[j-nsp] IGMP problem
Hi
I would like to setup static IGMP joins between Cisco and Juniper.
But it's not working. Juniper is not sending IGMP Joins.
Same configuration Cisco + Cisco working without issues. Any clues ?
Interface configuration for Juniper at Cisco side:
interface GigabitEthernet1/1/1
description Juniper
no switchport
ip address 10.10.10.21 255.255.255.252
ip pim passive
!
Here is output of IGMP membership - none :(
cisco#sh ip igmp membership | include GigabitEthernet1/1/1
cisco#
Here is JunOS configuration:
interfaces {
ge-0/0/0 {
unit 0 {
family inet {
address 10.10.10.22/30;
}
}
}
routing-options {
static {
route 0.0.0.0/0 next-hop 10.10.10.21;
}
}
protocols {
igmp {
interface ge-0/0/0.0 {
version 2;
static {
group 231.0.0.3;
group 231.0.0.4;
}
}
}
pim {
rp {
static {
address 10.10.10.255 {
version 2;
}
}
}
interface ge-0/0/0.0 {
mode sparse;
version 2;
}
join-load-balance;
}
}
Rob
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] IGMP problem
Hi Robert,
What you have below only adds the interface to the OIL for that group. No IGMP
joins are generated!
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 07:51, Robert Hass wrote:
Hi
I would like to setup static IGMP joins between Cisco and Juniper.
But it's not working. Juniper is not sending IGMP Joins.
Same configuration Cisco + Cisco working without issues. Any clues ?
Interface configuration for Juniper at Cisco side:
interface GigabitEthernet1/1/1
description Juniper
no switchport
ip address 10.10.10.21 255.255.255.252
ip pim passive
!
Here is output of IGMP membership - none :(
cisco#sh ip igmp membership | include GigabitEthernet1/1/1
cisco#
Here is JunOS configuration:
interfaces {
ge-0/0/0 {
unit 0 {
family inet {
address 10.10.10.22/30;
}
}
}
routing-options {
static {
route 0.0.0.0/0 next-hop 10.10.10.21;
}
}
protocols {
igmp {
interface ge-0/0/0.0 {
version 2;
static {
group 231.0.0.3;
group 231.0.0.4;
}
}
}
pim {
rp {
static {
address 10.10.10.255 {
version 2;
}
}
}
interface ge-0/0/0.0 {
mode sparse;
version 2;
}
join-load-balance;
}
}
Rob
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] IGMP problem
Hello,
Actually this config generates PIM (*,G) joins upstream to RP.
I'm not aware of static igmp joins(generated) or igmp proxies support in
junos (excluding junosE) - though there is a feature that translates PIM
to
IGMP/MLDhttp://www.juniper.net/techpubs/en_US/junos11.4/topics/topic-map/mcast-message-translation.html
Krasi
On 10 September 2013 12:55, Vladislav Vasilev
vladislavavasi...@gmail.comwrote:
Hi Robert,
What you have below only adds the interface to the OIL for that group. No
IGMP joins are generated!
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 07:51, Robert Hass wrote:
Hi
I would like to setup static IGMP joins between Cisco and Juniper.
But it's not working. Juniper is not sending IGMP Joins.
Same configuration Cisco + Cisco working without issues. Any clues ?
Interface configuration for Juniper at Cisco side:
interface GigabitEthernet1/1/1
description Juniper
no switchport
ip address 10.10.10.21 255.255.255.252
ip pim passive
!
Here is output of IGMP membership - none :(
cisco#sh ip igmp membership | include GigabitEthernet1/1/1
cisco#
Here is JunOS configuration:
interfaces {
ge-0/0/0 {
unit 0 {
family inet {
address 10.10.10.22/30;
}
}
}
routing-options {
static {
route 0.0.0.0/0 next-hop 10.10.10.21;
}
}
protocols {
igmp {
interface ge-0/0/0.0 {
version 2;
static {
group 231.0.0.3;
group 231.0.0.4;
}
}
}
pim {
rp {
static {
address 10.10.10.255 {
version 2;
}
}
}
interface ge-0/0/0.0 {
mode sparse;
version 2;
}
join-load-balance;
}
}
Rob
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] IGMP problem
Robert,
Just noticed you actually have ip pim passive under the interface...
