I believe that the backup switch is already in not-prsnt state so powering it off and powering it on should not have any further impact with respect to the master switch. The lag interface will still have the member from master fpc as part of the lag ae0.
Thank you for your answers. I wonder if it's a common thing to get EX2300 crashed, as it happened just a couple of days after running into production..
If the backup switch stopped working it could mostly be due to file system corrouption.
Once you poweroff/poweron the backup and if it comes back up then check in the console logs for any bad blocks etc... during the boot up of backup.
Also you can check for the logs /var/run/dmesg.detail file to confirm if there were bad blocks. Usually a power cycle will fix it and you may not see it.
Hope this helps
Your issue might be related to the following PR,
Very interesting PR.
Nice finding skills. Kudos.
Thanks for help. After some digging, it really looks like the mentioned issue: https://prsearch.juniper.net/InfoCenter/index?page=prcontent&id=PR1442376 and in the thread on forum - https://forums.juniper.net/t5/forums/v3_1/forumtopicpage/board-id/JUNOS/thread-id/14654/page/3
I rebooted the switch and it started to work again (probably temporarly), I'll try to make an upgrade soon to 18.2R3-S2 from
The previous ECalderon's post might be the issue since you console to the device and not get any output, the switch could still forward traffic for sometime, but later might stop... What is your current switch version?
Thank you, Franky