I personally love the "apply path" commands with wildcards
Therefore it ONLY looks for peers in your configuration
set policy-options prefix-list bgp-addresses apply-path \"protocols bgp
group <*> neighbor <*>\"
set policy-options prefix-list dns-addresses apply-path \"system
name-server
Hi Lee
Would Flap Damping fix it for you ?
I also find that BFD can cause more problems than it fixes if you go too
aggressive with it !
regards
Sean
On 27-Apr-24 9:44 AM, Lee Starnes via juniper-nsp wrote:
Hello everyone,
Having difficulty finding a way to prevent BGP from re-establishi
On 3/2/13 12:12 AM, Morgan McLean wrote:
Would it be safe to say I could pickup the relatively cheap J2350 boxes and
stick the same version of JunOS on them and have a pretty similar
experience? Not like the MX80 has any processing cards or anything special
like the higher end MX boxes can take.
vrf-table-label is your friend
On 1/25/11 4:32 PM, Leigh Porter wrote:
Hello all,
Does anybody know how to persuade interfaces routes into the L3VPN
routing table to be redistributed across the network?
At the moment, I can build a policy to allow static routes, OSPF routes
and interface rou
On 11/18/10 12:27 PM, Sergey wrote:
Hello.
I attempt to debug ospf on gre interface but I see incoming
traffic only. Is it normal behavour or is it a bug of JunOS ?
hardware: M7i, JunOS 9.2R3.5.
did you look at the real interface (tunnel source interface) to see the
outgoing OSPF stuff ?
Hi Michael
You cannot mix MPC and DPC in the same chassis before 10.2, so you'll
need either to upgrade to 10.2 (which is in beta) or remove the DPC in
slot 1
You *should* also have upgraded the fan trays
cheers
Sean
On 5/4/10 8:25 AM, Michael Phung wrote:
Hey Guys,
We just tired install
What is your config ?
Are you sure you are setting it right, i.e. check the rib
>set routing-options rib inet6.0 static route 49::/64 discard
>show route table inet6
inet6.0: 1 destinations, 1 routes (1 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both
49::/64
On 3/17/10 11:01 AM, Felix Schueren wrote:
confirmed and very, very annoying. I've filed this as a bug over a
year ago when my MBGP sessions broke in NSR on the backup RE -
apparently, it was always broken and not working properly when using
flowspec,
Not really a bug - it's just never been
On 3/12/10 8:32 AM, Pekka Savola wrote:
Hmm. I wonder how our PICs work fine with 10mbit/s as well? :-)
Officially it isn't supported, I think, though.
Yes officially not, but mostly it did ... and you didn't even need to
pay for a licence ;-)
__
On 3/9/10 4:40 PM, Richard A Steenbergen wrote:
Nonsense. The CPU usage might be ever so slightly higher in v9, but
neither is a major contributor to overall load even at the maximum
supported RE sampling rate. The only time you*need* a services card is
when you want to do a much higher rate tha
On 3/9/10 2:38 PM, Matjaž Straus wrote:
We are using 9.6R3.8 on our Juniper MX routers and we wish to enable netflow v9
for IPv6 traffic flow sampling. Our local Juniper partner claims that a special
Multiservices DPC card is needed for netflow v9. DPCE 4x 10GE R can run tunnels
like GRE, IP-i
What are you connecting too ? Another Juniper ?
Please send messages from both ends, also configs, and confirm
interfaces are UP on each end of the circuit.
cheers
Sean
On 2/15/10 2:43 AM, Hoogen wrote:
Hi All,
I am having some issues with L2Vpn.. The circuit stays down.. and error
messag
-2/0/4.614
Any hint?
Thank you,
John
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
--
Sean Clarke
Gerard Brandtstraat 29
2332AK Leiden
The Netherlands
uniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
--
Sean Clarke
Gerard Brandtstraat 29
2332AK Leiden
The Netherlands
+31 614 261 198
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.
On 1/27/10 5:54 PM, Abel Alejandro wrote:
Hello,
Is it possible to know if there is a loopback on a T1 interface on a
Juniper?
