Agreed, we had this happen on older M20s as well. Depending on your routing table, a quick bandaid fix might be to just clean up more specific prefixes if you can so that your forwarding table doesn't have as many entries. For example, I have had cases where we have a bunch of smaller internal routes in RR clients that really didn't need to be there as long as the RR server routers had the more specifics. More memory is certainly the long term solution though.
Regards, -Jeff ________________________________________ From: [email protected] [[email protected]] On Behalf Of [email protected] [[email protected]] Sent: Monday, January 19, 2009 10:35 AM To: [email protected] Cc: [email protected] Subject: Re: [j-nsp] M10i - %KERN-1-RT_PFE: RT msg op 1 (PREFIX ADD) failed, err 6 (No Memory) / RT msg op 3 (PREFIX CHANGE) failed, err 6 (No Memory) > i have a problem on one of our M10i. > System log continously shows following errors: > > Jan 19 15:59:03 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 3 (PREFIX > CHANGE) failed, err 5 (Invalid) > Jan 19 15:59:03 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 1 (PREFIX > ADD) failed, err 6 (No Memory) > Jan 19 15:59:07 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 3 (PREFIX > CHANGE) failed, err 6 (No Memory) > Jan 19 15:59:15 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 1 (PREFIX > ADD) failed, err 6 (No Memory) > Jan 19 15:59:23 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 3 (PREFIX > CHANGE) failed, err 5 (Invalid) > Jan 19 15:59:27 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 1 (PREFIX > ADD) failed, err 6 (No Memory) > Jan 19 15:59:27 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 3 (PREFIX > CHANGE) failed, err 5 (Invalid) > Jan 19 15:59:30 MYROUTER /kernel: %KERN-1-RT_PFE: RT msg op 1 (PREFIX > ADD) failed, err 6 (No Memory) > > But there is no traffic impact.... > I tried further investigation with some commands i found in older mailings. > But without success... > > u...@myrouter> start shell pfe network cfeb0 > vty: connect: Connection refused This probably doesn't work precisely because there is insufficient CFEB memory. We have seen exactly this happen. Sooner or later it probably *will* impact traffic. > I could imagine that a reload will clear the problem, but i would like > to avoid it... A reload is probably the only fix. And you *really* want to monitor your CFEB memory utilization (show chassis cfeb). We are in the process of upgrading all our M7i/M10i CFEBs to 256 MB. Steinar Haug, Nethelp consulting, [email protected] _______________________________________________ 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