The ip igmp join-group in Cisco IOS generates IGMP joins (and PIM joins
upstream), and packets sent to the group address get sent up to the CPU (the
router would reply back to icmp-echo packets sent to the group address -
convenient for troubleshooting).
On the other hand, the ip igmp static-group in Cisco IOS generates IGMP joins
(and PIM joins upstream), but packets sent to the group address do not get sent
up to the CPU.
As Krasi said, in JunOS, you still have the PIM joins upstream, but no IGMP
joins are generated.
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 11:24, Krasimir Avramski wrote:
Hello,
Actually this config generates PIM (*,G) joins upstream to RP.
I'm not aware of static igmp joins(generated) or igmp proxies support in
junos (excluding junosE) - though there is a feature that translates PIM to
IGMP/MLD
Krasi
On 10 September 2013 12:55, Vladislav Vasilev vladislavavasi...@gmail.com
wrote:
Hi Robert,
What you have below only adds the interface to the OIL for that group. No
IGMP joins are generated!
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 07:51, Robert Hass wrote:
Hi
I would like to setup static IGMP joins between Cisco and Juniper.
But it's not working. Juniper is not sending IGMP Joins.
Same configuration Cisco + Cisco working without issues. Any clues ?
Interface configuration for Juniper at Cisco side:
interface GigabitEthernet1/1/1
description Juniper
no switchport
ip address 10.10.10.21 255.255.255.252
ip pim passive
!
Here is output of IGMP membership - none :(
cisco#sh ip igmp membership | include GigabitEthernet1/1/1
cisco#
Here is JunOS configuration:
interfaces {
ge-0/0/0 {
unit 0 {
family inet {
address 10.10.10.22/30;
}
}
}
routing-options {
static {
route 0.0.0.0/0 next-hop 10.10.10.21;
}
}
protocols {
igmp {
interface ge-0/0/0.0 {
version 2;
static {
group 231.0.0.3;
group 231.0.0.4;
}
}
}
pim {
rp {
static {
address 10.10.10.255 {
version 2;
}
}
}
interface ge-0/0/0.0 {
mode sparse;
version 2;
}
join-load-balance;
}
}
Rob
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] IGMP problem
You should enable the sap protocol for the group you want to generate join
messages.
On Tue, Sep 10, 2013 at 2:18 PM, Vladislav Vasilev
vladislavavasi...@gmail.com wrote:
Robert,
Just noticed you actually have ip pim passive under the interface...
The ip igmp join-group in Cisco IOS generates IGMP joins (and PIM joins
upstream), and packets sent to the group address get sent up to the CPU
(the router would reply back to icmp-echo packets sent to the group address
- convenient for troubleshooting).
On the other hand, the ip igmp static-group in Cisco IOS generates IGMP
joins (and PIM joins upstream), but packets sent to the group address do
not get sent up to the CPU.
As Krasi said, in JunOS, you still have the PIM joins upstream, but no
IGMP joins are generated.
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 11:24, Krasimir Avramski wrote:
Hello,
Actually this config generates PIM (*,G) joins upstream to RP.
I'm not aware of static igmp joins(generated) or igmp proxies support in
junos (excluding junosE) - though there is a feature that translates PIM
to IGMP/MLD
Krasi
On 10 September 2013 12:55, Vladislav Vasilev
vladislavavasi...@gmail.com wrote:
Hi Robert,
What you have below only adds the interface to the OIL for that group.
No IGMP joins are generated!
Regards,
Vladislav A. VASILEV
On 10 Sep 2013, at 07:51, Robert Hass wrote:
Hi
I would like to setup static IGMP joins between Cisco and Juniper.
But it's not working. Juniper is not sending IGMP Joins.
Same configuration Cisco + Cisco working without issues. Any clues ?
Interface configuration for Juniper at Cisco side:
interface GigabitEthernet1/1/1
description Juniper
no switchport
ip address 10.10.10.21 255.255.255.252
ip pim passive
!
