Was notified of firmware update yesterday for all of my cam 3s. All were successful but one of them failed and appears to have bricked the wifi in the process. I contacted support who told me so sad too bad, out of warranty, not our problem.
I pulled a new cam v3 out of a box today to replace the bricked one. It did 2 firmware updates not problem. The last firmware update bricked the camera in the exact same way.
Now I am out 2 cameras through no fault of my own.
DONE WITH WYZE.
It is most likely that the firmware itself was not the cause of the brickage but instead a momentary loss in the WiFi signal to the cam while the download was in progress that caused lost packets.
In the firmware update process, the WiFi is the weakest link.
For cams that experience an update failure like this, you can recover them by manually flashing the firmware.
Well if that’s the case, it may not be the issue of firmware, but it sounds like the upgrade process is flawed, as the upgrade process won’t protect the devices from unreliable Wi-Fi connections, and that sounds like a bigger issue.
100% on point. I’m not sure what, if any, file verification protocols are in place on the cam after a download and before the application of the update. I would like to believe that the logic in the firmware would first do a thorough inspection and server verification of the downloaded package prior to an attempt to apply it. But, as long as I have been here on the forum, I haven’t read any posts that discuss that topic.
My experience, however, is that some updates fail. I have done bulk updates on many cams and the majority of the time they succeed. But occasionally, one or two fail. It is very rare for me though since I am running my cams on what I would consider a rock solid network for the cams.
Moreover, it supports my theory that it was a file download failure to the affected cams when multiple cams (16 V3 in my case), running identical firmware versions, undergo the exact same firmware update and it updates perfectly on the majority of the cams. To support this further, the cams will then successfully update by manual firmware download and flash to the exact firmware version that they were downloading during the OTA update process. This supports the theory that it isn’t the firmware but some random failure of either the download or install process.
One of my theories is that the bulk update process is part of the problem, blasting bits and bytes all over the 2.4GHz WiFi band to every cam at once. If the Router isn’t 100% on board and doing its expected duties, who knows what mayhem can be happening. Think US Postal Service during holiday rush. How many packages do they loose or destroy? Packet loss over wireless transfer is real.
With that in mind, and since Wyze can’t rely on the stability or effectiveness of all the cheap, poorly coded routers out there, I have requested that Wyze introduce logic that will complete Bulk Firmware updates progressively, one single cam at a time. Everybody in a nice, neat, que line being served one at a time rather than what appears to be the current shotgun blast approach.
@SlabSlayer , thank you, glad to join the forum. One of my wyze devices also got bricked during last night’s upgrade, thus lead me finding this forum.
Thanks for the detailed info and your theories.
Regardless whether the network/power/download server/devices/local network is reliable or not (any of them could fail), there should be a roll back mechanism which should automatically roll back the upgrade if it fails in the middle. Otherwise more or less devices will be bricked, sooner or later.
I am not sure if the number of devices getting upgrades at the same time plays a factor in the upgrade-failure-then-bricked scenario. I double it would be a big factor. You mentioned 2.4Ghz wifi and pointed out that could be the potential issues. I believe the network protocols in different layers should handle them gracefully, otherwise you’d see glitches when viewing live video here and there.
To me, to support rolling back to the original state is the #1 priority to fix this kind of issues.
Agreed. Having a failsafe fallback would be helpful. But that would require the cam to have two copies of firmware onboard simultaniously for the unfortunate circumstance that the update fails and it needs to fall back. I’m not sure the memory chip has enough space for the duplication. I will have to check the Wishlist to see if that has been suggested.
Thanks. The latest version Cam 3 firmware is 8.97 MB (9,412,736 bytes). I assume it isn’t a big size for today’s cam, given the size of SD card it can control is in GBs.