TL-SG2210P, Stability Issues after upgrade to 5.0.4
This thread has been locked for further replies. You can start a new thread to share your ideas or ask questions.
We have Omada devices on three sites. We recently updated the firmware on 4 out of 5 TL-SG2210P
from version 5.0.0 to 5.0.4 – since then, these 4 switches show frequent disconnects.
- Heartbeat missed about every 10 minutes.
- Disconnected, followed by Adopting like every 60 minutes.
We’ve tried to downgrade the firmware to 5.0.3 and 5.0.2 (older versions are not available on the website) – the issue remains.
The only device that is running stable is the one still sitting at version 5.0.0.
Does anyone encounter similiar issues?
Any chance to get these switches downgraded to 5.0.0?
- Copy Link
- Subscribe
- Bookmark
- Report Inappropriate Content
dognose wrote
We have Omada devices on three sites. We recently updated the firmware on 4 out of 5 TL-SG2210P
from version 5.0.0 to 5.0.4 – since then, these 4 switches show frequent disconnects.
- Heartbeat missed about every 10 minutes.
- Disconnected, followed by Adopting like every 60 minutes.
We’ve tried to downgrade the firmware to 5.0.3 and 5.0.2 (older versions are not available on the website) – the issue remains.
The only device that is running stable is the one still sitting at version 5.0.0.
Does anyone encounter similiar issues?
Any chance to get these switches downgraded to 5.0.0?
Dear @dognose
To better assist you, I've created a support ticket via your registered email address, and escalated it to our support engineer to look into the issue.
The ticket ID is TKID221213185, please check your email box and ensure the support email is well received. Thanks!
- Copy Link
- Report Inappropriate Content
- Copy Link
- Report Inappropriate Content
It depends on your environment i'd say.
The szenario where I encountered these Issues was the following:
- 3 sites
- connected through a Site2Site-VPN
- Omada Controller located on ONE Site only.
Together with the support we analyzed the issue for a while, but weren't able to figure out the root-cause.
Devices of Version 5.0.4 where perfectly working, answering on ping-requests - but when trying to access their http-page connection
timed out / they didn't respond. Devices of prior versions respond fine and show the "This device is managed by omada controller" page. (Through VPN - locally they worked as expected)
The heartbeat functionality was changed in that version, and maybe the new version encounters the same timeouts then.
-----
For me, the situation was kind of resolved because during all the tests I "accidently" missconfigured the vpn on a router on one site (which had a working heartbeat) and then noted, that in this situation the device is basically totally bricked. No VPN means no access to omada controller. No access to Omada Controller means no reconfiguration of the device. Had to completly reset a router, then first get a VPN running in Standalone mode - just to adopt the device again and reconfigure it once more in controller mode.
That is unacceptable for real-world-usage so I decided to add a dedicated controller per site. And that in turn (obviously) resolved the heartbeat-issues I was facing through VPN.
----
IMHO, before the Multi-Site-Management is really usable, TP-Link need to add the possibility to provide Controller-Mirrors on either site (like Domain-Controller-Replication) so any site stays manageable in case of controller-disconnects.
But that's complaining on a high level, given the low costs of a oc-200 (or even the possibility to run it free on a VM) I'm fine with running a dedicated controller per site.
Having every device just beeing local-network wired I'm running everything on the newest version without any (knwon) issues.
Hope this helps you to decide If you can upgrade or not :)
- Copy Link
- Report Inappropriate Content
Very thoughtful response, thank you!
I have a similar setup to your evolved one, ie two sites, connected via L2TP/IPSEC VPN. I have an OC200 at the remote site because it has a number of meshed APs and I cannot risk extended downtime in the event of an issue with the VPN tunnel or the local site (which runs a software controller in a Docker container). Sounds like I would not experience this issue, since I have a controller per site.
I would like to explore the idea of a controller 'mirror'...because with the new 'Global' view in Controller 5.8.4, it would be nice if one controller was the 'dominant' and other sites became 'subservient', or at least allowed their current status to roll up to the dominant for reporting. I agree that the OC200 is sized/priced appropriately enough to be the 'subservient' or 'mirror' controller and capable enough to manage a few remote sites all on its own too.
- Copy Link
- Report Inappropriate Content
Information
Helpful: 0
Views: 772
Replies: 4
Voters 0
No one has voted for it yet.