This is what I'm seeing on my Cisco NCS edge routers in case it's helpful to anybody: RP/0/RP0/CPU0:Oct 7 14:49:10.478 CDT: bgp[1089]: %ROUTING-BGP-3-MALFORM_UPDATE : Malformed UPDATE message received from neighbor 2001:504:27::d1af:0:2 (VRF: default) - message length 171 bytes, error flags 0x00000100, action taken "TreatAsWdr". Error details: "Error 0x00000100, Field "Attr-length", Attribute 35 (Flags 0xe0, Length 1024), Data [f0230400]". NLRIs: [IPv6 Unicast] 2402:2de0::/32 RP/0/RP0/CPU0:Oct 7 15:18:11.888 CDT: bgp[1089]: %ROUTING-BGP-3-MALFORM_UPDATE : Malformed UPDATE message received from neighbor 2001:504:27::d1af:0:1 (VRF: default) - message length 171 bytes, error flags 0x00000100, action taken "TreatAsWdr". Error details: "Error 0x00000100, Field "Attr-length", Attribute 35 (Flags 0xe0, Length 1024), Data [f0230400]". NLRIs: [IPv6 Unicast] 2402:2de0::/32 The router is discarding the malformed update so I'm not seeing any actual issues based on what appears to be a correct RFC7606 error handling implementation by IOS-XR. *Nathan Beard* *7Sigma Systems, Inc.* Senior Network Engineer (218) 820-7636 www.7sigma.com On Tue, Oct 7, 2025 at 4:03 PM Jay Hanke via mice-discuss < mice-discuss@lists.micemn.net> wrote:
Looks like bird 1.x isn't affected 2.x is.
Initial indications are that the route with the malformed attribute is coming from the philippines.
On Tue, Oct 7, 2025 at 3:46 PM Andrew Hoyos <hoyosa@gmail.com> wrote:
Similarly:
Oct 7 20:43:21 rpd[32393]: bgp_read_v4_update:12939: NOTIFICATION sent
to 2001:504:27::d1af:0:2 (External AS 53679): code 3 (Update Message Error) subcode 1 (invalid attribute list)
We do have bgp-error-tolerance enabled, however, so in theory, this
should throw out update itself, not teardown session.
We’re actually seeing BFD die to route servers as well.
Any commonality w/ the IX’s below and route server software? We’re only seeing this on MICE, and v6, FWIW.
Andrew Hoyos hoyosa@gmail.com
On Oct 7, 2025, at 15:40, Mark Kane <markk@honeycomb.net> wrote:
FWIW, we saw this from rs1/rs2 on a Cisco:
Oct 7 14:49:10 CDT: bgp[1087]: %ROUTING-BGP-3-MALFORM_UPDATE :
Malformed UPDATE message received from neighbor 2001:504:27::d1af:0:2 (VRF: default) - message length 171 bytes, error flags 0x00000100, action taken "TreatAsWdr". Error details: "Error 0x00000100, Field "Attr-length", Attribute 35 (Flags 0xe0, Length 1024), Data [f0230400]". NLRIs: [IPv6 Unicast] 2402:2de0::/32
Oct 7 15:18:11 CDT: bgp[1087]: %ROUTING-BGP-3-MALFORM_UPDATE : Malformed UPDATE message received from neighbor 2001:504:27::d1af:0:1 (VRF: default) - message length 171 bytes, error flags 0x00000100, action taken "TreatAsWdr". Error details: "Error 0x00000100, Field "Attr-length", Attribute 35 (Flags 0xe0, Length 1024), Data [f0230400]". NLRIs: [IPv6 Unicast] 2402:2de0::/32
Both sessions have remained up and appear normal.
-----Original Message----- From: Jay Hanke via mice-discuss <mice-discuss@lists.micemn.net> Sent: October 7, 2025 15:34 To: Jonathan Stewart <jonathan@les.net> Cc: mice-discuss@lists.micemn.net Subject: Re: v6 on rs1/rs2 fubar?
Confirmed at FLIX/CIX-ATL/PIT-IX so far.
All juniper but not all junipers affected
If someone affected can you send me a trace it would be helpful.
On Tue, Oct 7, 2025 at 3:32 PM Jonathan Stewart <jonathan@les.net> wrote:
I've seen BGP sessions resetting this afternoon on Juniper.
What more do we know from the global community?
Jonathan Stewart Network Engineer LES.NET - AS18451 Desk: 1-204-666-6191 Mobile: 1-204-990-2120 130 Portage Avenue E Winnipeg, MB R3C 0A1 CANADA
On Tue, 2025-10-07 3:11 p.m., Jay Hanke via mice-discuss wrote:
Looks like it might be a worldwide issue.
Running Juniper?
On Tue, Oct 7, 2025 at 3:07 PM Andrew Hoyos <hoyosa@gmail.com>
wrote:
Hi folks,
We dropped v6 to rs1/rs2. Spot checked a few other members we
manage, and same deal.
Any work taking place on route servers at the moment?
Andrew Hoyos hoyosa@gmail.com
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net
_______________________________________________ mice-discuss mailing list -- mice-discuss@lists.micemn.net To unsubscribe send an email to mice-discuss-leave@lists.micemn.net