Jump to content


21
[Offer]

vMX full public release



173 replies to this topic

#15 encimaDeIca

encimaDeIca

    Advanced Member

  • Members
  • PipPipPip
  • 55 posts
  • 12 thanks

Posted 09 December 2015 - 06:17 AM

Has anyone been able to get this FULLY (function-wise) running on/in KVM in Ubuntu 14.4.3 LTS? That is, is it possible to be able to ping out from one of the 10 1GE intf's to external interfaces? I can ping out when I explicitly set the next hop MAC addr. Otherwise I cannot. My Ubuntu VM is running in a vmware vm running on ESXi 5.5. There may be an issue with one of the dependent python scripts that vmx.sh --bind-dev uses. Just a remark, BTW.  When you chase down the interfaces with tcpdump -i, then there is an issue between the Junos ge-0-0-0 (say) interface and the vmx_link bridge. Thanks. (My setup is using 20 GB of RAM and it seems not to matter whether E1000 or VMXNET interfaces are used.)

The question becomes, if vMX in KVM on Ubuntu cannot be run on ESXi, then what are people using? Bare metal? KVM enabled in an Ubuntu VM in Virtual Box?

Edited by encimaDeIca, 09 December 2015 - 07:21 PM.


Thanked by 2 Members:
vikramdahiya , saugusli69

#16 msr99

msr99

    Junior Member

  • Members
  • PipPip
  • 1 posts
  • 416 thanks

Posted 12 December 2015 - 07:36 PM

You are running in nested virtualization.setup promisc mode on network card in esxi

You can also run thins fing :https://labs.vmware.com/flings/esxi-mac-learning-dvfilter

Edited by msr99, 12 December 2015 - 07:40 PM.


#17 tp2009

tp2009

    Junior Member

  • Members
  • PipPip
  • 2 posts
  • 626 thanks

Posted 22 December 2015 - 06:47 AM

thank you very much. very cool

#18 nu11

nu11

    Member

  • Members
  • PipPip
  • 18 posts
  • 41 thanks

Posted 28 December 2015 - 04:18 PM

Hey guys,

I'm trying to run jinstall-vmx-14.1R1.10-domestic.img on ubuntu via qemu and I'm not able to get ge-0/0/X interfaces to work.
Here is my startup script:
qemu-system-x86_64 -m 2G jinstall-vmx-14.1R1.10-domestic.img -nographic -enable-kvm \
    -netdev tap,ifname=tap0,id=hostnet0,script=no,downscript=no \
    -device virtio-net-pci,netdev=hostnet0,mac=12:12:12:12:12:12 \
    -netdev tap,ifname=tap1,id=hostnet1,script=no,downscript=no \
    -device virtio-net-pci,netdev=hostnet1,mac=A2:A2:A2:A2:A2:A2 \
    -netdev tap,ifname=tap2,id=hostnet2,script=no,downscript=no \
    -device e1000,netdev=hostnet2,mac=A3:A3:A3:A3:A3:A3

The above one brings em0 with 12:12:12:12:12:12 mac linked to tap0 and em1 with a2:a2... mac linked to tao1. Unfortunately ge-0/0/0 has not linked to tap2.
I've tried also to use virtio-net-pci for the last line, but it gives error during the boot (debugger prompt).
It's a little bit strange that after image is loaded it takes around a minute to "sh interface terse" to show ge-0/0/X interfaces and PIC 0 to be shown as online.

root> show chassis fpc pic-status
Slot 0   Online    10x1G FPC
  PIC 0  Online    Virtual 10x1GE PIC

root>

Can someone help to resolve it?

#19 lsergei

lsergei

    Advanced Member

  • Members
  • PipPipPip
  • 51 posts
  • 3202 thanks
  • LocationJNCIE-SP, CCIE SP

Posted 05 January 2016 - 10:41 PM

Guys, anyone tried to run 15.1F on vmware?
I see that it can't detect true platform ("olive" instead "vmx") and i get error "Unrecognized command (chassis-control)" when i am trying to do "show chassis hardware".

#20 Kravnik

Kravnik

    Junior Member

  • Members
  • PipPip
  • 4 posts
  • 4 thanks

Posted 07 January 2016 - 09:13 PM

tried it, ,had the same issue with olive. fixed it by installing all 3 images on the RE: jinstall64-vmx-15.1F4.15-domestic.vmdk, vmxhdd.vmdk, metadata_usb.vmdk (in this order). and on the PFE: vFPC-20151203.vmdk. Now it shows me vmx and connection between pfe-re is all ok, but the Gig interfaces show up/down with Alarm on LINK. i tried it on vmware workstation, still have no ideea how to fix that.

Edited by Kravnik, 07 January 2016 - 09:15 PM.


#21 Kravnik

Kravnik

    Junior Member

  • Members
  • PipPip
  • 4 posts
  • 4 thanks

Posted 07 January 2016 - 09:46 PM

made it to work:

You need to delete the 'scsi0.present="TRUE' and scsi0.virtualDev='lsilogic' from the Vmware vmx file.

