Reset Search
 

 

Article

How to perform netinstall on SLX platforms via NFS?

« Go Back

Information

 
TitleHow to perform netinstall on SLX platforms via NFS?
Objective
The netinstall process will re-partition and format the switch's SSD hard drive. So, all existing data (including licenses) on the local switch's SSD hard drive will be erased!

As a result, Netinstall will install the firmware like a newly shipped switch.
Environment
  • SLX-OS
  • All SLX Platforms
Requirement:
  • TFTP server (required for 9850 only)
  • NFS server
Procedure

9850


 
 Key sequence to access SLX consoles:
Ctrl+y-1 - HOST
Ctrl+y-2 - SLX VM (SLX-0S)
Ctrl+y-3 - VM (TPVM)
 
  1. Download and copy the 9850 boot files ("fusion host" folder) to your TFTP server root folder. The boot files are included in the zip with the utilities in the below article. 
  1. We want to perform netinstall on each MM, so we will need to power down the standby MM2 while we netinstall MM1
  2. On the "Active MM1 (Management Module)", run 'power-off linecard <1-8>' for all line cards
  3. On MM2, console to the HOST level via by pressing 'Ctrl-y+1', login as root and password fibranne,
  4. On MM2 HOST level, enter 'reboot -f' and stop the boot and get into the boot prom. Then, leave it there, so it does not disrupt the netinstall process for MM1
Hit ESC to stop autoboot: 0
FPGA f7600058 -> 0x12
 
1) Start system.
2) Recover password.
3) Enter command shell.
 
Option? 3
  1. On MM1, console to the host level via by pressing 'Ctrl-y+1', login as root and password fibranne.
  2. On MM1 HOST level, enter 'reboot -f' and stop the boot and get into the boot prom.
Hit ESC to stop autoboot: 0
FPGA f7600058 -> 0x12
 
1) Start system.
2) Recover password.
3) Enter command shell.
 
Option? 3
  1. Run, "printenv" to check the configured variables.
=> printenv
  1. We will need to set the below variables, if it's not set yet. Set the below variables to have the MM boot from the TFTP server, then boot the MM
NOTE: we do not need to save the variables as this will be a one time boot from the TFTP server for netinstall
setenv ipaddr 10.26.133.61
setenv netmask 255.255.128.0
setenv gatewayip 10.26.128.1
setenv serverip 10.26.176.97 <//--- TFTP server address

setenv ethact e1000e
setenv bootargs  'console=ttyS0,9600n8 initrd=0x4000000,20M quiet reboot=pci rootfstype=ext4 ip=${ipaddr}:${serverip}:${gatewayip}:${netmask}::eth0:on pci=noaer'
setenv bootcmd 'tftp 1000000 netinstall/fusion_host/bzImage-3.14.17; tftp 4000000 netinstall/fusion_host/ramdisk-3.14.17.image; zboot 0x1000000  0x1000000   0x4000000  0x2000000'

boot
  1. Once ramdisk is booted, you will get to the fusion prompt, login as root and password fibranne
fusion login: root
Password: (fibranne)
  1. Run 'ifconfig' to confirm the IP address on the MM. Then, try to ping the NFS server
  2. mount the firmware from the NFS server (In our setup, 10.26.176.97 is a TFTP and NFS server)
fusion ~: mount -o tcp,nolock 10.26.176.97:/tftp/sre/Released/slxosr /dist
  1. cd to SLX-OS firmware
fusion ~: cd /dist/slxos18r.1.00b/ 
  1. make folder and contents executable. 
fusion slxos18r.1.00b: chmod -R 755 install* 
  1. run the netinstall script for host, vm. (Note: Install.net will delete and re-add the TPVM partition)
fusion slxos18r.1.00b: ./install_cr/install.net; ./install_cr/install.host; ./install_cr/install.vm; reboot -f 
  1. Once MM1 boots up, power on the linecards and MM2 to have it sync the firmware from MM1

9540

 

 
  1. Perform this keyboard sequence at console:  'Ctrl-y+1' to get the SLX host level prompt/login. 
    1. At the SLX host level, enter 'reboot -f' to reload the host, then pres 'ESC' to stop the boot, and get into the boot prom by entering "3"
Ubuntu 14.04 LTS HOST ttyS0

HOST login: root
Password: (fibranne)
Last login: Wed Feb 13 20:08:04 BRST 2019 from pb_vm1 on pts/4
Welcome to Ubuntu 14.04 LTS (GNU/Linux 4.4.7 x86_64)

 * Documentation:  https://help.ubuntu.com/

  System information as of Wed Feb 13 20:10:59 BRST 2019

  System load:  1.97              Processes:          223
  Usage of /:   14.4% of 9.72GB   Users logged in:    0
  Memory usage: 39%               IP address for br1: 127.2.0.1
  Swap usage:   0%

  Graph this data and manage this system at:
    https://landscape.canonical.com/

root@HOST:~# reboot -f
...
Hit ESC to stop autoboot: 0
FPGA f7600058 -> 0x12
 
1) Start system.
2) Recover password.
3) Enter command shell.
 
