@Hank21 Thank you for your prompt response.

Did you have one OC300 management on each site? Or use one OC300 to manage 9 sites?

Yes , Every site has its own dedicated OC300.

If you only have one OC300, what is the error message when failed connections to this particular site?

"Failed to connect to Controller"

Could you confirm if you have a large data transmission within this site?

Are there any high-traffic applications on the back-end devices?

There are no back end services running. It's a simple setup, Gateway to CoreSwitch,  OC300 connected to Coreswitch and Coreswitch connects  - to All Access switches via SFP ports (Fiber) and All access switches to there respective EAPs 13 per floor. 4 floors + Ground floor with 6 EAPs.

Max Number of client devices (users), 250, Rate limited to 3Mbps max each. Max Gate way load reaches 21% during peak hours.

Earlier Core switch was TL-SX3008f it was removed as the site was not allowing the clients to browse at all. the Fiber switch was replaced with SG2428P in emergency , but the problem with connection to OC300 existed with the SX3008f as well.

Thanks and Regards

  @Hank21 Thank you for your quick and prompt response and support.

May I know how you solve the issue? By changing different web browsers, refreshing the page, or rebooting the controller?

I Keep trying to connect , several fail, attempts and i have to refresh the main page where all the controllers are listed coz it simply does not connect after 3 to 4 attempts , i refresh the page and than keep try till it eventually connects.

Did the local OC300 keep running properly when you see the error message on the Cloud website?

Local connection is perfectly fine,the site is approx 60 Km away from my office, so when i visit there i use LAN cable to connect it connects immediately.

Is there any different configuration or points for this issue site in terms of the devices managed by each controller and the client configuration carried by each site?

One thing i had to set different was the DNS , all sites use Cloud flare and Google DNS, how ever this site seems to only connect with DNS 1 is set to local isp DNS and DNS 2 can be either or Cloud flare or Google DNS.

Do you use one PC to manage all sites?

Yes , I have two devices and i usually use My Desktop or My laptop and occasionally the OMADA app for quick site monitoring and overview.

May I know the MAC address of this issue site?

Note: You may send the MAC address via private message (move the mouse over my avatar and click Message).

I have shared the MAC address in the Private message.

Thankyou and Regards

  @Hank21 

Hi Hank, Apologies for the delayed response and thank you for your continued support.

The issue is not resolved as of yet. By changing the DNS to Google i lost the site completely, as the site works with local DNS for some reason unknown to me.

I will be be visiting the site in the near future and updating the OC300 V1.0 to the latest 5.9.32 and will update you.

Thank you and Regards

Hank21 wrote

Hi @Kash01976 ，

 

Thanks for all the detailed information you provided.

Could you please change the Primary DNS Server to Google DNS on the issue site for testing?

  @Hank21 

Hi, hope you are doing well, I updated the site with OC300 V1.0 -- 5.7.6 --> 5.9.32 successfully, Tried the DNS and lost the connection, I than Left the Internet DNS blank and replaced all the LAN and VLAN DNS Refer Fig 1 and 1a to Google DNS,Refer Fig 1 , 1a and 1b. I was able to browse for a few seconds and disconnected and than reconnect and this repeats, connect and disconnect, while doing so. 

Fig 1 below:

Fig 1a below:

Fig 1b below

I was not able to get cloud access nor was able to get internet connection steady so i reverted back to local ISP DNS of 213.42.20.20 as shownin Fig 1,1a and 1b.

After this change and reversion i notcied something very odd that the TX and RX Swapped, TX increased and RX decreased. REFER Fig 2.

The first gaps in the graph above is during controller upgrade , second break or gap in the graph was when i saw the swap between TX and RX graph and rebooted the controller hoping it would normalize. which it did not still.

I was still having trouble with steady network connection, even after reverting to LOCAL ISP DNS 213.42.20.20 DNS 2 8.8.8.8 or 8.8.4.4 or 1.1.1.1 i tried all.

ON some of my WLAN i noticed PMF was set to capable and the Multi Cast broad cast was set to 100% and IPV6 was Multi cas to Uni cast was enabled.

I checked my other working sites which were recently upgraded as well to latest OC300 V1.0 5.9.32 from 5.7.6, and found both these values of PMF, Multicast were disabled , so i disabled them on this site as well.

Fig 3 below:

After doing all the above , the network connection to clients was still unstable and very jerky, connects, browses for a few seconds and looses connection and reconnects (REPEAT). WIFI CONNECTION was steady but connection to internet was lost, failed to browse.

I than set the Internet connection Refer fig 4 DNS server 1 and 2 to 213.42.20.20 and 8.8.8.8 respectively. Same issue no change.I had even called the ISP help desk to reset there back end / Source connections to this site 2ice and reboot the ONT and refresh the whole system during the trouble shooting.

The MTU value was set to 1492 on all my site and it was same here. I decreased the MTU Value to 1450 fearing packet drops may be the issue with this particular site.

to my surprise it worked, and the browsing and connection was now steady and the site clients were able to browse consistently with out jitter in the connections or downloads or uploads. but the graph was still inverted!! Refer Fig 4.

Fig 4 below:

I eventually found that decreasing the MTU value from 1492 to 1450 helped to stabilize the network connection, and allowed site clients to browse consistently without any issues. However, the graph of TX and RX remained inverted, which is concerning. I will be observing the Original issue of not being able to connect to OC300 V1.0 in one attempt for a few days and will get back to you with update of intermittent connection issue.

I'm not satisfied with my troubleshooting results and would appreciate your insight and recommendations on what has caused the inverted graph and how can i fix this (Do i need to start another Thread?).

Thank you for your help.

  @Tedd404 @Hank21

I'm not certain whether the issue is a blocking one, given that all other sites are using the same DNS and performing well without any problems.

Recently, I noticed that the graphs on all my sites have inverted, with the TX and RX swapped. At first, I only focused on this issue on one site and didn't check the others. I assumed that I had made a mistake and caused the swap, but now I believe it's likely related to an issue with the 5.9.32 update.

I realized that I was "beating a dead horse" by trying to find a solution to this problem, as it is not something that I can fix on my end. Hopefully, this issue will be resolved in subsequent updates to the SDN controller.

I contacted the ISP and discovered that the only difference between this site and the others was the managed router they provided, which resulted in me having a dynamic internet connection type instead of PPOE. I asked them to reconfigure the DMZ settings on their managed router and enable DMZ. The site is now very responsive, and I haven't experienced any intermittent connection issues so far. Next, I plan to change the DNS to Google DNS and Cloud flare and observe if this is possible now that the DMZ is enabled. For now, I'll leave the site as is, running for two days and see if the intermittent connectivity issue has been resolved by the DMZ setting on the ISP-provided managed router.

Thank you for your prompt responses and support.

Fig below shows the swap post upgrade to 5.9.32 (OC300 V1.0) - OMADA APP

Best Regards