[j-nsp] JUNOS 9,6 AND IS-IS ISSUE

[Mario Andres Rueda Jaimes]

Actually there are no RE switchover during the event, a remote equipment
goes Down and inmediately RPD crash on this equipment, this is the log
chronology for the issue

Mar 31 20:21:50  x2 /kernel: BAD_PAGE_FAULT: pid 4837 (rpd), uid 0: pc
0x826d91d got a read fault at 0xf, x86 fault flags = 0x4
Mar 31 20:22:02  x2 /kernel: pid 4837 (rpd), uid 0: exited on signal 11
(core dumped)
Mar 31 20:22:02  x2 rpd[7209]: L2CKT acquiring mastership for standby
Mar 31 20:22:02  x2 rpd[7209]: L2VPN acquiring mastership for standby
Mar 31 20:22:04  x2 rpd[7209]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
97 for route 10.10.200.1
Mar 31 20:22:04  x2 rpd[7209]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
91 for route 172.31.20.5
Mar 31 20:22:04  x2 rpd[7209]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
72 for route 192.168.8.5
Mar 31 20:22:04  x2 rpd[7209]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
72 for route 192.168.8.9
Mar 31 20:22:35  x2 dumpd: Core and context for rpd saved
in /var/tmp/rpd.core-tarball.4.tgz




El lun, 04-04-2011 a las 11:39 -0400, Stefan Fouant escribió:
> > -----Original Message-----
> > From: Mario Andres Rueda Jaimes [mailto:maeve2009 at gmail.com]
> > Sent: Monday, April 04, 2011 11:28 AM
> > To: Stefan Fouant
> > Cc: juniper-nsp at puck.nether.net
> > Subject: RE: [j-nsp] JUNOS 9,6 AND IS-IS ISSUE
> > 
> > Hi
> > 
> > Basicly these are the log lines prompted:
> > 
> > Mar 31 16:43:19  x2 rpd[4564]: L2CKT acquiring mastership for standby
> > Mar 31 16:43:19  x2 rpd[4564]: L2VPN acquiring mastership for standby
> > Mar 31 16:43:20  2 rpd[4564]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
> > 97
> > for route 10.10.200.1
> > Mar 31 16:43:20  x2 rpd[4564]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
> > 91 for route 172.31.20.5
> > Mar 31 16:43:20  x2 rpd[4564]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
> > 72 for route 192.168.8.5
> > Mar 31 16:43:20  x2 rpd[4564]: RPD_KRT_KERNEL_BAD_ROUTE: KRT: lost ifl
> > 72 for route 192.168.8.9
> 
> Is this just coincidental that there are Layer 2 VPN failures at precisely the same time as you have routing issues?  It also appears this might be related to an RE switchover.
> 
> Looks like you've got RPD indicating bad kernel routing table information due to the loss of interfaces and these happen around the same time that a possible switchover took place...
> 
> I don't think you've given us all the information.  Can you describe your set up and the events leading up to this in a bit more detail?
> 
> Stefan Fouant, CISSP, JNCIEx2
> www.shortestpathfirst.net
> GPG Key ID: 0xB4C956EC
>