Here is output of IGMP membership - none :(
cisco#sh ip igmp membership | include GigabitEthernet1/1/1
cisco#
Here is JunOS configuration:
interfaces {
ge-0/0/0 {
unit 0 {
family inet {
address 10.10.10.22/30;
}
}
}
routing-options {
static {
route 0.0.0.0/0 next-hop 10.10.10.21;
}
}
protocols {
igmp {
interface ge-0/0/0.0 {
version 2;
static {
group 231.0.0.3;
group 231.0.0.4;
}
}
}
pim {
rp {
static {
address 10.10.10.255 {
version 2;
}
}
}
interface ge-0/0/0.0 {
mode sparse;
version 2;
}
join-load-balance;
}
}
Rob
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion
That's my thought too. However even the 12.3 VPLS configuration guide states FEC128 multihoming. But again showing with BGP Thanks Darren http://www.mellowd.co.uk/ccie Date: Mon, 9 Sep 2013 23:30:08 +0300 Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion From: kr...@smartcom.bg To: darre...@outlook.com CC: juniper-nsp@puck.nether.net Hello, IMHO there is mess with docs/terms. FEC 128 multihoming as described has nothing to do with ldp. It's bgp signaling and autodiscovery. Krasi On 8 September 2013 22:37, Darren O'Connor darre...@outlook.com wrote: Hi list. I'm going over the VPLS multihoming options on Juniper's web site. I'm not concerned with LAG and MC-LAG for the moment. As far as I'm aware, FEC128 is when you are using manual discovery of pseudowires (LDP) - FEC129 is when you are using BGP auto-discovery. Juniper techpub for FEC129 multihoming I don't have a problem with as it shows how to multihome with BGP: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html The FEC128 multihome techpub says that you cannot enable LDP signalling, you have to use BGP signalling: http://www.juniper.net/techpubs/en_US/junos/topics/usage-guidelines/vpns-configuring-vpls-multihoming.html I know that you can use LDP for manual discovery and LDP will then signal VC labels. You can also use BGP for auto-discovery and LDP for VC label signalling. You can also use BGP for both. What I don't get is how you could use FEC128 with BGP signalling. Junos doesn't give you the option to only signal through BGP but manual discovery through LDP. So my question is, when exactly would the FEC128 config be used over the FEC129 config? If you are using BGP for signalling are you not using BGP for discovery at the same time? Or maybe I'm just misunderstanding something. Thanks Darren http://www.mellowd.co.uk/ccie ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion
Perhaps this is useful: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html There are two places in the configuration where you can configure VPLS multihoming. One is for FEC 128, and the other is for FEC 129: For FEC 128-routing-instances instance-name protocols vpls site site-name multi-homing For FEC 129-routing-instances instance-name protocols vpls multi-homing -Original Message- From: juniper-nsp [mailto:juniper-nsp-boun...@puck.nether.net] On Behalf Of Darren O'Connor Sent: Tuesday, September 10, 2013 3:53 PM To: Krasimir Avramski Cc: juniper-nsp@puck.nether.net Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion That's my thought too. However even the 12.3 VPLS configuration guide states FEC128 multihoming. But again showing with BGP Thanks Darren http://www.mellowd.co.uk/ccie Date: Mon, 9 Sep 2013 23:30:08 +0300 Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion From: kr...@smartcom.bg To: darre...@outlook.com CC: juniper-nsp@puck.nether.net Hello, IMHO there is mess with docs/terms. FEC 128 multihoming as described has nothing to do with ldp. It's bgp signaling and autodiscovery. Krasi On 8 September 2013 22:37, Darren O'Connor darre...@outlook.com wrote: Hi list. I'm going over the VPLS multihoming options on Juniper's web site. I'm not concerned with LAG and MC-LAG for the moment. As far as I'm aware, FEC128 is when you are using manual discovery of pseudowires (LDP) - FEC129 is when you are using BGP auto-discovery. Juniper techpub for FEC129 multihoming I don't have a problem with as it shows how to multihome with BGP: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html The FEC128 multihome techpub says that you cannot enable LDP signalling, you have to use BGP signalling: http://www.juniper.net/techpubs/en_US/junos/topics/usage-guidelines/vpns-configuring-vpls-multihoming.html I know that you can use LDP for manual discovery and LDP will then signal VC labels. You can also use BGP for auto-discovery and LDP for VC label signalling. You can also use BGP for both. What I don't get is how you could use FEC128 with BGP signalling. Junos doesn't give you the option to only signal through BGP but manual discovery through LDP. So my question is, when exactly would the FEC128 config be used over the FEC129 config? If you are using BGP for signalling are you not using BGP for discovery at the same time? Or maybe I'm just misunderstanding something. Thanks Darren http://www.mellowd.co.uk/ccie ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion
