mirror of
https://github.com/OpenEPaperLink/OpenEPaperLink.git
synced 2026-03-21 00:04:28 +01:00
[GH-ISSUE #54] blockrequest: couldn't find taginfo 0000000000000000, tags not updating #570
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @bartbh on GitHub (May 31, 2023).
Original GitHub issue: https://github.com/OpenEPaperLink/OpenEPaperLink/issues/54
Describe the bug
After flashing the lastest version of the esp32 flasher on an esp32 (simple AP version), the tags won't receive the new image. I've tried multiple tags, all with their own, original mac address. In this example I use 0000021C5D4E3B14 as an example.
To Reproduce
Expected behavior
A new image on the selected tag. As it did in earlier versions.
Error on webpage
20:28:10 Updating 0000021C5D4E3B14
20:27:37 Tag not found, this shouldn't happen.
20:27:36 Tag not found, this shouldn't happen.
20:27:36 Tag not found, this shouldn't happen.
20:27:36 Tag not found, this shouldn't happen.
20:26:54 Tag not found, this shouldn't happen.
20:26:53 Tag not found, this shouldn't happen.
20:26:52 Tag not found, this shouldn't happen.
20:26:11 Tag not found, this shouldn't happen.
20:26:11 Tag not found, this shouldn't happen.
20:26:10 Tag not found, this shouldn't happen.
20:25:29 Tag not found, this shouldn't happen.
20:25:28 Tag not found, this shouldn't happen.
20:25:27 Tag not found, this shouldn't happen.
20:25:27 Tag not found, this shouldn't happen.
20:24:40 new image: /current/0000021C5D4E3B14.pending
20:24:40 Updating 0000021C5D4E3B14
Error in serial console:
<ADR 0000021C5D4E3B14
blockrequest: couldn't find taginfo 0000000000000000
blockrequest: couldn't find taginfo 0000000000000000
blockrequest: couldn't find taginfo 0000000000000000
<ADR 0000021A0FBF3B1B
Software versions
FW version tag: 0.1.5-clut
FW version ap: 0016
version esp32 ap: 30-5 22:55 (CEST) downloaded
tagDB.json.txt
@jjwbruijn commented on GitHub (May 31, 2023):
Thank you for your detailed report!
Good news: There really isn't an awful lot that could go wrong here; I can only see this happening if either the tag sends a block request with a 0-mac, or if the AP-tag somehow zeroes the mac upon receipt.
Bad news, I'm not sure what happens, here. Have you tried force-flashing the AP to the same version again, and maybe reprogram the ESP32 with a flash-erase?
To zero down on the issue, would you mind compiling and running the blockrequest-debug firmware? It'll show a dump of the blockrequest messages as sent by the AP.
https://github.com/jjwbruijn/OpenEPaperLink/tree/blockrequest-debug
It'll barf up messages like these:
@jjwbruijn commented on GitHub (May 31, 2023):
Also, since you're currently the only one experimenting with a 2.9" tag as AP: please try to run the 1.54" AP or nodisplay firmware, and see if that helps
@bartbh commented on GitHub (May 31, 2023):
I've tried 2 different esp32 boards with the same result. Also different tags, with different fw version (0.1.5 and 0.1.3).
So I'd suspect it must be the combination esp32/AP-tag.
I'll force flash the AP tag again, with different versions (2.9 and 1.54). And then try debug firmware.
@bartbh commented on GitHub (Jun 4, 2023):
Hereby the error from the debug firmware
@jjwbruijn commented on GitHub (Jun 9, 2023):
Have you tried using the 1.54" AP firmware? This debug log is very interesting :) It's either the AP or a bad tag; since it's at least working for one tag, I suspect one tag that's bad
@bartbh commented on GitHub (Jun 9, 2023):
No, not yet. Just found out both of my esp32's we're giving brown out errors at startup. Even with no pins attached and only the esp32 AP firmware.
So I want to swap them with a mini S2 first. To rule out a faulty esp32.
@bartbh commented on GitHub (Jun 11, 2023):
Just swapped the doit esp32 for an esp32 s2 and the AP and the tags are working fine now. So I suspect a faulty (or well, 2) esp32, as reverting back to the older version of the esp32 AP didn't resolve the issue.