LACP Aggregate on Switch Stack - Slow failover?

[JimBean]

Hello

we have a 2 switch stack (MS425-32)

Stacked with Port 33 SW1 going to Port 34 on SW2 and Port 34 SW1 going to Port 33 SW2

Stack is good and healthy.

We have a aggregate on port 1 on both switches that face a firewall

Port status	Enabled
Type	Trunk
Native VLAN	666
Allowed VLANs	123
Access policy	Open
Link negotiation	Auto negotiate (10 Gbps)
RSTP	Enabled (Forwarding)
Port schedule	Unscheduled
Port isolation	Disabled
Trusted DAI	Disabled
UDLD	Alert only
Tags	none
PoE	n/a
Port mirroring	Not mirroring traffic
Stacking port	Disabled

and 

We have a aggregate on port 2 on both switches that face my machine

Port status	Enabled
Type	Trunk
Native VLAN	666
Allowed VLANs	123
Access policy	Open
Link negotiation	Auto negotiate (10 Gbps)
RSTP	Enabled (Forwarding)
Port schedule	Unscheduled
Port isolation	Disabled
Trusted DAI	Disabled
UDLD	Alert only
Tags	none
PoE	n/a
Port mirroring	Not mirroring traffic
Stacking port	Disabled

if we do a switch reboot (eg reboot switch 1 or 2) , we are seeing a 30 second outage before the firewall interface on VLAN123 is available again , what we are expecting is almost a instant / 1-2 second outage

We made a change on the firewall side ,  changing it from Passive SLOW LACP to Passive FAST LACP

and it made a big improvement - approx 10 Second outage before the firewall interface is available again on switch reboot.

What we are trying to get to is normal ~1 second outage on a switch reboot - where the links fail to the other switch

please can you check if our Agg towards the firewall is configured as best practise , 

i suspect we may need to disable RSTP ,  and maybe some other aggregate settings - any help or assistance appreciated.as this is our first Meraki Switch Stack and we are struggling to find best practises on LACP Aggregates towards devices such as a firewall

thanks!

[GIdenJoe]

Yes I believe that has been recognized as a bug.
I also recently did a single stack member poweroff to test something and we also had the packet loss.

Even the most recent MS 15.21.1 firmware has this issue according to the firmware release notes.

Excerpt:

Ms2xx/35x/4xx known issues

• Cross-stack LACP bundles experiencing a switch reboot will cause the remaining online port to experience an outage for up to 30 seconds. The same is seen again when the switch comes back online (present since MS 10)

[JimBean]

Thanks for this info,

does anyone know if there is a official bugID number or some kind of reference number? 

[JacekJ]

Good question, would like to know that as well - to be honest since a few major issues we had with Meraki switches firmwares I'm reading each and every changelog and discovering bugs that we run into.

It has happened a lot that bugs were discovered and noted as known in 15.xx releases, saying that they are known since 14.xx or even 12.xx but they are not shown on the 14.xx affected versions changelogs (seems that they just add that on the current release when this has been acknowledged without updating the old affected ones).

If you have stacks with more than two switches, watch out for this one, its very nasty: 

• Connecting a stacking cable to a stack that is online may result in a stack member going offline (present since MS 12)

[GIdenJoe]

Meraki does not use any public facing bug ID's like Cisco does.  Perhaps internal they use them but you should ask them directly.

For now be assured that this is a known issue listed in the Firmware issues.