I understand that part, but it doesn't answer the original question. Thanks Darren http://www.mellowd.co.uk/ccie From: per.gran...@gcc.com.cy To: darre...@outlook.com; kr...@smartcom.bg CC: juniper-nsp@puck.nether.net Subject: RE: [j-nsp] VPLS Multihoming on Junos - FEC confusion Date: Tue, 10 Sep 2013 13:20:17 + Perhaps this is useful: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html There are two places in the configuration where you can configure VPLS multihoming. One is for FEC 128, and the other is for FEC 129: For FEC 128-routing-instances instance-name protocols vpls site site-name multi-homing For FEC 129-routing-instances instance-name protocols vpls multi-homing -Original Message- From: juniper-nsp [mailto:juniper-nsp-boun...@puck.nether.net] On Behalf Of Darren O'Connor Sent: Tuesday, September 10, 2013 3:53 PM To: Krasimir Avramski Cc: juniper-nsp@puck.nether.net Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion That's my thought too. However even the 12.3 VPLS configuration guide states FEC128 multihoming. But again showing with BGP Thanks Darren http://www.mellowd.co.uk/ccie Date: Mon, 9 Sep 2013 23:30:08 +0300 Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion From: kr...@smartcom.bg To: darre...@outlook.com CC: juniper-nsp@puck.nether.net Hello, IMHO there is mess with docs/terms. FEC 128 multihoming as described has nothing to do with ldp. It's bgp signaling and autodiscovery. Krasi On 8 September 2013 22:37, Darren O'Connor darre...@outlook.com wrote: Hi list. I'm going over the VPLS multihoming options on Juniper's web site. I'm not concerned with LAG and MC-LAG for the moment. As far as I'm aware, FEC128 is when you are using manual discovery of pseudowires (LDP) - FEC129 is when you are using BGP auto-discovery. Juniper techpub for FEC129 multihoming I don't have a problem with as it shows how to multihome with BGP: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html The FEC128 multihome techpub says that you cannot enable LDP signalling, you have to use BGP signalling: http://www.juniper.net/techpubs/en_US/junos/topics/usage-guidelines/vpns-configuring-vpls-multihoming.html I know that you can use LDP for manual discovery and LDP will then signal VC labels. You can also use BGP for auto-discovery and LDP for VC label signalling. You can also use BGP for both. What I don't get is how you could use FEC128 with BGP signalling. Junos doesn't give you the option to only signal through BGP but manual discovery through LDP. So my question is, when exactly would the FEC128 config be used over the FEC129 config? If you are using BGP for signalling are you not using BGP for discovery at the same time? Or maybe I'm just misunderstanding something. Thanks Darren http://www.mellowd.co.uk/ccie ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] Curly braces in AS-path
On Sep 10, 2013, at 12:45 PM, Vincent De Keyzer vinc...@dekeyzer.net wrote:
Hello all,
the following comes from the TeliaSonera looking glass (http://lg.telia.net/
):
Router: Hong Kong
Command: show route protocol bgp 1.38.0.0/17 table inet.0
inet.0: 518574 destinations, 1103564 routes (518516 active, 113 holddown,
1207 hidden)
+ = Active Route, - = Last Active, * = Both
1.38.0.0/17*[BGP/170] 1w2d 17:44:34, MED 0, localpref 150
AS path: 1273 55410 38266 {38266} ?
to 80.239.130.206 via ge-2/1/1.0
What do the curly braces in the AS path mean? More generally, does someone
know a document explaining how to interpret the output of sh route prot
bgp?
AS-SET
http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080094826.shtml#aggregatingwiththeassetargument
- Jared
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] Curly braces in AS-path
Hello all,
the following comes from the TeliaSonera looking glass (http://lg.telia.net/
):
Router: Hong Kong
Command: show route protocol bgp 1.38.0.0/17 table inet.0
inet.0: 518574 destinations, 1103564 routes (518516 active, 113 holddown,
1207 hidden)
+ = Active Route, - = Last Active, * = Both
1.38.0.0/17*[BGP/170] 1w2d 17:44:34, MED 0, localpref 150
AS path: 1273 55410 38266 {38266} ?
to 80.239.130.206 via ge-2/1/1.0
What do the curly braces in the AS path mean? More generally, does someone
know a document explaining how to interpret the output of sh route prot
bgp?