If we need to verify if there is a loopback we activate a BERT test and
if the BERT synchronizes
then we know there is a loopback but I am wondering if there is an
easie
On 1/27/10 8:32 AM, Taqdir Singh wrote:
Hi All,
could anyone please clear me what is the actual diff between
JNCIS-E and JNCIS-M
I know M stands for M series routers.
which one is most latest ?
what is the exam fee for JNCIS ? can we do it directly without giving JNCIA
?
JNC
eriencing. This KB describes a
case where the directly connected subnet is not advertised over the
L3VPN and how to 'fix' this.
Thanks for any clues.
Kind regards,
-Jeroen-
PS: JunOS version on the PE routers is 9.3R2.8
[1] http://kb.juniper.net/index?page=content&id=KB1243
On 1/26/10 12:06 PM, Mark Tinka wrote:
Really? The -R cards won't do QoS? I haven't tried the -R
cards, but was thinking of giving them a shot!
Documentation suggests they'll do QoS, but not as deep as
the -R-Q models.
Cheers,
Mark.
R Cards do QOS, but not per unit .. only per port.
Fire
I may be wrong on this one but I think you'd do better to open a case
with customer care, rather than JTAC, as your case is not technical.
Maybe you did already.
cheers
Sean
On 19/01/2010 18:50, Stefan Fouant wrote:
I know this is a long shot, but does anyone have any contacts into any IT
fol
On 1/4/10 12:31 PM, Tore Anderson wrote:
Interesting. So you're still able to use all physical ports (ge-0/1/0
through ge-0/1/9) as before; you've actually oversubscribed the PFE by
11:10?
Correct .. the PFE has a capability of a bit more than 11G .. hence the
1G tunnel can be configure
IRB will keep the instance "UP" ... therefore use the following knob
set routing-instances VPLS protocols vpls connectivity-type ce
This will ignore the IRB being up.
cheers
Sean
On 11/9/09 2:45 PM, Clarke Morledge wrote:
Sean says:
Hi Clarke .. what's the config ? Are you using irb inter
Hi Clarke .. what's the config ? Are you using irb interfaces ?
If no irb, and the PE1 is directly connected to CE1, on interface down
the VPLS primary should switchover.
cheers
Sean
I have a VPLS multi-homed environment with two MX routers (PE1 and
PE2) connected to a single ethernet switch
>
> The juniper router can't be a NTP server in the network,the Juniper
> website is show the truth.
>
It can be used ... but the pre-requisite is it needs to peer to a "real"
NTP source .. it will then pass this info on as stratum value minus 1
___
ans-1-2;
then {
next-hop self;
}
}
l...@r3>
Thanks for your help guys..
-Hoogen
On Thu, Oct 29, 2009 at 3:36 AM, Sean Clarke
mailto:s...@clarke-3.demon.nl>> wrote:
What is in your ibgp
What is in your ibgp export policy from R1 to R3 ? Are you putting
something in there to cause an issue ?
On 10/29/09 10:43 AM, Hoogen wrote:
Hi Felix,
Thank you for the reply..
I am not sure how that 17 hidden routes came into play... But its not
there now.. I still see the issue..
Are you doing "Next hop self" on R1 to advertise to R3, or are you
trying to send the routes without R3 knowing anything about the eBGP
next-hop 10.0.5.254 ? If the latter, advertise the link between R1 and
P1 passively towards R3
On 10/29/09 9:27 AM, Hoogen wrote:
Well I am working with
Hi Bill
Not sure if it makes a difference, but try telling the switch that it's
a 1G link , i.e.
aggregated-ether-options {
link-speed 10g;
lacp {
active;
}
}
Failing that, maybe if you look in the logs, or at the show interface
ae0 extensive and show lacp interfaces out
No chance with logical routers .. but virtual routers should be possible,
What are you trying to achieve with them ?
cheers
Sean
On 10/13/09 1:51 PM, Walaa Abdel razzak wrote:
Hi Experts
Do you have ideas about increasing the number of logical routers per
physical router. I know that maxi
09 1:27 PM, Matthias Gelbhardt wrote:
Hi!
We are using only iBGP between our routers on different locations.
