Meraki

RaphaelL · ‎Mar 10 2022

I think 28.6 fixes some bugs related to PKMID which ''could'' slow association time. That might be worth a try !

RaphaelL · ‎Mar 10 2022

Out of curiosity , what MR firmware version are you running ?

RaphaelL · ‎Mar 9 2022

Hi , I wanted to create a simple SSID for my guest network. Upon entering the correct PSK , I want my guess to receive my splash page with a simple addition : an embed youtube video on autoplay ( rickroll'd lol ). So I created a Custom Theme in the Wireless -> Splash page. In the continue.html file I wanted to insert this snippet : <iframe src="https://www.youtube.com/embed/dQw4w9WgXcQ?autoplay=1" allow='autoplay'></iframe> The code is working fine on Chrome, but Meraki won't let me finish my evil plan. Everytime I hit save , it goes back to the default ''continue.html'' file. I was able to edit it ( ex : remove the footer ) , so that proves that save button is working , but it might not like my insertion of my iframe. Is there any limitation to that feature ? Maybe Miles has the answer 🙂 ? Thanks !

RaphaelL · ‎Mar 9 2022

Are you not using auto-vpn between your 2 Meraki products ? I'm so confused...

RaphaelL · ‎Mar 9 2022

Hi , "As soon as I disable site to site tunnel , client VPN works" this was a good indicator that you were in full tunnel and lan trafic from 172.10.0.0/16 is routed through the S2S. If thats not the case , do you have a firewall rule to allow the outbound trafic ( on Site B )

RaphaelL · ‎Mar 9 2022

You are probably in full tunnel mode. Your Client VPN flow is going thru your VPN tunnel and not over the internet.

RaphaelL · ‎Mar 9 2022

Hi , As stated by rymiles this is the SSID used for Mesh. You can disable it ( it is a Network-Wide setting ) : Meraki allows the ability to disable mesh on the Meraki Access Points. The toggle is a network-wide setting and is configured under Network-wide > Configure > General > Device configuration Disabling mesh will stop the APs from broadcasting the Mesh SSID to save airtime. https://documentation.meraki.com/MR/WiFi_Basics_and_Best_Practices/Wireless_Mesh_Networking

RaphaelL · ‎Mar 8 2022

Hi , There is no manual priority to elect a master switch. To my knowledge the switch with the lowest MAC will be the master.

RaphaelL · ‎Mar 8 2022

That rule needs to be a Site-To-Site VPN rule and not a L3 firewall rule . Correct me if I'm wrong

RaphaelL · ‎Mar 8 2022

Hi , I don't think there is any limitation to the reservation limit. Are you planning to run DHCP on the MX ? The only issue I see is : Note that the largest DHCP pool the MX will serve is equivalent in size to a /19 subnet, even on a VLAN configured with a larger subnet https://documentation.meraki.com/MX/DHCP/DHCP_Services But I hope your 5 vlans are smaller than a /19 🙂

RaphaelL · ‎Mar 7 2022

@PhilipDAth , I'm trying to move from RSTP ( spanning-tree mode rapid-pvst ) to MSTP single instance as suggested. What are the benefits ? The convergence time should be the same ( 3 hello x 2 seconds ) , no ? Am I missing something obvious ? Ps : I have read your posts on cisco also , but I can't make my mind on it 🙂 Thanks ,

RaphaelL · ‎Mar 7 2022

That doesn't make any sense. If your networks are combined , getNetworkClients will return ALL clients from the network, including those from the MX , MS , MR.

RaphaelL · ‎Mar 7 2022

Hi @Ameya_Ahir , I was able to update Meraki Case 07733973 , if you ever want to have all the specific details ( S/N , timestamps and so on ). The latest update is : "It looks like the Wireless Health Connection Log pulls from the same info as get Network Wireless Connection Stats, however Connection Log does not count for bursts of connection failure attempts. For example, if a client fails to connect 3 times in a very short amount of time, the API call will log all 3 of those attempts, however Connection Log sees this as only 1 failed connection attempt. If we have enough clients attempting to connect, that would explain why these two are giving us such different answers." But I was able to find some case that the reported info is different from SSID to SSID and versus the dashboard and the API. Thanks 🙂

RaphaelL · ‎Mar 7 2022

Thanks for the info , but this endpoint is not related to wireless Health.

RaphaelL · ‎Mar 7 2022

I was able to test both endpoints and they are returning the correct info. Are you clients displayed in Network-Wide -> Clients ?

RaphaelL · ‎Mar 7 2022

No mention of clients being wireless in this post

RaphaelL · ‎Mar 7 2022

The logs are simply indicating that the interface towards your printer is ''flapping'' . Spanning-Tree is simply putting the interface in the forwarding state when the interface is Up/Up and disabled when the interface is down/down. The ''issue'' is not Meraki related. This is a normal behavior for any port. You know have to find why the port is flapping every 30 mins but this is related to the printer.