Thanks,
Vincent
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] Curly braces in AS-path
Vincent De Keyzer writes:
What do the curly braces in the AS path mean? More generally, does someone
know a document explaining how to interpret the output of sh route prot
bgp?
I don't see anything under show route protocol bgp:
http://www.juniper.net/techpubs/en_US/junos/topics/reference/command-summary/show-route-protocol.html
But under show route advertising-protocol you get decoder ring:
http://www.juniper.net/techpubs/en_US/junos/topics/reference/command-summary/show-route-advertising-protocol.html
AS path through which the route was learned. The letters at the
end of the AS path indicate the path origin, providing an
indication of the state of the route at the point at which the
AS path originated:
I - IGP.
E - EGP.
? - Incomplete; typically, the AS path was aggregated.
When AS path numbers are included in the route, the format is as follows:
[ ] - Brackets enclose the local AS number associated with the
AS path if configured on the router, or if AS path prepending
is configured.
{ } - Braces enclose AS sets, which are groups of AS numbers
in which the order does not matter. A set commonly results from
route aggregation. The numbers in each AS set are displayed in
ascending order.
( ) - Parentheses enclose a confederation.
( [ ] ) - Parentheses and brackets enclose a confederation set.
Note: In Junos OS Release 10.3 and later, the AS path field
displays an unrecognized attribute and associated hexadecimal
value if BGP receives attribute 128 (attribute set) and you
have not configured an independent domain in any routing instance.
Thanks,
Phil
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] Curly braces in AS-path
Hi, AS-set normally. Used in context of aggregation.
Cheers
mh
Message d'origine
De : Vincent De Keyzer vinc...@dekeyzer.net
Date :
A : juniper-nsp@puck.nether.net
Objet : [j-nsp] Curly braces in AS-path
Hello all,
the following comes from the TeliaSonera looking glass (http://lg.telia.net/
):
Router: Hong Kong
Command: show route protocol bgp 1.38.0.0/17 table inet.0
inet.0: 518574 destinations, 1103564 routes (518516 active, 113 holddown,
1207 hidden)
+ = Active Route, - = Last Active, * = Both
1.38.0.0/17 *[BGP/170] 1w2d 17:44:34, MED 0, localpref 150
AS path: 1273 55410 38266 {38266} ?
to 80.239.130.206 via ge-2/1/1.0
What do the curly braces in the AS path mean? More generally, does someone
know a document explaining how to interpret the output of sh route prot
bgp?
Thanks,
Vincent
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] Curly braces in AS-path
Hi, This is an AS Set that is generated when a route is aggregated. http://www.rfc-editor.org/rfc/rfc4271.txt Section 5.1.6 Thanks, Antonio Sánchez-Monge Juniper Networks Want to win up to 200 Junosphere VM days? Take the Mastering Junos Configuration Challenge! http://juni.pr/14yzj5v ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] WIN Junosphere VM credits by taking the Mastering Junos Configuration Challenge!
Hi, Apologies for the wide distribution, I thought it may be of interest to someone. We just launched a remote hands-on worldwide Junos challenge. It's a tricky one, hope you like it. Take the Mastering Junos Configuration Challenge! http://juni.pr/14yzj5v Also you can subscribe to that blog to get notified of possible future hands-on challenges on different topics (networking, security, etc...). Thanks, Ato ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] Help: Learning routes from same ASN, cisco vs juniper
I've found an interesting issue and I wanted to get some thoughts before talking to JTAC about it. I have a few of MX480s. In the past, I've advertised a dedicated /24 from my lab to my providers upstream. That /24 was never learned by my primary MX. The issue comes down to either the MX or the Cisco filtering routes that are from the same ASN. It's been a couple of years since I ran across this and I can't remember who was at fault. This behavior is biting my with regard to my DR site. At my DR, I have a SRX with say ASN 1234. It's advertising a /24. At my primary site, I also use ASN1234. I do not receive the /24 via BGP. So, either the Cisco (7600 I think) isn't advertising the route to me because it's from my ASN - OR - The MX is filtering it because it's from my ASN and coming in on a eBGP link. If it's the MX, I'm certain I can write an import filter, but I'm having an issue hunting down syntax on that. If it's the Cisco, then I can yell at the provider to have them open a TAC case. Like I said, I ran across this a few years ago, but can't remember who was at fault. I could build a multi-hop neighbor relationship to get around this, but surely there's a simpler solution... ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] Help: Learning routes from same ASN, cisco vs juniper
On 9/10/2013 1:28 PM, OBrien, Will wrote:
I've found an interesting issue and I wanted to get some thoughts before
talking to JTAC about it.