I added 10 GB Ram for the PFE (as you can see it is using quite alot), 4 vCPU, and for the RE 2 GB RAM and 1 vCPU

[email protected]> ping 172.16.3.1
PING 172.16.3.1 (172.16.3.1): 56 data bytes
64 bytes from 172.16.3.1: icmp_seq=0 ttl=128 time=2.001 ms
64 bytes from 172.16.3.1: icmp_seq=1 ttl=128 time=3.694 ms
^C
--- 172.16.3.1 ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max/stddev = 2.001/2.848/3.694/0.846 ms


[email protected]> show configuration interfaces ge-0/0/0 | display set
set interfaces ge-0/0/0 mac 00:0c:29:d4:13:54
set interfaces ge-0/0/0 unit 0 family inet address 172.16.3.2/24


[email protected]> show chassis hardware
Hardware inventory:
Item             Version  Part number  Serial number     Description
Chassis                                VMXeb             VMX
Midplane
Routing Engine 0                                         RE-VMX
CB 0                                                     VMX SCB
CB 1                                                     VMX SCB
FPC 0                                                    Virtual FPC
  CPU            Rev. 1.0 RIOT         123XYZ987
  MIC 0                                                  Virtual
    PIC 0                 BUILTIN      BUILTIN           Virtual


[email protected]> show chassis routing-engine
Routing Engine status:
  Slot 0:
    Current state                  Master
    Election priority              Master (default)
    DRAM                      2025 MB (2048 MB installed)
    Memory utilization          41 percent


On the PFE you should see at the end of the boot:

CONFIG: NIC TX ports:  CONFIG: 0  CONFIG: ;
CONFIG: I/O lcore 2 (socket 0): CONFIG: Input rings per TX port  CONFIG: 0 (CONFIG: 0x7f53cdbcf240  CONFIG: )  CONFIG: ;
CONFIG: Worker lcore 3 (socket 0) ID 0:
CONFIG: Output rings per TX port  CONFIG: 0 (0x7f53cdbcf240)  CONFIG: ;
CONFIG: Ring sizes: NIC RX = 1024; Worker in = 1024; Worker out = 1024; NIC TX = 1024;
CONFIG: Burst sizes: I/O RX (rd = 32, wr = 32); Worker (rd = 32, wr = 32); I/O TX (rd = 32, wr = 32)
RUNTIME: Logical core 2 (I/O) main loop.

RUNTIME: Logical core 3 (worker 0) main loop.

RPIO: Accepted connection from localhost <-> localhost:3000
RPIO: Accepted connection from localhost <-> localhost:3001
RIOT: Received bandwidth config: b/w : 125000

RIOT: Initializing policer for bank 0, bucket : 0 rate: 125000

METER: Low level srTCM config:
        CIR period = 19155, CIR bytes per period = 1
RIOT: New policer index: 0

HOSTIF: Accepted connectionRUNTIME: Detected port  0 status changed to UP


[email protected]:~# free -m
             total       used       free     shared    buffers     cached
Mem:          9999       7265       2734          0         13       2775
-/+ buffers/cache:       4476       5523
Swap:            0          0          0

#22 Alu01

Alu01

    Newbie

  • Members
  • Pip
  • 0 posts
  • 22 thanks

Posted 14 January 2016 - 06:20 PM

you guy's are Awesome! big thanks you made my day!

#23 blackandrose

blackandrose

    Junior Member

  • Members
  • PipPip
  • 4 posts
  • 953 thanks

Posted 20 January 2016 - 07:54 AM

i don't understand how to setup network interfaces mapping with 20x1G + 4x10G .Can we set more 20+ network interfaces in PFE VM ?.  pls help me. i can't ping anywhere

#24 HermanToothrot

HermanToothrot

    Member

  • Members
  • PipPip
  • 32 posts
  • 17143 thanks
  • LocationMonkey Island

Posted 20 January 2016 - 06:28 PM

For the full vMX release, did you set up the virtio-net interfaces as mentioned in the guide I posted?   If you're using the engineering releases in GNS3, the interface mapping is wonky. Eth0 and Eth1 are the internal interfaces, while ge-0/0/0 is actually Eth2, ge-0/0/1 is Eth3, etc...

Hope that answered your question. If not, I'll try to get some testing done, when I get some free time in the next couple of weeks, assuming one of the Juniper Gods here doesn't beat me to it.

#25 blackandrose

blackandrose

    Junior Member

  • Members
  • PipPip
  • 4 posts
  • 953 thanks

Posted 21 January 2016 - 08:38 AM

@hermanToothrot. I tried with VMware worktation included two VM Instances vMX14.2-MPC2-vm-01 , vMX14.2-RE0-vm-01 for a vMX router . For topology testing, i used two vMX routers to ping together. they've not worked .

Attached Files



Thanked by 3 Members:
singhlallan , nunziamai , vikramdahiya

#26 HermanToothrot

HermanToothrot

    Member

  • Members
  • PipPip
  • 32 posts
  • 17143 thanks
  • LocationMonkey Island

Posted 21 January 2016 - 02:00 PM

That's a good question. I haven't had the time to try using the split VMs in VMware, so I dunno. Hopefully someone who has will post the steps they used to get it working.

#27 blackandrose

blackandrose

    Junior Member

  • Members
  • PipPip
  • 4 posts
  • 953 thanks

Posted 22 January 2016 - 01:26 AM

@Herman.thanks you. anyone else help me ,pls

#28 Feb

Feb

    Member

  • Members
  • PipPip
  • 42 posts
  • 3483 thanks

Posted 25 January 2016 - 05:46 AM

Anybody owns latest version? 15.1F4 vMX ESXi

can be downloaded here

http://www.juniper.net/support/downloads/?p=vmx#sw

Thanked by 1 Member:
COTOCALI



0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

Organization

Community

Downloads

Test Providers

Site Info


Go to top