RaphaelL · ‎Mar 7 2022

Could that be related to a Eco Energy Saving feature ? This seems to occur every 30 mins. Does this happen on every printer ?

RaphaelL · ‎Mar 4 2022

Hi , Thanks for the quick reply ! I was able to compare these endpoints (the output is identical 😞 /devices/{serial}/wireless/connectionStats : {"serial":"Q3AJ-XXXX-ZEAP","connectionStats":{"assoc":12,"auth":179,"dhcp":3,"dns":0,"success":203}} /networks/{networkId}/wireless/devices/connectionStats : [{"serial":"Q3AJ-XXXX-ZEAP","connectionStats":{"assoc":12,"auth":179,"dhcp":3,"dns":0,"success":203}}] Also when comparing the Wireless Health page from the network vs the Access Point I do get different results. Wireless -> Access Points -> ( Selecting my AP Q3AJ-XXXX-ZEAP ) -> Connections ( selecting last day 😞 says 0 errors Wireless -> Health -> Connections Log ( Filtering Selecting my AP Q3AJ-XXXX-ZEAP ) : got 54 results Finaly , the test that you just suggested ( the timeline page ) showed me 54 results which is the same as the Wireless -> Health -> Connections Log ( good news ) TLDR ; The 2 pages are not showing the same info and the API is not giving me the same info also. Thanks ! 🙂

RaphaelL · ‎Mar 4 2022

Awww I was 10 points away from the top 4 ! Better luck next month 🙂 Well done everyone !

RaphaelL · ‎Mar 4 2022

Hi , We are doing a huge revamp of our Wireless infrastructure. We are now deploying MR36 accross all sites ( 1000++ sites ) and I'm currently looking at the stats available in the dashboard. In the Dashboard , Wireless -> Health -> Connection Logs , I can easily see which SSID is in trouble and which client and so on. So I was curious to automate this process with the API. I'm using this endpoint : /networks/{networkID}/wireless/devices/connectionStats?timespan=86400&ssid={SSIDid} But it is giving me completly different stats from the Dashboard. Ex : {"serial":"Q3AJ-XXXX-XXXX","connectionStats":{"assoc":1,"auth":163,"dhcp":0,"dns":0,"success":84}} I only have 46 Auth errors for that given AP according to the dashboard. 46 != 163. Same period This behavior is present will all error types ( assoc , auth , dhcp , dns ) and seen on Networks with or without a template. I have opened a case about that. Has anyone ever experienced that ? I would expect the connectionstats and the connectionlogs to give me the same amount of errors. In the dashboard it is named ''failure type'' , but I guess failure = error in my book. Thanks ,

RaphaelL · ‎Feb 28 2022

Hi , I was able to do 2 tests. 1- Capture L2 trafic from SW1 on a vlan that is defined on the MX ( MX has a SVI on that vlan ) 2- Capture L2 trafic from SW1 on a vlan that is not defined on the MX ( MX has no SVI on that vlan ) Both captures were taken the interfaces on the MS leading to the MX. When the vlan is not present on the MX , the MX doesn't forward the trafic to SW2 There were multiple clients on SW1 doing ARP requests on vlans 3000-3100 ( which doesn't exist on the MX ). The capture on SW1 shows the L2 broadcast , and at the same time a capture on SW2 was taken and not a single L2 broadcast was seen from SW1 on vlans 3000-3100.

RaphaelL · ‎Feb 25 2022

It is allowed everywhere. But I don't see how would the MX forward the trafic from Vlan 3020 on port '4' to port '3'. The MX has no knowledge of vlan 3020.

RaphaelL · ‎Feb 25 2022

Hi , We have a super weird topology like this one : The MX on the top does Routing and the MS are simply Layer 2 switches. The MX on the bottom is strickly for the guest network. Layer 3 : MXs Layer 2 : MS Let's say we have a SSID on vlan 3020. The vlan is tagged on the SSID and clients are bridged. Client on the AP sends a DHCP discover. Will the MX on the bottom left receive it ? ( the flow would follow the red arrows ) Let's not forget that the vlan 3020 is ONLY on the MX on the bottom. The MX on the top has no idea what vlan 3020. Will the MX flood the trafic from vlan 3020 to his other ports ? I don't think so. The question may sound stupid , but sometimes the MX product line is tricky and is not really behaving like a standard switch. Thanks ,

RaphaelL · ‎Feb 24 2022

Nothing in the change logs ? A network rebind , template rebind ? Nothing in the event logs ''IP conflict'' ? If you answer no to all these questions , I think you will need to open a case to shed some light on that issue. I have over 6000 networks and none of these alerts were received today.

About RaphaelL

RaphaelL

Raphael Letourneau

Community Record

Badges