I have a few of MX480s. In the past, I've advertised a dedicated /24 from my
lab to my providers upstream.
That /24 was never learned by my primary MX.
The issue comes down to either the MX or the Cisco filtering routes that are
from the same ASN. It's been a couple of years since I ran across this and I
can't remember who was at fault.
This behavior is biting my with regard to my DR site.
At my DR, I have a SRX with say ASN 1234. It's advertising a /24.
At my primary site, I also use ASN1234. I do not receive the /24 via BGP.
So, either the Cisco (7600 I think) isn't advertising the route to me because
it's from my ASN - OR - The MX is filtering it because it's from my ASN and
coming in on a eBGP link.
If it's the MX, I'm certain I can write an import filter, but I'm having an
issue hunting down syntax on that.
If it's the Cisco, then I can yell at the provider to have them open a TAC
case.
Like I said, I ran across this a few years ago, but can't remember who was at
fault. I could build a multi-hop neighbor relationship to get around this,
but surely there's a simpler solution...
In Juniper:
https://www.juniper.net/techpubs/en_US/junos/topics/reference/configuration-statement/loops-edit-protocols-bgp-family.html
protocols {
bgp {
neighbor 10.2.3.4 {
family inet {
unicast {
loops 1;
}
}
}
}
}
-set-
set protocols bgp neighbor 10.2.3.4 family inet unicast loops 1
^^ Will allow AS in path 1 time (can be set higher).
-DMM
signature.asc
Description: OpenPGP digital signature
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] virtual router on mx
Hello,
On my mx240 ( 11.4R5.5) i try to make virtual router with additional bgp
session to my isp. Main bgp session in main routing table goes via
ae0.74, extra session in vr blackhole should go via ae0.77. Problem is
that there is no communication with peer in this virtual router. When i
ping remote site from interface in vr blackhole i see this packets on
interface in main router.. I don't know why junos transmit packets via
main routing table ?
Appreciate the help
many thanks
Peter
below config and some output
p2p addresses:
me: 10.10.7.154/30
remote: 10.10.7.153/30
# show interfaces ae0.74
description main_interface;
vlan-id 74;
family inet {
address 10.10.7.130/30;
}
#show interfaces ae0.77
description vr-blackholing;
vlan-id 77;
family inet {
address 10.10.7.154/30;
}
# show routing-instances blackhole
instance-type virtual-router;
interface ae0.77;
routing-options {
static {
route 10.10.0.243/32 next-hop 10.10.7.153;
}
autonomous-system 138;
}
protocols {
bgp {
group blackhole {
type external;
multihop;
local-address 10.10.7.154;
import blackhole-in;
export blackhole-out;
peer-as 123;
neighbor 10.10.0.243;
}
}
}
# run show route 10.10.7.153
inet.0: 466059 destinations, 1693894 routes (466053 active, 6 holddown,
2 hidden)
10.10.0.0/16 *[BGP/170] 1w5d 12:40:05, MED 0, localpref 1141, from
10.10.0.243
AS path: 123 I
to 10.10.7.129 via ae0.74
blackhole.inet.0: 4 destinations, 4 routes (4 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both
10.10.7.152/30 *[Direct/0] 00:47:23
via ae0.77
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] m10 re-333 latest version of junos
Hi list, I have a couple of old m10s with re-333s in them and I would like to upgrade them to whatever the last version of Junos code they can run. These are lab routers and not production. Can someone point me in the right direction? I've seen a few folks say 10.4 will work, but I only see the m120, m160, m320 etc on the downloads page. I assume the m120 image would work in this case? Thanks All! ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] m10 re-333 latest version of junos
Depends a lot of which flavor you have with what memory This RE came with either 246M or 768M of memory and an 80MB flash Note: For M-series, MX-series, and T-series routing platforms, the CompactFlash card memory requirement for JUNOS software Release 9.0 is 1 GB. For M7i and M10i routing platforms with only 256 MB of memory, see the Customer Support Center JTAC Technical Bulletin PSN-2007-10-001: https://www.juniper.net/alerts/viewalert.jsp?txtAlertNumber=PSN-2007-10-001actionBtn=Search. Hope this helps On Sep 10, 2013, at 2:49 PM, N. Max Pierson nmaxpier...@gmail.com wrote: Hi list, I have a couple of old m10s with re-333s in them and I would like to upgrade them to whatever the last version of Junos code they can run. These are lab routers and not production. Can someone point me in the right direction? I've seen a few folks say 10.4 will work, but I only see the m120, m160, m320 etc on the downloads page. I assume the m120 image would work in this case? Thanks All! ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
