Re: [j-nsp] Stiching L2 to L3 on MX480

2017-12-21 Thread Wojciech Janiszewski
Hi Pshem, Yes, you can. Just as I described in my previous post. Regards, Wojciech 22.12.2017 00:43 "Pshem Kowalczyk" napisał(a): Hi, Can we use this for non-BNG functions as well? Despite a large number all of those VLANs are 'static' customers. kind regards Pshem On Thu, 21 Dec 2017 at

[j-nsp] MX104 won't recognize fpc, mpc and onboard interfaces

2017-12-21 Thread Caio
Hello there, We had a MX104 router in production for a very long time and recently we replaced it with a brand new MX480 and we took it to our lab in order to load the default configuration on it and start to configure it over to another POP. However, after the factory default it won't recognize t

Re: [j-nsp] Stiching L2 to L3 on MX480

2017-12-21 Thread Pshem Kowalczyk
Hi, Can we use this for non-BNG functions as well? Despite a large number all of those VLANs are 'static' customers. kind regards Pshem On Thu, 21 Dec 2017 at 23:01 wrote: > Hi Pshem, > > The feature designed specifically for what you are looking for is > pseudowire > headend (or pseudowire h

Re: [j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread adamv0025
It has the same (well 31.8M DWORDs/254.3MB vs 32M DWORDs/256MB) EDMEM RLDRAM as gen1 and 2 trio chips But it's not storing ENCAPS in there which gives it additional 4.1M DWORDs in the shared memory which next hops can take from. adam netconsultings.com ::carrier-class solutions for the telec

Re: [j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread adamv0025
Hi Sebastian, Is this 10M FIB within the next hop 4M DWORD allotted space or also using recommended 80% of the shared EDMEM space too or even all of it actually not leaving anything for FW and Counters or proper functioning of the LU? Being a vendor I'd boot it up to a theoretical max just to loo

Re: [j-nsp] Subscriber Management with RADIUS authentication question

2017-12-21 Thread Alex D.
Hello Alex, You use the delete-binding-on-renegotiation statement to override the default action, and to specify that DHCP tear down the existing matching client entry and to process the message as a new client entry. https://www.juniper.net/documentation/en_US/junos/topics/reference/configu

[j-nsp] Subscriber Management with RADIUS authentication question

2017-12-21 Thread Alex D.
Hi, i have a running Subscriber Management setup in my testlab with Juniper MX5 running JUNOS15.1R6.7. MX is functioning as DHCP proxy and forwards all DHCP requests to an external DHCP server after successful authentication against a RADIUS server based on DHCP Option 82 remote-id. RADIUS ser

Re: [j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread Olivier Benghozi
The use of NH DMEM might also slightly vary with various features (LFA/PIC/multipath), I guess. > On 21 dec. 2017 at 12:19, adamv0...@netconsultings.com wrote : > > Junos code version: > Number of VRFs: > Number of destinations (total or average per VRF): > Output from: request pfe execute targe

Re: [j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread Sebastian Becker
Hi Jason, you’re right … the claims means that there is no (extra) software restriction besides what the hardware can deliver itself. The MX204 should have the following scale limits: FIB:10M (IPv4 and IPv6 combined) RIB:80M (IPv4); 50M (IPv6) VRFs: 6050 Mostly the gi

Re: [j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread Jason Lixfeld
Hey there, General question - MX204-IR, for example, claims no RIB/FIB scale restrictions. While I’m sure with that claim, RIB scale is limited to the amount of physical memory available on the box, I’m not sure what the physical limits are around the FIB. My understanding is that it’s Trio G

[j-nsp] Poll Question (VRF scale on MX)

2017-12-21 Thread adamv0025
Hi folks, I have this large scale rollout and while doing scaling testing and Juniper recommendations will get you some confidence, I'd like to understand where we land on the graph in comparison with other operators (can't get this info from Juniper folks unfortunately). But we can build our own

Re: [j-nsp] Stiching L2 to L3 on MX480

2017-12-21 Thread adamv0025
Hi Aaron, In Carrier Ethernet VLAN IDs are arbitrary at both ends of the PW. In other words, you can do whatever you want at each end pop/push/translate one/two VLAN headers (QinQ or 802.1ad). The only thing that both ends need to agree on with regards to VLAN tagging is the resulting MTU. adam

Re: [j-nsp] Stiching L2 to L3 on MX480

2017-12-21 Thread adamv0025
Hi Pshem, The feature designed specifically for what you are looking for is pseudowire headend (or pseudowire headed termination PWHT on junos). I'm just really surprised no one mentioned it already. The limit is 2048 in 15.1 and should be raised to around 15k on 17.1. But as always you should do

Re: [j-nsp] SRX and http/https proxy

2017-12-21 Thread Benoit Plessis
On 20/12/2017 23:00, Roger Wiklund wrote: > You can download the latest signature here: > > https://kb.juniper.net/InfoCenter/index?page=content&id=KB27038 > > Try this: > > 1. unzip the file, then gunzip all gz files: gzip -d *.gz > 2. copy all files to the device with scp: scp -r * > root@ip:/var