What procedure do I follow to upgrade XOS on all switches in a pre-existing stack (X460s)?

  • 0
  • 1
  • Question
  • Updated 12 months ago
  • Answered
  • (Edited)
What procedure do I follow to upgrade XOS on all switches in a pre-existing stack (X460s)?
Photo of Stanley Riley Jr.

Stanley Riley Jr.

  • 950 Points 500 badge 2x thumb

Posted 3 years ago

  • 0
  • 1
Photo of Jeremy

Jeremy, Embassador

  • 9,788 Points 5k badge 2x thumb
download image 1.2.3.4/name-of-image.xos vr VR-Default

Put the image on a tftp server.

http://opensourceph.com/2014/07/14/extreme-networks-xos-upgrade/
Photo of Patrick Voss

Patrick Voss, Alum

  • 11,594 Points 10k badge 2x thumb
Here you go Stanley:

https://gtacknowledge.extremenetworks.com/articles/How_To/How-to-upgrade-EXOS-on-a-Summit-Stack

Please let me know if you need anything else.
(Edited)
Photo of Stanley Riley Jr.

Stanley Riley Jr.

  • 950 Points 500 badge 2x thumb
Thanks so much guys for the super fast response!
Photo of Bruce Garlock

Bruce Garlock

  • 654 Points 500 badge 2x thumb
Out of curiosity, what do most people use as a best practice before updating the Primary image?  When I deploy a switch from scratch, I usually have both PRI/SEC be the same version.  When updating to a new version, you boot into the SEC image, but how long before you make the PRI the same as the SEC? Or, do you just wait until the next release, and install that to the PRI and make the PRI the boot image?
Photo of Patrick Voss

Patrick Voss, Alum

  • 11,594 Points 10k badge 2x thumb
Hi Bruce,

In my opinion you just leave the other partition the way it is until you need to upgrade again. A lot of people will have two different version just in case one isn't working you can quickly revert to another.

Hope this helps!
Photo of EtherMAN

EtherMAN, Embassador

  • 6,960 Points 5k badge 2x thumb
1600 purple switches of various models and code revs.  We put latest and greatest general release on primary and secondary on new install... After switch has been put into service only time it is ever updated is if there is a bug we need to deal with or new feature we want to deploy.  Then the new image is preloaded on secondary in-active bucket and we will schedule a reboot during a window.  We leave older image intact along with older configuration file for about 2 or 3 weeks and then match backup with active configuration and newer image.  In 19 years there have been very few corrupted or failed updates where we had to roll back... We run a pure layer two network with very little Policies or advanced layer 3 services so I would imagine that has a lot to do with our not having more problems between image versions and configuration files...  
Photo of Lutha

Lutha

  • 304 Points 250 badge 2x thumb
Hi Guys Im doing an upgrade for the 1st time ono EXOS i have read the articles but i would like to get personal experince from the community. when i upgrade to new version do i have to reapply my config to new os. I know extreme says create a backup after you reboot into new image does it use the config from previous version. im moving from 21.1.1.4 to 22.3.1.4

feel to suggest better methods im also running stacking in some parts of my network so the upgrade on the stack will be similar to standalone switch

Thanks
Photo of Patrick Voss

Patrick Voss, Alum

  • 11,594 Points 10k badge 2x thumb
Hello Luthango,

The reason we recommend backing up the configuration is not only for upgrades but to protect yourself from any unpredictable issues during the process. The configuration is separate to the OS and will be loaded after the reboot on the new image automatically. You can treat the stacks the same considering the master node will propagate the image to the rest of the nodes.

In the future I would recommend making a new post for any questions or issue. This way everyone in the community will be notified instead of just the people who have responded to this post.