Re: [pmacct-discussion] bgp_daemon wrong next-hop

[Alberto Santos]

no problem, were you able to find something?


cheers


Alberto Santos CCIE #26648
JNCIS-SP
*"...Fix your DNS, make it dual-stack, take your mail server and make it
dual-stack, take your web server and make it dual-stack..." by Randy
Bush/RIPE IPv6*

On 25 February 2017 at 13:39, Paolo Lucente  wrote:

>
> Hi Alberto,
>
> Thanks very much for that. Catalin has been gracious enough to provide
> access to his environment (on which i should make progress very soon). I
> believe you two are bumping into the very same issue (MPLS VPN routes,
> same BGP next-hop also). I should be able to look into it during this.
> Let me include you in any (troubleshooting, debugging) conversation and
> then we can summarise all back on the list.
>
> Cheers,
> Paolo
>
> On Thu, Feb 23, 2017 at 11:00:36AM +0100, Alberto Santos wrote:
> > Hi Paolo, I want to help with this, I was quite busy in jan, but now I
> have
> > the setup almost ready.
> > I just need to install pmacct in the VM where we will have the BGP
> peering,
> > looks like the same issue I reported before.
> >
> > BR
> > Al
> >
> >
> >
> > Alberto Santos CCIE #26648
> > JNCIS-SP
> > *"...Fix your DNS, make it dual-stack, take your mail server and make it
> > dual-stack, take your web server and make it dual-stack..." by Randy
> > Bush/RIPE IPv6*
> >
> > On 17 February 2017 at 16:10, Paolo Lucente  wrote:
> >
> > >
> > > Hi Catalin,
> > >
> > >
> > > Any chance we can go unicast and arrange a way for me to troubleshoot
> > > this? Like either you point the BGP session towards one of my boxes or
> > > you can grant me temporary access to your environment? Happy to support
> > > you.
> > >
> > > Thanks,
> > > Paolo
> > >
> > > On Fri, Feb 17, 2017 at 02:53:25PM +, Catalin Petrescu wrote:
> > > > >
> > > > > Currently i have only type 0 and i can see in the logs the correct
> > > > > next-hop:
> > > >
> > > >
> > > > mistake form my side , the next hop is still not correct but has
> changed
> > > to
> > > > another randon ip.
> > > >
> > > > Regards,
> > > > Catalin
> > >
> > > > ___
> > > > pmacct-discussion mailing list
> > > > http://www.pmacct.net/#mailinglists
> > >
> > >
> > > ___
> > > pmacct-discussion mailing list
> > > http://www.pmacct.net/#mailinglists
> > >
>
___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

Re: [pmacct-discussion] bgp_daemon wrong next-hop

[Alberto Santos]

Hi Paolo, I want to help with this, I was quite busy in jan, but now I have
the setup almost ready.
I just need to install pmacct in the VM where we will have the BGP peering,
looks like the same issue I reported before.

BR
Al



Alberto Santos CCIE #26648
JNCIS-SP
*"...Fix your DNS, make it dual-stack, take your mail server and make it
dual-stack, take your web server and make it dual-stack..." by Randy
Bush/RIPE IPv6*

On 17 February 2017 at 16:10, Paolo Lucente  wrote:

>
> Hi Catalin,
>
>
> Any chance we can go unicast and arrange a way for me to troubleshoot
> this? Like either you point the BGP session towards one of my boxes or
> you can grant me temporary access to your environment? Happy to support
> you.
>
> Thanks,
> Paolo
>
> On Fri, Feb 17, 2017 at 02:53:25PM +, Catalin Petrescu wrote:
> > >
> > > Currently i have only type 0 and i can see in the logs the correct
> > > next-hop:
> >
> >
> > mistake form my side , the next hop is still not correct but has changed
> to
> > another randon ip.
> >
> > Regards,
> > Catalin
>
> > ___
> > pmacct-discussion mailing list
> > http://www.pmacct.net/#mailinglists
>
>
> ___
> pmacct-discussion mailing list
> http://www.pmacct.net/#mailinglists
>
___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

Re: [pmacct-discussion] bgp_daemon wrong next-hop

[Paolo Lucente]

Hi Catalin,


Any chance we can go unicast and arrange a way for me to troubleshoot
this? Like either you point the BGP session towards one of my boxes or
you can grant me temporary access to your environment? Happy to support
you.

Thanks,
Paolo 

On Fri, Feb 17, 2017 at 02:53:25PM +, Catalin Petrescu wrote:
> >
> > Currently i have only type 0 and i can see in the logs the correct
> > next-hop:
> 
> 
> mistake form my side , the next hop is still not correct but has changed to
> another randon ip.
> 
> Regards,
> Catalin

> ___
> pmacct-discussion mailing list
> http://www.pmacct.net/#mailinglists


___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

Re: [pmacct-discussion] bgp_daemon wrong next-hop

[Catalin Petrescu]

>
> Currently i have only type 0 and i can see in the logs the correct
> next-hop:


mistake form my side , the next hop is still not correct but has changed to
another randon ip.