[j-nsp] M5 or M10 AC power supplies
I have an old M10 (not M10i) with DC power supplies. Does anyone have any AC power supplies they'd be willing to part with or trade for the 2 DC ones I have? This is just for playing around in the home lab... Alternatively, does anyone know of a cheap way to get enough DC power for these in a lab that doesn't have DC power? Each power supply needs 14A at 48V, about 700W. This needs to be really cheap or free, because otherwise I'm just going to trash the whole router. Thanks, Chuck ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion
Hi, Note the junos 7.5 release introduction of multi-hominghttp://www.juniper.net/techpubs/en_US/junos/topics/reference/configuration-statement/multi-homing-edit-protocols-vpls.htmlstanza (under routing-instances instance-name protocols vpls site site-name) specified in FEC 128 doc. Believe me that with 7.5 release only BGP signaling/autodiscovery was supported - I remember that at this time there was VPLS standard battlehttp://www.lightreading.com/document.asp?doc_id=589392 Krasi On 10 September 2013 17:18, Darren O'Connor darre...@outlook.com wrote: I understand that part, but it doesn't answer the original question. Thanks Darren http://www.mellowd.co.uk/ccie From: per.gran...@gcc.com.cy To: darre...@outlook.com; kr...@smartcom.bg CC: juniper-nsp@puck.nether.net Subject: RE: [j-nsp] VPLS Multihoming on Junos - FEC confusion Date: Tue, 10 Sep 2013 13:20:17 + Perhaps this is useful: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html There are two places in the configuration where you can configure VPLS multihoming. One is for FEC 128, and the other is for FEC 129: For FEC 128-routing-instances instance-name protocols vpls site site-name multi-homing For FEC 129-routing-instances instance-name protocols vpls multi-homing -Original Message- From: juniper-nsp [mailto:juniper-nsp-boun...@puck.nether.net] On Behalf Of Darren O'Connor Sent: Tuesday, September 10, 2013 3:53 PM To: Krasimir Avramski Cc: juniper-nsp@puck.nether.net Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion That's my thought too. However even the 12.3 VPLS configuration guide states FEC128 multihoming. But again showing with BGP Thanks Darren http://www.mellowd.co.uk/ccie Date: Mon, 9 Sep 2013 23:30:08 +0300 Subject: Re: [j-nsp] VPLS Multihoming on Junos - FEC confusion From: kr...@smartcom.bg To: darre...@outlook.com CC: juniper-nsp@puck.nether.net Hello, IMHO there is mess with docs/terms. FEC 128 multihoming as described has nothing to do with ldp. It's bgp signaling and autodiscovery. Krasi On 8 September 2013 22:37, Darren O'Connor darre...@outlook.com wrote: Hi list. I'm going over the VPLS multihoming options on Juniper's web site. I'm not concerned with LAG and MC-LAG for the moment. As far as I'm aware, FEC128 is when you are using manual discovery of pseudowires (LDP) - FEC129 is when you are using BGP auto-discovery. Juniper techpub for FEC129 multihoming I don't have a problem with as it shows how to multihome with BGP: https://www.juniper.net/techpubs/en_US/junos/topics/topic-map/vpls-bgp-multihoming.html The FEC128 multihome techpub says that you cannot enable LDP signalling, you have to use BGP signalling: http://www.juniper.net/techpubs/en_US/junos/topics/usage-guidelines/vpns-configuring-vpls-multihoming.html I know that you can use LDP for manual discovery and LDP will then signal VC labels. You can also use BGP for auto-discovery and LDP for VC label signalling. You can also use BGP for both. What I don't get is how you could use FEC128 with BGP signalling. Junos doesn't give you the option to only signal through BGP but manual discovery through LDP. So my question is, when exactly would the FEC128 config be used over the FEC129 config? If you are using BGP for signalling are you not using BGP for discovery at the same time? Or maybe I'm just misunderstanding something. Thanks Darren http://www.mellowd.co.uk/ccie ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] m10 re-333 latest version of junos
