Attribute Length Error today
Anyone else see this today with both MICE route reflectors? Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:44:36 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:45:23 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Frank
On Fri, Dec 08, 2017 at 10:10:00PM +0000, Frank Bulk wrote:
Anyone else see this today with both MICE route reflectors?
Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:44:36 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:45:23 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED)
Which AS are you? No, but at the same time on the BIRD side we saw 2017-12-08 13:39:55 <RMT> NASN_26794: Received: Invalid attribute length: 400200400304ce6c 2017-12-08 13:39:55 <RMT> NASN_54578: Received: Invalid attribute length: 400200400304ce6c 2017-12-08 13:39:55 <RMT> NASN_32609: Received: Malformed AS_PATH: 400200 2017-12-08 13:39:55 <RMT> NASN_22402: Received: Invalid attribute length: 400200400304ce6c 2017-12-08 13:39:55 <RMT> NASN_16842: Received: Malformed AS_PATH: 400200 2017-12-08 13:39:55 <RMT> NASN_10242: Received: Malformed AS_PATH: 400200 Which in the past meant somebody had a cisco, and wasn't doing no bgp enforce-first-as as BIRD doesn't insert the MICE AS into the AS path, and that confuses Cisco devices. -- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
Yup Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8 ________________________________________ From: MICE Discuss [MICE-DISCUSS@LISTS.IPHOUSE.NET] on behalf of Frank Bulk [fbulk@MYPREMIERONLINE.COM] Sent: Friday, December 08, 2017 4:10 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: [MICE-DISCUSS] Attribute Length Error today Anyone else see this today with both MICE route reflectors? Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:38:56 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:39:47 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:39:49 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:41:28 192.168.0.85 BGP: Peer 206.108.255.1 DOWN (Attribute Length Error) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Dec 8 13:41:36 192.168.0.85 BGP: Peer 206.108.255.2 DOWN (Attribute Length Error) Dec 8 13:44:36 192.168.0.85 BGP: Peer 206.108.255.1 UP (ESTABLISHED) Dec 8 13:45:23 192.168.0.85 BGP: Peer 206.108.255.2 UP (ESTABLISHED) Frank
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session. Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0 Looks like Minnesota WiFi should check on their route announcements. HTH, Andy
Hey all, I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks My cell is 507-261-7690 <(507)%20261-7690> My email to consultant: Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE. https://forum.mikrotik.com/viewtopic.php?t=55798 https://forum.mikrotik.com/viewtopic.php?t=43498 Consultant response: Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic. On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote:
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session.
Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0
Looks like Minnesota WiFi should check on their route announcements.
HTH, Andy
-- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi <http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
In looking at it, do you happen to know why you were sending a default route? Has that been corrected? From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:16 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today Hey all, I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks My cell is 507-261-7690 My email to consultant: Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE. https://forum.mikrotik.com/viewtopic.php?t=55798 https://forum.mikrotik.com/viewtopic.php?t=43498 Consultant response: Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic. On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote: On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session. Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0 Looks like Minnesota WiFi should check on their route announcements. HTH, Andy -- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi Like us on Facebook To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
Looking at route filters right now and that I attached, I don't see that I'm advertising a default route unless I'm looking in the wrong spot. I did not setup any portion of the BGP as I know I'm not an expert or comfortable setting it up, thus why I hired a consultant to do it for me. They're a big Mikrotik consultant from Missouri. I only see us advertising 204.73.77.0/24, 162.255.252.0/22 in our route filters right now that I sent. On Fri, Dec 8, 2017 at 5:29 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote:
In looking at it, do you happen to know why you were sending a default route? Has that been corrected?
*From:* MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] *On Behalf Of *Darin Steffl *Sent:* Friday, December 08, 2017 5:16 PM *To:* MICE-DISCUSS@LISTS.IPHOUSE.NET *Subject:* Re: [MICE-DISCUSS] Attribute Length Error today
Hey all,
I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks
My cell is 507-261-7690 <(507)%20261-7690>
My email to consultant:
Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE.
https://forum.mikrotik.com/viewtopic.php?t=55798
https://forum.mikrotik.com/viewtopic.php?t=43498
Consultant response:
Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic.
On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote:
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session.
Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0
Looks like Minnesota WiFi should check on their route announcements.