Regards,
Catalin
___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

Re: [pmacct-discussion] bgp_daemon wrong next-hop

[Catalin Petrescu]

Hi Paolo ,

Looks to be related to RD type #1. If i remove all the RDs that are #1 from
the bgp update towards the pmacct , then all is working as expected.

Currently i have only type 0 and i can see in the logs the correct next-hop:

before:
{"seq": 4790, "log_type": "delete", "timestamp": "2017-02-17
14:17:34.594582", "peer_ip_src": "10.91.0.30", "event_type": "log",
"ip_prefix": "10.101.6.16/29", "bgp_nexthop": "10.91.0.148", "as_path": "",
"ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
"0:29009:3801"}

after:
{"seq": 5696, "log_type": "update", "timestamp": "2017-02-17
14:18:06.646529", "peer_ip_src": "10.91.0.30", "event_type": "log",
"ip_prefix": "10.101.6.16/29", "bgp_nexthop": "10.91.0.108", "as_path": "",
"ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
"0:29009:3801"}


Hope this helps.

Catalin

On Fri, Feb 17, 2017 at 2:18 PM, Catalin Petrescu 
wrote:

> Hi Paolo,
>
> Wondering if you can share some light on a issue i'm seeing with pmacct
> bgp_daemon.
>
> lab@lab:~$ /opt/pmacctd/bin/pmacct -V
> pmacct, pmacct client 1.6.1 (20161001-00+c5)
>  '--prefix=/opt/pmacctd' '--enable-kafka' '--enable-jansson'
>
>
> I have a vpnv4 ibgp session with on jnp box. All the routes received have
> the same bgp_nexthop of : 10.91.0.148.
>
> in the logs
> {"seq": 4391, "log_type": "update", "timestamp": "2017-02-17
> 13:57:02.149110", "peer_ip_src": "10.91.0.30", "event_type": "log",
> "ip_prefix": "10.101.6.16/29", "bgp_nexthop": "10.91.0.148", "as_path":
> "", "ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
> "0:29009:3801"}
>
> {"seq": 4392, "log_type": "update", "timestamp": "2017-02-17
> 13:57:02.149110", "peer_ip_src": "10.91.0.30", "event_type": "log",
> "ip_prefix": "10.101.6.8/29", "bgp_nexthop": "10.91.0.148", "as_path":
> "", "ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
> "0:29009:3801"}
>
> while if i do a packet capture i can see the correct next-hop in the bgp
> update.
>
> also correct on the router:
>
> cpetrescu@rtr> show route advertising-protocol bgp 10.91.92.198
> 10.101.6.16/29
>
> bgp.l3vpn.0: 1958 destinations, 2992 routes (1958 active, 0 holddown, 0
> hidden)
>   Prefix  Nexthop  MED LclprefAS path
>   29009:3801:10.101.6.16/29
> * 10.91.0.203  100I
>
> What is strange is i have infact a RD 10.91.0.148:3801 with next hop
> 10.91.0.148 advertised to the pmacct but i can't find any record of that in
> the log.
>
> This happens every time , even after the bgp session is cleared.
>
> Regards,
> Catalin
>
___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists

[pmacct-discussion] bgp_daemon wrong next-hop

[Catalin Petrescu]

Hi Paolo,

Wondering if you can share some light on a issue i'm seeing with pmacct
bgp_daemon.

lab@lab:~$ /opt/pmacctd/bin/pmacct -V
pmacct, pmacct client 1.6.1 (20161001-00+c5)
 '--prefix=/opt/pmacctd' '--enable-kafka' '--enable-jansson'


I have a vpnv4 ibgp session with on jnp box. All the routes received have
the same bgp_nexthop of : 10.91.0.148.

in the logs
{"seq": 4391, "log_type": "update", "timestamp": "2017-02-17
13:57:02.149110", "peer_ip_src": "10.91.0.30", "event_type": "log",
"ip_prefix": "10.101.6.16/29", "bgp_nexthop": "10.91.0.148", "as_path": "",
"ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
"0:29009:3801"}

{"seq": 4392, "log_type": "update", "timestamp": "2017-02-17
13:57:02.149110", "peer_ip_src": "10.91.0.30", "event_type": "log",
"ip_prefix": "10.101.6.8/29", "bgp_nexthop": "10.91.0.148", "as_path": "",
"ecomms": "RT:29009:3801", "origin": 0, "local_pref": 100, "rd":
"0:29009:3801"}

while if i do a packet capture i can see the correct next-hop in the bgp
update.

also correct on the router:

cpetrescu@rtr> show route advertising-protocol bgp 10.91.92.198
10.101.6.16/29

bgp.l3vpn.0: 1958 destinations, 2992 routes (1958 active, 0 holddown, 0
hidden)
  Prefix  Nexthop  MED LclprefAS path
  29009:3801:10.101.6.16/29
* 10.91.0.203  100I

What is strange is i have infact a RD 10.91.0.148:3801 with next hop
10.91.0.148 advertised to the pmacct but i can't find any record of that in
the log.

This happens every time , even after the bgp session is cleared.

Regards,
Catalin
___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists