Omada Hardware Controller v5.5.7 issues
This thread has been locked for further replies. You can start a new thread to share your ideas or ask questions.
I installed the newly released Omada Hardware Controller v5.5.7 firmware on 2 September. It seemed to run Ok until this morning. When I say, 'OK', the OC200 has been losing connection/readopting a device every few days, but not constantly.
Overnight, the OC200 started dropping, reconnecting, and readopting all of my equipment. Then it would do it again a few minutes later. I unplugged the OC200 from its POE switch, waited a few minutes, then replugged it in again. Within an hour, it started the dropping/reconnecting/readopting loop again. I was forced to revert to V5.4.7 and rebuild from my backup from 1 September. It's now been back up for about an hour and 30 minutes without issue.
Here.s my equipment - all except the OC200 now running the latest firmware:
- TP-Link Router (TL-R605) (V1)
- TP-Link Hardware Controller (OC-200) (V1)
- TP-Link 8-Port Gigabit Managed PoE Switch (TL-SG2210P) (V3.20)
- (2) TP-Link Managed 16-Port Switches (TL-SG2218)
- (3) TP-Link EAP245 Access Point (V3)
- TP-Link EAP 225 Outdoor (V1)
Is anyone else experiencing any difficulties with v5.5.7?
- Copy Link
- Subscribe
- Bookmark
- Report Inappropriate Content
Rebooting everything (all devices) seems to have done the trick - both for me and another person with similar issues on Reddit. It has now been 24 hours since I rebooted everything and there has not been a single error since. I'll mark this as the solution.
I guess that may mean that a reboot is needed for every firmware update (?)
- Copy Link
- Report Inappropriate Content
Update:
After downgrading the OC200's firmware to v5.4.7 (as per my original post), the looping drop/connect/readopt cycles started again. I then remembered that I upgraded the TL-SG2210P switch's firmware from 3.20.2 to 3.20.4 on the same day as I upgraded the OC20's firmware - and everything is connected to this switch. I have now downgraded the switch to 3.20.2 and am waiting to see if the looping drop/connect/readopt cycles start again. If they stop, it will mean that it's the switch firmware 3.20.4 that's causing the issue.
I also checked the power consumption on the POE ports of the TL-SG2210P that are currently providing power to (4) EAPs and the (1) OC200. It's only running at 28-30% of capacity. So, it shouldn't be overtaxed. The other ports are not using POE power.
If this TL-SG2210P firmware downgrade fails to resolve the issue, it will probably mean either the switch or the OC200 hardware is failing. I believe if we get here, it'll be the OC200 that's failing because other than a slowdown with all the looping drop/connect/readopt cycles, I never lost my internet or internal LAN/VLAN connectivity.
Any other ideas?
- Copy Link
- Report Inappropriate Content
@lflorack Ditto here!! I see all my devices lose connection here and there. Last night both my AP's lost their connection. Yesterday it was my 2210...
Would love to go back to 5.4.7
- Copy Link
- Report Inappropriate Content
- Copy Link
- Report Inappropriate Content
Update 2:
After downgrading both the OC200's firmware to v5.4.7 (in my update #1) and now the TL-SG2210P switch's firmware to 3.20.4 the looping drop/connect/readopt cycles continued overnight. From 22:25 last night until 08:18 this morning (~10 hours), here's a count of drop/connect/readopt cycles for each of my devices:
| Device drop/connect/readopt cycles | ||
| 1. Router/Gateway | TL-R605 (V1) | 6 |
| 2. AP - Study (POE) | EAP 245 (V3) | 6 |
| 3. AP - AV/LAN Closet (POE) | EAP 245 (V3) | 5 |
| 4. Controller (POE) | OC200 (V1) | N/A |
| 5. Switch - Main (POE Switch) | TL-SG2210P (V3.20) | 7 |
| 6. AP - Porch (POE) | EAP 225 Outdoor (V1) | 5 |
| 7. AP - Kitchen (POE) | EAP 245 (V3) | 5 |
| 8. Switch - Study | TL-SG2218 | 2 |
| 9. Switch - AV/LAN Closet | TL-SG2218 | 3 |
| Total | 39 |
So, it looks like there were (7) events overnight that affected all of the devices to varying amounts. Therefore, the problem persists - in spite of reverting to the previous firmware versions for both the OC200 controller and the TL-SG2210P main switch. Although these drop/connect/readopt cycles were occurring before yesterday, they were happening less frequently - probably once every day or two.
I believe all of this means that either the TL-SG2210P switch or the OC200 controller is failing. I will attempt to move some of the POE devices off of the TL-SG2210P (POE) main switch and see if that helps. If not, I'll have to replace either the OC200 (likely) or the TL-SG2210P.
Any other suggestions?
- Copy Link
- Report Inappropriate Content
See my update #2. In my case, it doesn't appear that it's the new firmware for either the OC200 or the 2210P that's causing the dropping/connecting/readopting device issues. But, if you decide to revert to the earlier firmware on the OC200, make sure you have a backup ready because downgrading will wipe out your settings.
- Copy Link
- Report Inappropriate Content
@lflorack I just completed the rollback of the 2210P to 20.2 from 20.4 and gonna let that sit for a while to see what happens.
I didnt rollback the OC200 yet since as you said, I need to make sure I have a good backup. I do auto backups every night so what I did do..and perhaps somewhat unorthodox, is to take the 5.4.7 config I have from Friday and just restored that on top of 5.5.7 to see if anything changes...with the last choice being to reset the OC200 and roll it back.
Little of a pain to do the 2210P since I had to un-adopt it first. Even though I had the file and tried to upload and upgrade, the system would not let me until I took it out of the controller and then I could do the downgrade. The adopted it back again.
Should know more tomorrow if it acts up. But I believe I did note that the controller did not note any update available. I just found it on the website and decided to download and install on my own and not let the controller do it...
- Copy Link
- Report Inappropriate Content
@Fleegle61 said
"Little of a pain to do the 2210P since I had to un-adopt it first. Even though I had the file and tried to upload and upgrade, the system would not let me until I took it out of the controller and then I could do the downgrade. The adopted it back again. "
Hmm. I wonder why you needed to do it that way... I updated my 2210P via the OC200 and it went fine without doing anything special.
Anyway, regarding the drop/connect/readopt cycle issue, my latest attempt was to just reboot all my devices. It's been 4.5 hours without any issues. I'll post again on how long it lasts.
- Copy Link
- Report Inappropriate Content
Update #3
At the risk of jinxing myself....
At 6:30 pm (EDT) last night, I rebooted all of my Omada devices from the OC200's device screen. Everything came back up, reconnected, and readopted fine. It is now 07:41 (EDT) and there hasn't been a single drop, connect, readopt cycle since the mass reboot. I know it's a short time but honestly, if it was still like yesterday, I'd have over 200 notifications by now.
Good luck!
- Copy Link
- Report Inappropriate Content
@lflorack I woke up this morning and had the disconnects in the event log. So just rolling back 2210 to 3.20.2 didnt do it.
Since I am off today, guess I will start on the OC200. It does autobackups to USB so I hope it will go smoothly.
@Fae may need to find out whats going on since there is at least 2 of us having the exact same issue with 5.5.7 and 3.20.4 causing the random disconnects on all devices.
- Copy Link
- Report Inappropriate Content
Did you try rebooting everything like I did?
- Copy Link
- Report Inappropriate Content
Information
Helpful: 0
Views: 3738
Replies: 29
Voters 0
No one has voted for it yet.