HTH, Andy
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
<http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
-- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi <http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
I do not know much about how Mikrotik works. I agree that the filters do not look like you are sending default, I was just seeing it from Andy's post. From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:39 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today Looking at route filters right now and that I attached, I don't see that I'm advertising a default route unless I'm looking in the wrong spot. I did not setup any portion of the BGP as I know I'm not an expert or comfortable setting it up, thus why I hired a consultant to do it for me. They're a big Mikrotik consultant from Missouri. I only see us advertising 204.73.77.0/24, 162.255.252.0/22 in our route filters right now that I sent. On Fri, Dec 8, 2017 at 5:29 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote: In looking at it, do you happen to know why you were sending a default route? Has that been corrected? From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:16 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today Hey all, I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks My cell is 507-261-7690 My email to consultant: Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE. https://forum.mikrotik.com/viewtopic.php?t=55798 https://forum.mikrotik.com/viewtopic.php?t=43498 Consultant response: Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic. On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote: On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session. Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0 Looks like Minnesota WiFi should check on their route announcements. HTH, Andy -- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi Like us on Facebook To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 -- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi Like us on Facebook To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
It's possible it was sending default earlier and our consultant fixed it when he logged in between 1:27pm and 1:45pm. If someone wants to verify, that'd be great. On Fri, Dec 8, 2017 at 5:48 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote:
I do not know much about how Mikrotik works. I agree that the filters do not look like you are sending default, I was just seeing it from Andy's post.
*From:* MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] *On Behalf Of *Darin Steffl *Sent:* Friday, December 08, 2017 5:39 PM
*To:* MICE-DISCUSS@LISTS.IPHOUSE.NET *Subject:* Re: [MICE-DISCUSS] Attribute Length Error today
Looking at route filters right now and that I attached, I don't see that I'm advertising a default route unless I'm looking in the wrong spot. I did not setup any portion of the BGP as I know I'm not an expert or comfortable setting it up, thus why I hired a consultant to do it for me. They're a big Mikrotik consultant from Missouri.
I only see us advertising 204.73.77.0/24, 162.255.252.0/22 in our route filters right now that I sent.
On Fri, Dec 8, 2017 at 5:29 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote:
In looking at it, do you happen to know why you were sending a default route? Has that been corrected?
*From:* MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] *On Behalf Of *Darin Steffl *Sent:* Friday, December 08, 2017 5:16 PM *To:* MICE-DISCUSS@LISTS.IPHOUSE.NET *Subject:* Re: [MICE-DISCUSS] Attribute Length Error today
Hey all,
I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks
My cell is 507-261-7690 <(507)%20261-7690>
My email to consultant:
Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE.
https://forum.mikrotik.com/viewtopic.php?t=55798
https://forum.mikrotik.com/viewtopic.php?t=43498
Consultant response:
Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic.
On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote:
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session.
Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0
Looks like Minnesota WiFi should check on their route announcements.
HTH, Andy
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
<http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
<http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
------------------------------
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
-- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi <http://www.facebook.com/minnesotawifi> Like us on Facebook <http://www.facebook.com/minnesotawifi>
The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff. On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening. And yes, "set-bgp-prepend=0” on the inbound filter on the session is the correct way to handle a similar behavior to ‘no bgp enforce first-as’ in Tik world. This causes the MICE ASN to be removed from the AS Path. -- Andrew Hoyos hoyosa@gmail.com
On Dec 8, 2017, at 5:55 PM, Darin Steffl <darin.steffl@MNWIFI.COM> wrote:
It's possible it was sending default earlier and our consultant fixed it when he logged in between 1:27pm and 1:45pm. If someone wants to verify, that'd be great.
On Fri, Dec 8, 2017 at 5:48 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote: I do not know much about how Mikrotik works. I agree that the filters do not look like you are sending default, I was just seeing it from Andy's post.
From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:39 PM
To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
Looking at route filters right now and that I attached, I don't see that I'm advertising a default route unless I'm looking in the wrong spot. I did not setup any portion of the BGP as I know I'm not an expert or comfortable setting it up, thus why I hired a consultant to do it for me. They're a big Mikrotik consultant from Missouri.
I only see us advertising 204.73.77.0/24, 162.255.252.0/22 in our route filters right now that I sent.
On Fri, Dec 8, 2017 at 5:29 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote:
In looking at it, do you happen to know why you were sending a default route? Has that been corrected?
