping from the box reliably to the destination that has issues pinging to the box? ...and so on... -DMM > > On Wed, Jul 18, 2012 at 9:14 AM, Xu Hu <[email protected]> wrote: > >> Does the Juniper RE not the same as Cisco RSP. I think the control plane >> information all need to go to the RE, if RE had any issue, why the traffic >> don't have any issue? >> >> Thanks and regards, >> Xu Hu >> >> On 18 Jul, 2012, at 22:32, "OBrien, Will" <[email protected]> wrote: >> >>> Check your fxp0 configuration. You may be shipping return traffic out >> random interfaces... >>> We are leaning toward putting all production traffic inside a virtual >> routing instance/chassis and using the main routing instance just for >> management. >>> ________________________________________ >>> From: [email protected] [ >> [email protected]] on behalf of Morgan McLean [ >> [email protected]] >>> Sent: Wednesday, July 18, 2012 1:34 AM >>> To: [email protected] >>> Subject: [j-nsp] MX80 poor monitoring, packet loss to RE, SNMP not >> responding >>> I have a pair of MX80's that both are very unreliable in terms of trying >> to >>> monitor them. Any traffic destined to the RE, be it ICMP or SNMP seems to >>> be very hit or miss. Sometimes SNMP won't respond, pinging it gives me >>> maybe 50% loss on average, but it passes traffic fine. >>> >>> This causes issues with monitoring, false alerts, etc. I realize the >>> traffic destined for the RE is not as important, but the box is hardly >>> loaded and among maybe 50 other juniper devices I have, EX, SRX, only >> these >>> are giving me issues. >>> >>> Can anybody give me any insight? >>> >>> Thanks, >>> Morgan >>> _______________________________________________ >>> juniper-nsp mailing list [email protected] >>> https://puck.nether.net/mailman/listinfo/juniper-nsp >>> >>> _______________________________________________ >>> juniper-nsp mailing list [email protected] >>> https://puck.nether.net/mailman/listinfo/juniper-nsp > _______________________________________________ > juniper-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/juniper-nsp _______________________________________________ juniper-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/juniper-nsp