There is a working BGP and data-connection between the two systems.
Perhaps I can somehow restart the BFD-daemon? Maybe it crashed?
Matthias
Sean Clarke schrieb:
Are you not using an IGP
Are you not using an IGP ?
Can you ping between the 2 routers ?
On 9/8/09 1:07 PM, Matthias Gelbhardt wrote:
Hi!
I see now only outgoing BFD packets... Perhaps I should better think
about using an IGP for the internal communication.
Matthias
___
You should open a JTAC case so they can debug the stack trace file ...
this may tell you what's actually wrong with the box.
As you are running 8.0R2 you may be asked to upgrade :-)
cheers
Sean
Clue Store wrote:
Hi All,
Last friday we had some nastiness on one of our m10i's. As I am not a
J
Hi Bill
the keyword "except" is what is not allowed on the EX .. maybe you need
to write one to accept only the NMS-NETWORKS prefix list and deny the
rest ... it should do the same job
i.e.
filter ROUTER-PROTECT {
term SEQ-100-accept {
from {
source-prefix-
On 7/3/09 4:30 PM, Bit Gossip wrote:
Unfortunately I have tested it but the result is that the policer
operates independently on the 2 interfaces with the result that the
total out of the 2 GE is 2000k and not 1000k.
Any idea way and how I can get it to work in aggregate fashion.
don't
On 6/3/09 8:56 AM, Richard A Steenbergen wrote:
Psst... JUNOS image downloads have been busted for like almost a day
now, somebody might wanna fix that. :)
# telnet download.juniper.net 443
Trying 207.17.137.230...
telnet: connect to address 207.17.137.230: Operation timed out
telnet: Unable to
framing unframed;
}
unit 0 {
family inet {
address xxx.xxx.xxx.xxx/30;
}
family iso;
family mpls;
}
}
Transmission using DDF
Rgrds,
Faizal
On Fri, Apr 17, 2009 at 2:53 PM, Sean Clarke wrote:
And on the other
psulation cisco-hdlc;
e1-options {
framing unframed;
}
unit 0 {
family inet {
address xxx.xxx.xxx.xxx/30;
}
family iso;
family mpls;
}
}
Transmission using DDF
Rgrds,
Faizal
On Fri, Apr 17, 2009 at 2:53 PM, Sea
On 4/17/09 8:15 AM, Faizal Rachman wrote:
But if we ignore the restarted router, why should i have to unplug/plug E1
cable facing to peer router to make it work again.
I believe that keepalive is kind of setting to make interface know while it
is up or down. But after several times, link cannot
The way you have done it, the bandwidth will be shared
Adding filter-specific knob to the policer will make them unique ... i.e.
policer P {
filter-specific;<
if-exceeding {
bandwidth-limit 1000k;
burst-size-limit 15k;
}
then discard;
}
On 4/15/09 1:33 PM,
You might find using "instance-import" and "instance-export" much easier
to use than RIB-groups.
With a simple policy you can import/export routes from and to the inet.0
table, and use the same command within the routing instances to pull
routes from inet.0 into the VRF.
This works fine with
If I were you I'd turn on some OSPF traceoptions ... it might tell you
why it's not working
cheers
Sean
On 3/27/09 11:58 AM, Aamir Saleem wrote:
Hi All,
I am having problem to established NBMA relationship between two P2P SONET
interfaces with frame-relay encapsulation. Can anyone tell me
On 2/23/09 4:19 PM, Chris Adams wrote:
Once upon a time, Sean Clarke said:
GRE keepalives are not supported.
Does BFD work on GRE?
Not as a protocol to bring down the GRE tunnel.
You can use OSPF with BFD though and make the tunnel endpoints known via
OSPF.
I know it&#
GRE keepalives are not supported.
You'd need to use some dynamic protocol , OSPF for instance, and build
GRE tunnels based on the dynamically learned addresses (loopback for
instance), if you want to achieve some sort of dynamic GRE tunnel.