From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:16 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
Hey all,
I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks
My cell is 507-261-7690
My email to consultant:
Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE.
https://forum.mikrotik.com/viewtopic.php?t=55798
https://forum.mikrotik.com/viewtopic.php?t=43498
Consultant response:
Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic.
On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote:
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session.
Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0
Looks like Minnesota WiFi should check on their route announcements.
HTH, Andy
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
-- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
The route servers are dropping it, however I wonder if the error is being generated in the process of dropping it. -----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Andrew Hoyos Sent: Friday, December 08, 2017 6:03 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff. On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening. And yes, "set-bgp-prepend=0” on the inbound filter on the session is the correct way to handle a similar behavior to ‘no bgp enforce first-as’ in Tik world. This causes the MICE ASN to be removed from the AS Path. -- Andrew Hoyos hoyosa@gmail.com
On Dec 8, 2017, at 5:55 PM, Darin Steffl <darin.steffl@MNWIFI.COM> wrote:
It's possible it was sending default earlier and our consultant fixed it when he logged in between 1:27pm and 1:45pm. If someone wants to verify, that'd be great.
On Fri, Dec 8, 2017 at 5:48 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote: I do not know much about how Mikrotik works. I agree that the filters do not look like you are sending default, I was just seeing it from Andy's post.
From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:39 PM
To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
Looking at route filters right now and that I attached, I don't see that I'm advertising a default route unless I'm looking in the wrong spot. I did not setup any portion of the BGP as I know I'm not an expert or comfortable setting it up, thus why I hired a consultant to do it for me. They're a big Mikrotik consultant from Missouri.
I only see us advertising 204.73.77.0/24, 162.255.252.0/22 in our route filters right now that I sent.
On Fri, Dec 8, 2017 at 5:29 PM, Jeremy Lumby <jlumby@mnvoip.com> wrote:
In looking at it, do you happen to know why you were sending a default route? Has that been corrected?
From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Darin Steffl Sent: Friday, December 08, 2017 5:16 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
Hey all,
I sent the following to my Mikrotik consultant who set this up for me and his response is below. Anyone familiar with Mikrotik that knows the easy way to do this? We have MICE filters setup but not sure the correct way to do this without messing anything up further. He had the route filters out of order in our Mikrotik which is why I wasn't seeing MICE traffic. Once he fixed that, we're now receiving 500+ mbps from MICE and its stable on our end. But if we're still sending incorrect route announcements, I want to fix that so I don't cause issues. I found two related forum posts on how to change this in my router but I have a couple MICE filters as shown in the attachment. Thanks
My cell is 507-261-7690
My email to consultant:
Under the technical page of their website http://micemn.net/technical.html I see the following note "Note that your BGP sessions to the route servers likely need no bgp enforce-first-as or similar. The route servers do not add the MICE AS to the AS path." A quick search of Mikrotik forums shows that the equivalent to this setting is to "set-bgp-prepend=0" for the MICE peer. I'm not sure of this so want clarification to see if we need to make this change in order for traffic to start going through MICE.
https://forum.mikrotik.com/viewtopic.php?t=55798
https://forum.mikrotik.com/viewtopic.php?t=43498
Consultant response:
Your MICE connection should have had traffic assuming that you were advertising your prefixes to them. Their BGP session should not add their AS, so that should not matter. You should not need to change that, I would have to look to see why you are not seeing traffic.
On Fri, Dec 8, 2017 at 4:55 PM, Andy Koch <akoch@wiscnet.net> wrote:
On 08 December 2017 at 16:15:59, Justin Krejci wrote:
Yup
Dec 8 13:41:40.262 CST: %BGP-3-NOTIFICATION: sent to neighbor 206.108.255.1 3/11 (invalid or corrupt AS path) 3 bytes 400200 Dec 8 13:41:40.262 CST: BGP: 206.108.255.1 Bad attributes FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF 002D 0200 0000 0E40 0101 0040 0200 4003 04CE 6CFF 5918 CC49 4D16 A2FF E8
From this UPDATE dump, this parser helps quite a bit to decode what is going on: http://bgpaste.convergence.cx/
Looks like the AS_PATH was 0 bytes long, which is invalid and Frank's router did the only thing it knows to do on BGP errors - drop the session.