Hi, Am 10.09.2013 21:49, schrieb N. Max Pierson: I have a couple of old m10s with re-333s in them and I would like to upgrade them to whatever the last version of Junos code they can run. These are lab routers and not production. Can someone point me in the right direction? I've seen a few folks say 10.4 will work, but I only see the m120, m160, m320 etc on the downloads page. I assume the m120 image would work in this case? We are currently running 10.4 (R9 i should upgrade them) on our lab M5s with RE2.0 or RE3.0. You can use the M-series, MX high end series T-series Install Package which is a universal image for all the m-series devices. In theory newer releases (11 or even 12) should work as well, but i have not tried it myself. It might be they removed support for EOL Hardware somewhere down the line. Depending on the release you come from you might need/want to use the M-series, MX high end series T-series Install Media if you do not want to perform a nearly endless upgrade chain. You might need additional compact flash to install the newer releases. We used off the shelf cf cards which just worked. If you can find some old memory it can be quite handy because the newer releases use lots of memory. This Router is currently idle with basically factory default configuration lab@LR2 show chassis routing-engine Routing Engine status: Temperature 36 degrees C / 96 degrees F CPU temperature 36 degrees C / 96 degrees F DRAM 384 MB Memory utilization 77 percent CPU utilization: User 0 percent Background 0 percent Kernel 1 percent Interrupt 0 percent Idle 98 percent Model RE-2.0 -- Kind Regards Tobias Heister ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] M5 or M10 AC power supplies
I was pretty sure the old M5/M10 were around 500W, max, total. You sure that 700W isn't just the rating on the PSU? I'd bet you only need A couple hundred total to run it unless it's fully configured...Something like this - http://www.trcelectronics.com/View/Mean-Well/HRP-200-48.shtml On Tue, Sep 10, 2013 at 2:16 PM, Chuck Anderson c...@wpi.edu wrote: I have an old M10 (not M10i) with DC power supplies. Does anyone have any AC power supplies they'd be willing to part with or trade for the 2 DC ones I have? This is just for playing around in the home lab... Alternatively, does anyone know of a cheap way to get enough DC power for these in a lab that doesn't have DC power? Each power supply needs 14A at 48V, about 700W. This needs to be really cheap or free, because otherwise I'm just going to trash the whole router. Thanks, Chuck ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp -- Genius might be described as a supreme capacity for getting its possessors into trouble of all kinds. -- Samuel Butler ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
Re: [j-nsp] m10 re-333 latest version of junos
I have the 768M flavor and have already upgraded the CF to 2GB. I'm currently on 8.5 and wanted to get to the latest version for these 2 lab boxes. I'll give 10.4 a shot for now. Thanks for all of the responses! On Tue, Sep 10, 2013 at 4:10 PM, Jerry Jones jjo...@danrj.com wrote: Depends a lot of which flavor you have with what memory This RE came with either 246M or 768M of memory and an 80MB flash *Note: *For M-series, MX-series, and T-series routing platforms, the CompactFlash card memory requirement for JUNOS software Release 9.0 is 1 GB. For M7i and M10i routing platforms with only 256 MB of memory, see the Customer Support Center JTAC Technical Bulletin PSN-2007-10-001: https://www.juniper.net/alerts/viewalert.jsp?txtAlertNumber=PSN-2007-10-001actionBtn=Search . Hope this helps On Sep 10, 2013, at 2:49 PM, N. Max Pierson nmaxpier...@gmail.com wrote: Hi list, I have a couple of old m10s with re-333s in them and I would like to upgrade them to whatever the last version of Junos code they can run. These are lab routers and not production. Can someone point me in the right direction? I've seen a few folks say 10.4 will work, but I only see the m120, m160, m320 etc on the downloads page. I assume the m120 image would work in this case? Thanks All! ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp ___ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