On 2/23/09 12:52 AM, Stefan Fouant wrote:
I second
You have to enable it ... by default the PIC only has 4 queues available.
i.e.
set chassis fpc 1 pic 0 max-queues-per-interface 8
cheers
On 2/16/09 6:33 PM, Andrew Jimmy wrote:
Thanks for your reply. Can you let me know the way to use 8 supported Queues
instead of 4 usable queues.
CoS que
Hi Samit
Still sounds like a Layer 2 protocol thing, generally CDP or Spanning
Tree BPDU's .. probably nothing to worry about, if you can get a sniffer
on the cisco port it could tell you what it's sending out.
Not sure a "monitor traffic interface ge-x/y/z" will help as the Juniper
probably
Juniper's stance would be if it's not a Juniper part, and if there's an
issue with that part (or RE for instance) the JTAC probably wouldn't
touch it - which I would say is fair enough.
Obviously if the memory is not Juniper's and you've a PIC fault, you
will get support on the PIC issue.
W
Faizal Rachman wrote:
Hi All,
Did anywone know how to config 2 lsp in 2 back-to-back routers while
implementing mpls loadbalancing?
Thank you.
Faizal R
___
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/ju
Hi Dave
I find using the "instance-import" (or export) command much easier than
messing about with ribs
for instance .. very simple example.
routing-options {
instance-import test;
}
# show interfaces ge-1/0/0
unit 0 {
family inet {
address 1.1.1.1/30;
}
}
# show routing-ins
I would say common practice - if you're using VRF's in a multivendor
environment with draft-rosen for instance, I believe you MUST use the
same IP address in the lo0.x interface, as the "other" vendor expects
the lo0.0 address (which the Juniper won't send). Of course all Juniper
network won't
Marlon Duksa wrote:
Thanks Sean! Please see in-line:
> Don't do it on the RE, unless you want problems.
Marlon: But would this be an option if I wanted to? It seems to me
that this is not even an option.
only for GRE or IP-IP for instance ... but don't do it .. you may screw
up your protocols
Marlon Duksa wrote:
I remember running mcast traffic on MX without converting a port to a
'tunneling' module.
Only if you are running native multicast as a receiver.
Or your DR is connected directly to the RP
Or you're running PIM-SSM
If the DR and RP are different then you need tunnel to enc
Hi Dan
You may want to use the "external" keyword in your policy. This way you
only set nhs to the external bgp routes, not ones learned via iBGP.
i.e.
policy-options {
policy-statement nhs {
from {
protocol bgp;
external;
}
then {
ne
Indeed there is
"request support information"
cheers
Sean
Abhi wrote:
Hi All
Is their any command similar to "show tech" on Junos platform which can collect
all the software and hardware details on M/T Series Chassis.
Thanks
regards
abhijeet.c
Try turning on some traceoptions in OSPF on the Juniper ... it'll
probably tell you what the problem is ... maybe a typo'ed md5 key, or an
unexpected router-id ?
cheers
Matthew Crocker wrote:
> I need some Juniper-Foo for my OSPF config.
>
> I have 3 routers connected to a GigE switch.
>
>
Hi David
You'd be better off downloading the jinstall-8.2R4.5-domestic-signed
version to be honest .. more reliable with such a "jump" in 3 versions
Once you have this on the router, Juniper always recommend doing a
"request system snapshot" first , and then
request system software add /j
Depends what you want to do.
Juniper supports static BFD, or BFD for OSPF & ISIS ... even PIM
BFD is compatible with Cisco, just watch the version number, Juniper
supports v0 and v1, not sure Cisco does v1 (didn't the last time I looked)
static : (looks a bit strange due to next hop and neigh
Hi David .. yes this is supported across multiple FPC's (and is the
normal way of doing this I would say, for port/pic/fpc redundancy)
cheers
Sean
David Ball wrote:
> Anyone know if there are restrictions regarding what ports can be
> used in an aggregated ethernet bundle on a T640? Meaning
Not sure what you're after here, but if you just do a "commit
synchronise" on the master RE, the backup RE will have the same config
(which is of course recommended)
cheers
Sean
Youness El Wardi wrote:
> Hi guys,
>
> We are installing a new RE on our J20 juniper GGSN. I believe this new
> RE
Hi Benny,
They are physically different sizes and fit into different PICs
The XFP 10GE would be IQ2 PICs, also supporting WAN-PHY and all sorts
of clever queueing/shaping.
