5.2 and ha on fgt100d

[hwhunsr]

After update a 2 cluster fgt100d active-passive, the slave gateway is down and its not coming up until the 2 ha lines are disconnected.

[rwpatterson]

When you split them, are they on the same version of code?

[hwhunsr]

Yes, after split they have same code version. I the meantime I have downgraded, because of availability. But I have other installations for example two 80c as active-passive ha, working without any problem after upgrade.

[Warren_Olson_FTNT]

I' ve seen a couple instances where the boot menu automatically chose TFTP option on secondary unit after an upgrade and got stuck in a sort of boot loop...Not saying this is the issue but what I would recommend is attaching a console to the secondary unit and watch it post-upgrade/reboot to see what' s happening and/or it' s getting stuck somewhere.

[ede_pfau]

I agree to Warren. Just to rule out that the upgrade somehow " stuck" I' d take the second unit out completely, boot, format the disk from the boot menu, reload 5.2 firmware via TFTP, reload the config (to set the HA parameters) and reconnect, HA link first. Sounds complicated but won' t take more than half an hour, and no downtime. Some precautions: - make absolutely sure that the slave unit will be secondary on reconnect - set it' s HA priority lower than the master' s. - of course, save the config of the slave first - do not confuse with the config from the cluster! Easiest would be via TFTP or USB. - formatting will erase all log files as well (which sometimes is not a real loss) At least, after this and still seeing the problems, you will know it' s a real bug pertaining to this model.