Further from the decode: The NEXT_HOP was 206.108.255.89 And the Routes were 204.73.77.0/24, 162.255.252.0/22 and 0.0.0.0/0
Looks like Minnesota WiFi should check on their route announcements.
HTH, Andy
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
--
Darin Steffl
Minnesota WiFi
www.mnwifi.com
507-634-WiFi
Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
-- Darin Steffl Minnesota WiFi www.mnwifi.com 507-634-WiFi Like us on Facebook
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote:
The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route. I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly. -- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
On Dec 8, 2017, at 6:14 PM, Doug McIntyre <merlyn@IPHOUSE.NET> wrote:
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote:
The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route.
Cool.
I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
Gotcha, while extra work, would it be easy enough to add first hop AS path filters on sessions on the route servers? Or do these already exist as well? Without testing, I’m not sure if that even would have prevented this specific scenario, but it seems like one we should prevent in the future, somehow. -- Andrew Hoyos hoyosa@gmail.com
If the issue was that the AS PATH was zero length, can BIRD filter those out? Frank -----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Doug McIntyre Sent: Friday, December 8, 2017 6:15 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote:
The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route. I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly. -- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too. -- Richard
On Dec 8, 2017, at 23:04, Frank Bulk <fbulk@MYPREMIERONLINE.COM> wrote:
If the issue was that the AS PATH was zero length, can BIRD filter those out?
Frank
-----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Doug McIntyre Sent: Friday, December 8, 2017 6:15 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote: The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route.
I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly.
-- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
it's time for secure route servers. Job sent out a list of ixp and quite a few of them are already doing rpki or irr filtering. On Dec 8, 2017 11:06 PM, "Richard Laager" <rlaager@wiktel.com> wrote:
Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too.
-- Richard
On Dec 8, 2017, at 23:04, Frank Bulk <fbulk@MYPREMIERONLINE.COM> wrote:
If the issue was that the AS PATH was zero length, can BIRD filter those out?
Frank
-----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Doug McIntyre Sent: Friday, December 8, 2017 6:15 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote: The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route.
I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly.
-- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
On Dec 9, 2017, at 7:00 PM, Jason Hanke <jayhanke@NEUTRALPATH.NET> wrote:
it's time for secure route servers. Job sent out a list of ixp and quite a few of them are already doing rpki or irr filtering.
I second this. IXPmanager makes this easy.
On Dec 8, 2017 11:06 PM, "Richard Laager" <rlaager@wiktel.com> wrote: Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too.
-- Richard
On Dec 8, 2017, at 23:04, Frank Bulk <fbulk@MYPREMIERONLINE.COM> wrote:
If the issue was that the AS PATH was zero length, can BIRD filter those out?
Frank
-----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Doug McIntyre Sent: Friday, December 8, 2017 6:15 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote: The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route.
I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly.
-- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
Here’s a tweet to the list that Jay is referring to: https://twitter.com/JobSnijders/status/939280110607327232 Frank From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET] On Behalf Of Jason Hanke Sent: Saturday, December 9, 2017 7:01 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET Subject: Re: [MICE-DISCUSS] Attribute Length Error today it's time for secure route servers. Job sent out a list of ixp and quite a few of them are already doing rpki or irr filtering. On Dec 8, 2017 11:06 PM, "Richard Laager" <rlaager@wiktel.com<mailto:rlaager@wiktel.com>> wrote: Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too. -- Richard
On Dec 8, 2017, at 23:04, Frank Bulk <fbulk@MYPREMIERONLINE.COM<mailto:fbulk@MYPREMIERONLINE.COM>> wrote:
If the issue was that the AS PATH was zero length, can BIRD filter those out?
Frank
-----Original Message----- From: MICE Discuss [mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET<mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET>] On Behalf Of Doug McIntyre Sent: Friday, December 8, 2017 6:15 PM To: MICE-DISCUSS@LISTS.IPHOUSE.NET<mailto:MICE-DISCUSS@LISTS.IPHOUSE.NET> Subject: Re: [MICE-DISCUSS] Attribute Length Error today
On Fri, Dec 08, 2017 at 06:03:21PM -0600, Andrew Hoyos wrote: The more important question - why didn’t the route servers drop that? I’d assume there should be inbound filters to drop bogons+default+$otherbadstuff.