Cheers
Sean
--
not so long ago , you wrote:
> Hi List,
> Is there any differences between 10GE interface using XFP and XENP
Hi David,
On what box ? It was broken on certain boxes .. but it's fixed now in
8.3R2
Cheers
Sean
--
not so long ago , you wrote:
> For those interested, I was able to get pbit recognition working in
> an L2VPN, and DSCP working in an L3VPN/VRF. It was, and still is, VPLS
> I'm having the
Hi Aaron,
OSPFv2 in 8.3R2 still has it, I don't know why you don't see this
option
[EMAIL PROTECTED] show protocols ospf
area 0.0.0.0 {
interface ge-0/0/0.0 {
interface-type p2p;
}
}
For IPv6 ... take a look at RFC 2740.
2.11. Identifying neighbors by Router ID
In OSPF fo
At 6:11 PM +0300 8/6/07, Amos Rosenboim wrote:
>Hello All,
>
>We are trying to migrate a frame relay network onto our mpls network,
>based on juniper M10i routers.
>We are using frame-relay-ccc encapsulation and l2circuit configuration.
>The problem is that we are unable to configure dlci values b
Hi Hamid,
Just add "set system backup-router 10.6.78.65" .. this is assuming
10.6.78.65 is your next-hop of the fxp0 interface
It's so the RE can get to the management default gateway at all
times ... the route will be installed on reboot until a "proper"
route is known
Cheers
Sean
--
Hi Chris,
Yes .. different "hump" is needed to house the RE .. It was an
orderable item once, but I guess it's EOL ..
Cheers
Sean
--
Not so long ago you wrote :
CC> Hey
CC> What parts do you need to install an RE333 into an M40 with original RE?
CC> It doesn't seem to fit into the compactp
Hi Alexander,
You can try and do what it suggests - run FSCK on the harddisk
If that's not possible, try rebuilding from an external PCMCIA card,
sometimes you can recover the disk this way.
Not really feasible to run without a harddisk - theoretically
possible, but you need to turn of
Hi Jonas,
Considering the IQ2 PIC is oversubscribed 4 to 1 ... there isn't
much problem I'd say ... you'd be lucky to get much more the 1G
through it ..
Of course if you have 4 of them, then you also oversubscribe the
FPC - but that issue's been known for ages (i.e. 4 x gig PIC's in an
Hi Steiner ...
It's not that Junipers BGP based (Kompella) L2 P2P is proprietary -
all the drafts are there, it's just not many vendors have implemented
it as the LDP version is a lot easier to code .. :-)
Cheers
Sean
--
Not so long ago you wrote :
snn> - Juniper supports both standard Marti
Hi Jacob,
H-VPLS is on the roadmap .. can't say exactly when but should be
this year ... contact your Juniper sales folks for more assistance
Cheers
Sean
--
Not so long ago you wrote :
BJJC> Hi All,
BJJC> I am having some trouble finding Juniper's official stance on H-VPLS, and
BJJC> its i
Hi Ihsan,
How have you put the route into inet.3 ?
I generally see it configured as a "discard" route, then it's not
hidden
Cheers
Sean
--
Not so long ago you wrote :
IJI> Thanks Ariff,
IJI> I've put in the prefix consisting of our loopback IP range into inet.3 and
IJI> now the VPNs nex
Hi Pekka,
There is nothing - there's a feature request in for about 8.5 (but
don't quote me as it keeps getting deferred) for a MIB trap when the
SFP/XENPAK etc is inserted or removed, but nothing else.
Cheers
Sean
--
Not so long ago you wrote :
PS> Hello all,
PS> I couldn't find a way
--
Not so long ago you wrote :
BS> The configuration is the same like configuring normal route
BS> reflector except you must enable family inet-vpn for L3VPN. The
BS> important thing is you must have a LSP to all PEs because without
BS> LSP or reachability, the RR will mark the routes as unusable
70 matches
Mail list logo
