I have 2 border routers that mesh connect to 2 vendor routers on their own vlan. The border routers connect to another pair of internal routers on another vlan. So bdr => vendor eBGP bdr => internal iBGP I have bgp peers connected bdr -> internal Now I am only seeing a very limited range of prefixes. 1) I think my aggregate is affecting my iBGP 2) I think my outfilter on my instance is affecting my iBGP as well I thought iBGP would be treated differently. I am thinking of instead of sharing my AS between inside and the border I create a new private AS. As I don't want to change my BGP filters Alex
Hi Alex, iBGP does not share routes received from iBGP to other iBGP peers, as they are supposed to be meshed. Is that likely to be what is causing the issue you are seeing? Using a private AS for internal BGP would get around this issue. Regards, Philip Loenneker | Network Engineer | TasmaNet 40-50 Innovation Drive, Dowsing Point, Tas 7010, Australia P: 03 6165 2542 | M: 0404 097 816 philip.loenneker@tasmanet.com.au www.tasmanet.com.au -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Alex Samad - Yieldbroker Sent: Monday, 14 November 2016 9:54 AM To: public@talk.mikrotik.com.au Subject: [MT-AU Public] bgp question I have 2 border routers that mesh connect to 2 vendor routers on their own vlan. The border routers connect to another pair of internal routers on another vlan. So bdr => vendor eBGP bdr => internal iBGP I have bgp peers connected bdr -> internal Now I am only seeing a very limited range of prefixes. 1) I think my aggregate is affecting my iBGP 2) I think my outfilter on my instance is affecting my iBGP as well I thought iBGP would be treated differently. I am thinking of instead of sharing my AS between inside and the border I create a new private AS. As I don't want to change my BGP filters Alex _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au
Hmm. Okay that makes sense I guess my vendor A router direct connects to my bdr routers but not my internal routers, I had thought the bdr's would redistribute the routes. Cause I have my routers in the DC that connect to ISP ISP A -> rtr A -> rtr B ISP A -> rtrA is eBGP but rtrA -> rtrB is iBGP and rtrB has all of the routes advertised from ISP A in its routing table..... A -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Philip Loenneker Sent: Monday, 14 November 2016 10:09 AM To: MikroTik Australia Public List <public@talk.mikrotik.com.au> Subject: Re: [MT-AU Public] bgp question Hi Alex, iBGP does not share routes received from iBGP to other iBGP peers, as they are supposed to be meshed. Is that likely to be what is causing the issue you are seeing? Using a private AS for internal BGP would get around this issue. Regards, Philip Loenneker | Network Engineer | TasmaNet 40-50 Innovation Drive, Dowsing Point, Tas 7010, Australia P: 03 6165 2542 | M: 0404 097 816 philip.loenneker@tasmanet.com.au www.tasmanet.com.au -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Alex Samad - Yieldbroker Sent: Monday, 14 November 2016 9:54 AM To: public@talk.mikrotik.com.au Subject: [MT-AU Public] bgp question I have 2 border routers that mesh connect to 2 vendor routers on their own vlan. The border routers connect to another pair of internal routers on another vlan. So bdr => vendor eBGP bdr => internal iBGP I have bgp peers connected bdr -> internal Now I am only seeing a very limited range of prefixes. 1) I think my aggregate is affecting my iBGP 2) I think my outfilter on my instance is affecting my iBGP as well I thought iBGP would be treated differently. I am thinking of instead of sharing my AS between inside and the border I create a new private AS. As I don't want to change my BGP filters Alex _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au
Route reflection might be the best solution -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Alex Samad - Yieldbroker Sent: Monday, 14 November 2016 9:18 AM To: MikroTik Australia Public List <public@talk.mikrotik.com.au> Subject: Re: [MT-AU Public] bgp question Hmm. Okay that makes sense I guess my vendor A router direct connects to my bdr routers but not my internal routers, I had thought the bdr's would redistribute the routes. Cause I have my routers in the DC that connect to ISP ISP A -> rtr A -> rtr B ISP A -> rtrA is eBGP but rtrA -> rtrB is iBGP and rtrB has all of the routes advertised from ISP A in its routing table..... A -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Philip Loenneker Sent: Monday, 14 November 2016 10:09 AM To: MikroTik Australia Public List <public@talk.mikrotik.com.au> Subject: Re: [MT-AU Public] bgp question Hi Alex, iBGP does not share routes received from iBGP to other iBGP peers, as they are supposed to be meshed. Is that likely to be what is causing the issue you are seeing? Using a private AS for internal BGP would get around this issue. Regards, Philip Loenneker | Network Engineer | TasmaNet 40-50 Innovation Drive, Dowsing Point, Tas 7010, Australia P: 03 6165 2542 | M: 0404 097 816 philip.loenneker@tasmanet.com.au www.tasmanet.com.au -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Alex Samad - Yieldbroker Sent: Monday, 14 November 2016 9:54 AM To: public@talk.mikrotik.com.au Subject: [MT-AU Public] bgp question I have 2 border routers that mesh connect to 2 vendor routers on their own vlan. The border routers connect to another pair of internal routers on another vlan. So bdr => vendor eBGP bdr => internal iBGP I have bgp peers connected bdr -> internal Now I am only seeing a very limited range of prefixes. 1) I think my aggregate is affecting my iBGP 2) I think my outfilter on my instance is affecting my iBGP as well I thought iBGP would be treated differently. I am thinking of instead of sharing my AS between inside and the border I create a new private AS. As I don't want to change my BGP filters Alex _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au
So I just checked the advertisements section and I am only advertising whats in the instance filter... Going to clean up my filters -- instead of doing it at the instance going to bring it back to the bgp peer level first and then see what happens from there. Alex -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Joseph B Sent: Monday, 14 November 2016 10:31 AM To: public@talk.mikrotik.com.au Subject: Re: [MT-AU Public] bgp question
Using a private AS for internal BGP would get around this issue.
Making the internal routers route reflector clients of the "internet facing" routers is another option that might work. _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au
Okay - for the list . I have redone my bdr -> vendor bgp setup - with the standard filters I am using. I had to add in nexthop-choice=force-self on the bdr to internal peerings, but now I have my routers from vendor -> bdr -> internal bgp. So the outbound filter I had on the instance was affecting it. Now to weight the bgp routes better then my internal OSPF routes .. thanks Alex -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Alex Samad - Yieldbroker Sent: Monday, 14 November 2016 10:42 AM To: MikroTik Australia Public List <public@talk.mikrotik.com.au> Subject: Re: [MT-AU Public] bgp question So I just checked the advertisements section and I am only advertising whats in the instance filter... Going to clean up my filters -- instead of doing it at the instance going to bring it back to the bgp peer level first and then see what happens from there. Alex -----Original Message----- From: Public [mailto:public-bounces@talk.mikrotik.com.au] On Behalf Of Joseph B Sent: Monday, 14 November 2016 10:31 AM To: public@talk.mikrotik.com.au Subject: Re: [MT-AU Public] bgp question
Using a private AS for internal BGP would get around this issue.
Making the internal routers route reflector clients of the "internet facing" routers is another option that might work. _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au _______________________________________________ Public mailing list Public@talk.mikrotik.com.au http://talk.mikrotik.com.au/mailman/listinfo/public_talk.mikrotik.com.au
participants (4)
-
Alex Samad - Yieldbroker -
Joseph B -
Philip Loenneker -
Tim Warnock