They do have filters for bogons + default route.
I suspect bad AS attribute processing is part of what made it get leaked onwards. The BIRD servers were logging that as well during this period.
On a larger scale, this sort of thing begs the question - do we need to have folks in some sort of isolated VLAN with test sessions to the route servers upon turnup? SIX does this, as well as others, I suspect to prevent these exact issues from happening.
Possibly.
-- Doug McIntyre <merlyn@iphouse.net<mailto:merlyn@iphouse.net>> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
________________________________ To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1
On Fri, Dec 08, 2017 at 11:06:50PM -0600, Richard Laager wrote:
Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too.
BIRD is probably the best tool for the job for doing the import BGP filtering for conditions based on either AS Path = 0 or start with their AS. It sounds like people want the BGP import function to change to include something like function route_import(int asn) { .... if bgp_path.first != asn then return false; if bgp_path.len > 25 then return false; if bgp_next_hop != from then return false; return true; } Are there other conditions that should be filtered on? (this is taken mostly from Ondřej Surý's examples). As for doing RPKI, it looks like BIRD v2.0 supports that in some fashion, but that would have to be tested out in a lab. How many members would have an ROA already? -- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
This is the tool we're using at IX Denver: https://arouteserver.readthedocs.io/en/latest/index.html RPKI is currently turned off at IX Denver. Doug,If you're interested then I'll unicast you the output of the script. It's a bit large for the list. On Wed, Dec 13, 2017 at 2:46 PM, Doug McIntyre <merlyn@iphouse.net> wrote:
On Fri, Dec 08, 2017 at 11:06:50PM -0600, Richard Laager wrote:
Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too.
BIRD is probably the best tool for the job for doing the import BGP filtering for conditions based on either AS Path = 0 or start with their AS. It sounds like people want the BGP import function to change to include something like
function route_import(int asn) { .... if bgp_path.first != asn then return false; if bgp_path.len > 25 then return false; if bgp_next_hop != from then return false; return true; }
Are there other conditions that should be filtered on? (this is taken mostly from Ondřej Surý's examples).
As for doing RPKI, it looks like BIRD v2.0 supports that in some fashion, but that would have to be tested out in a lab. How many members would have an ROA already?
-- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
-- Jay Hanke CTO Neutral Path Communications 3 Civic Center Plaza, Suite 204 Mankato, MN 56001 (507) 327-2398 mobile jayhanke@neutralpath.net www.neutralpath.net
On 12/13/2017 02:46 PM, Doug McIntyre wrote:
if bgp_path.first != asn then return false; if bgp_path.len > 25 then return false; if bgp_next_hop != from then return false;
While I am not familiar with BIRD's syntax, if I am reading these correctly, they seem like reasonable conditions. -- Richard
Akamai has no plans to publish ROAs at this time. The reality is that in its current state RPKI is a high overhead method for explaining to the world which origin ASN to prepend in order to more convincingly hijack routes. Owen
On Dec 13, 2017, at 12:46 , Doug McIntyre <merlyn@IPHOUSE.NET> wrote:
On Fri, Dec 08, 2017 at 11:06:50PM -0600, Richard Laager wrote:
Zero length is a subset of “doesn’t start with their AS”, so if we filter on that condition, hopefully we will catch that too.
BIRD is probably the best tool for the job for doing the import BGP filtering for conditions based on either AS Path = 0 or start with their AS. It sounds like people want the BGP import function to change to include something like
function route_import(int asn) { .... if bgp_path.first != asn then return false; if bgp_path.len > 25 then return false; if bgp_next_hop != from then return false; return true; }
Are there other conditions that should be filtered on? (this is taken mostly from Ondřej Surý's examples).
As for doing RPKI, it looks like BIRD v2.0 supports that in some fashion, but that would have to be tested out in a lab. How many members would have an ROA already?
-- Doug McIntyre <merlyn@iphouse.net> ~.~ ipHouse ~.~ Network Engineer/Provisioning/Jack of all Trades
At the last UG, we discussed IRR-based filtering. Does Akamai publish IRR records? -- Richard
participants (10)
-
Andrew Hoyos -
Andy Koch -
Darin Steffl -
DeLong, Owen -
Doug McIntyre -
Frank Bulk -
Jason Hanke -
Jeremy Lumby -
Justin Krejci -
Richard Laager