Option? 3
  1. Enter, 'onie_rescue' to get into recovery mode
=> run onie_rescue
 
Info: Mounting kernel filesystems... done.
scsi 6:0:0:0: Direct-Access     SanDisk  Ultra Fit        1.00 PQ: 0 ANSI: 6
sd 6:0:0:0: [sdc] 121307136 512-byte logical blocks: (62.1 GB/57.8 GiB)
 
ONIE:#
NOTE: Once in ONIE environment, you may see auto-discovery messages like below continuously scrolling. Run, “onie-discovery-stop” at ONIE shell to stop the messages.
Info: Fetching http://10.26.167.74/onie-installer-x86_64-avalanche-r0 ...
Info: Fetching http://10.26.167.74/onie-installer-x86_64-avalanche ...
Info: Fetching http://10.26.167.74/onie-installer-avalanche ...
Info: Fetching http://10.26.167.74/onie-installer-x86_64-brcd ...
Info: Fetching http://10.26.167.74/onie-installer-x86_64 ...
Info: Fetching http://10.26.167.74/onie-installer ...

ONIE:/ # onie-discovery-stop
  1. Run 'ifconfig' to for IP address. If it's not set, then run the following to set the IP and gateway address. 
ONIE:/ # ifconfig eth0 10.26.143.56 netmask 255.255.128.0 up; route add default gw 10.26.128.1
  1. Mount the firmware from the NFS server to /dist
ONIE:/ # mount -o tcp,nolock 10.26.176.97:/tftp/sre/Released/slxosr /dist
  1. cd to SLX-OS firmware
ONIE:/ # cd /dist/slxos18r.1.00/
  1. Make folder and contents executable
ONIE:/dist/slxos18r.1.00 # chmod -R 755 install*
  1. run the netinstall script for host, vm, and reboot
ONIE:/dist/slxos18r.1.00 # ./install_pbr/install.host; ./install_pbr/install.vm; reboot -f

9030/9140/9240
/9150/9250

 
 
  1. Connect to switch console to reload switch via 'reload system
  1. Wait for the GRUB menu, then select '*ONIE' -> '*ONIE: Rescue'
GNU GRUB  version 2.02~beta2+e4a1fe391

 +----------------------------------------------------------------------------+
 | SLX-OS                                                                     |
 | Offline Diagnostic                                                         |
 |*ONIE                                                                       |
 |                                                                            |
 |                                                                            |
 +----------------------------------------------------------------------------+


                      GNU GRUB  version 2.02~beta2+e4a1fe391

 +----------------------------------------------------------------------------+
 |*ONIE: Rescue                                                               |
 |                                                                            |
 |                                                                            |
 +----------------------------------------------------------------------------+
  1. After it boots into 'ONIE: Rescue' environment, press enter to get the prompt
Info: Mounting ONIE-BOOT on /mnt/onie-boot ...
Info: Using eth0 MAC address: 60:9c:9f:ec:b3:a9
Info: eth0:  Checking link... up.
Info: Trying DHCPv4 on interface: eth0
ONIE: Using DHCPrandom: dropbear urandom read with 108 bits of entropy available
v4 addr: eth0: 10.26.130.127 / 2random: nonblocking pool is initialized
55.255.128.0
Starting: dropbear ssh daemon... done.
Starting: telnetd... done.
discover: Rescue mode detected.  Installer disabled.

Please press Enter to activate this console.
NOTE: Once in ONIE environment, you may see auto-discovery messages like below continuously scrolling. Run, “onie-discovery-stop” at ONIE shell to stop the messages.
Info: Attempting http://10.26.142.85/onie-installer-x86_64.bin ...
Info: Attempting http://10.26.142.85/onie-installer ...
Info: Attempting http://10.26.142.85/onie-installer.bin ...

ONIE:/ #
ONIE:/ # onie-discovery-stop
discover: installer mode detected.
Stopping: discover... done.
 
  1. Run 'ifconfig' to for IP address. If it's not set, then run the following to set the IP and gateway address. 
ONIE:/ # ifconfig eth0 10.26.143.56 netmask 255.255.128.0 up; route add default gw 10.26.128.1
  1. Create /dist and mount the firmware from NFS server to /dist
ONIE:/ # mkdir /dist
ONIE:/ # mount -o tcp,nolock 10.26.176.97:/tftp/sre/Released/slxoss /dist
  1. cd to SLX-OS firmware
ONIE:/ # cd /dist/slxos18s.1.01
  1. Make folder and contents executable
ONIE:/dist/slxos18s.1.01 # chmod -R 755 *install*

9140/9240

  1. Run, 'onie-slxos-install' to start netinstall process. After, switch will automatically reboot
ONIE:/dist/slxos18s.1.01 # ./onie-slxos-install

9030/9150/9250

  1. Run, 'install_url file://SLXOS_SWBD3006-18x.100-ONIE.bin' to start netinstall process. After, switch will automatically reboots
ONIE:/dist # install_url file://SLXOS_SWBD3006-18x.100-ONIE.bin
discover: Rescue mode detected. No discover stopped.
      After a 2-3 minutes the install should start
ONIE: Executing installer: file://SLXOS_SWBD3006-18x.1.00a-ONIE.bin
Verifying image checksum ... OK.
Preparing image archive ... OK.
 
  
Additional notes

Feedback

 

Was this article helpful?


   

Feedback

Please tell us how we can make this article more useful.

Characters Remaining: 255