Known issues with R80.40 Take 102?

[Alex-]

I've installed Take 102 and got the following issues in different environments:

- DHCP relay new services  would stop with messages in drop debug: @;98963267;[kern];[tid_26];[fw4_0];fw_log_drop_ex: Packet proto=17 0.0.0.0:68 -> 255.255.255.255:67 dropped by fwmultik_handle_no_match Reason: Drop template (inbound);

Switching to the Take 94 node and everything works again

- Traffic to Internet via gateway running all TP blades would become jittery, solved by failing over to machine still in Take 94 and uninstalling the Take so the upgraded node also goes back to Take 94. Then everything is OK again on both nodes. 

- Some traffic would apparently stop after upgrade from R80.40 base to Take 102 when upgrading machine via clean install (all routes and interfaces imported and checked, ARP entries, policy install successful and so on). Not much time to troubleshoot as highly impacting, had to switch back to other appliance still in R80.30. But did similar upgrades in the same environment (multiple clusters in the same SMS) on other clusters but with lower takes and never had an issue.

[PhoneBoy · Accepted solution]

Believe there is a known issue under investigation with R80.40 JHF 102.
I recommend a TAC case if you you don’t have one already.

[Alex-]

Thanks @PhoneBoy. I'll stick with 94 for now and wait for the next HF, as it's all production environments it's a bit difficult to leave the issue on for investigation.

[Nicolas_Vanhoek]

Do we know if Tac got to the bottom of this?

I have looked at the list of fixes for R80.40 GA Take_118 , there is no reference of DHCP relay problem, or problem with t_112

[rdevarak]

Current issue shows up in /var/log/messages as ""dst_release: dst:ffff88052d4c68c0 refcnt:-480". It is coming from Linux stack.

[Alex-]

Thanks, if it's just the issue under investigation I will try anyway to replicate and make a TAC case in case it would be something else.