ESP32 BLE server supports only 1 bonded device

[oumayma-ba]

Answers checklist.

• I have read the documentation ESP-AT Programming Guide and the issue is not addressed there.
• I have used the latest released firmware or have updated my ESP-AT branch (master or release) to the latest version and checked that the issue is present there.
• I have searched the issue tracker for a similar issue and not found a similar issue.

General issue report

AT version:4.2.0.0-dev
ESP32-C6
What is the expected behavior?
ESP32 supports up to 13 bonded devices.
What is the actual behavior?

• Connect to esp32 with smartphone 1 and perform the pairing with bonding.
• Disconnect.
• Connect to esp32 with smartphone 2and perform the pairing with bonding.
• Disconnect.
• Reconnect with smartphone 1 : Connection error iOS msg : peer removed pairing information

[XionHsieh1025]

Hi @oumayma-ba

Is the firmware you're using an official release firmware built via CI compilation, or one you recompiled yourself? Have you modified any configuration options? If neither applies, could you provide the complete output of the AT+GMR command?

[oumayma-ba]

Hi @XionHsieh1025 , It's a custom build not an official release.

• Branch: master
• Commit id: 4416bb7
• Git repo link: https://github.com/espressif/esp-at
• The latest official release is v4.1.1.0, which has a known BLE bonding issue: ESP32-C6 BLE bonding lost after reconnect (iOS, AT v4.1.1.0) #1004
• This development version includes fixes available in the master branch, including the bonding issue.
  AT+ GMR response is: version:4.2.0.0-dev(ae01d89 - ESP32C6 - Apr 7 2026 12:38:06)
• No special config or changes were applied.

[XionHsieh1025]

Hi @oumayma-ba

The issues mentioned in our earlier discussions have all been fixed and merged into the master branch. I have tested the master branch, and the limitation that only one device could be bonded should no longer exist. Would it be acceptable for you to temporarily use the master branch for the time being? Below is the version information I tested:

AT+GMR
AT version:4.2.0.0-dev(7b6d97b - ESP32C6 - Jun 26 2026 09:11:02)
SDK version:v5.4.4
compile time(fc3628b4):Jul  1 2026 11:54:22
Bin version:v4.2.0.0-dev(ESP32C6-4MB)

OK

[oumayma-ba]

Hi @XionHsieh1025 ,

I generated a new firmware version for the ESP32-C6 from the master branch, but the issue persists.

This is the AT firmware version:

AT version:4.2.0.0-dev(7b6d97b - ESP32C6 - Jun 26 2026 09:11:02)
SDK version:v5.4.4

[XionHsieh1025]

Hi @oumayma-ba

tried reproducing the issue you reported again, but I still couldn’t reproduce it on my side. There may be some differences in the setup or the exact steps taken, so could you help review whether my reproduction steps need to be adjusted?

Here are the steps I followed:

AT+GMR
AT version:4.2.0.0-dev(7b6d97b - ESP32C6 - Jun 26 2026 09:11:02)
SDK version:v5.4.4-dirty
compile time(fc3628b4):Jul  8 2026 11:22:52
Bin version:v4.2.0.0-dev(ESP32C6-4MB)

OK
AT+BLEINIT=3

OK
AT+BLEADVPARAM=50,50,0,0,7

OK
AT+BLEADVDATA="020102030398FE171698FE647265616D6562632D3131303933393633372D30"

OK
AT+BLEADVSTART

OK
AT+BLEENCDEV?

OK
+BLECONN:0,"58:1a:16:92:51:31"
+BLESETPHY:"58:1a:16:92:51:31",2,2
+BLECFGMTU:0,203
AT+BLEENC=0,3

OK
+BLESECNCREQ:0,153147
AT+BLECONFREPLY=0,1

OK
+BLEAUTHCMPL:0,0
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"

OK
+BLEDISCONN:0,"58:1a:16:92:51:31"
AT+BLEADVSTART

OK
+BLEAUTHCMPL:0,0
+BLECONN:0,"b4:ae:c1:34:9b:a3"
+BLESETPHY:"b4:ae:c1:34:9b:a3",2,2
+BLECFGMTU:0,203
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"

OK
+BLEDISCONN:0,"b4:ae:c1:34:9b:a3"
AT+BLEADVSTART

OK
+BLECONN:0,"84:2f:57:18:cd:6c"
+BLESETPHY:"84:2f:57:18:cd:6c",2,2
+BLECFGMTU:0,203
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"

OK
AT+BLEENC=0,3

OK
+BLESECNCREQ:0,684257
AT+BLECONFREPLY=0,1

OK
+BLEAUTHCMPL:0,0
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"
+BLEENCDEV:1,"84:2f:57:18:cd:6c"

OK
+BLEDISCONN:0,"84:2f:57:18:cd:6c"
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"
+BLEENCDEV:1,"84:2f:57:18:cd:6c"

OK
AT+BLEADVSTART

OK
+BLECONN:0,"84:2f:57:18:cd:6c"
+BLESETPHY:"84:2f:57:18:cd:6c",2,2
+BLEAUTHCMPL:0,0
+BLECFGMTU:0,203
AT+BLEADVSTART

OK
+BLECONN:1,"b4:ae:c1:34:9b:a3"
+BLESETPHY:"b4:ae:c1:34:9b:a3",2,2
+BLEAUTHCMPL:1,0
+BLECFGMTU:1,203
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"
+BLEENCDEV:1,"84:2f:57:18:cd:6c"

OK
AT+BLECONN?
+BLECONN:0,"84:2f:57:18:cd:6c"
+BLECONN:1,"b4:ae:c1:34:9b:a3"

OK
+BLEDISCONN:1,"b4:ae:c1:34:9b:a3"
+BLEDISCONN:0,"84:2f:57:18:cd:6c"
AT+BLEENCDEV?
+BLEENCDEV:0,"b4:ae:c1:34:9b:a3"
+BLEENCDEV:1,"84:2f:57:18:cd:6c"

OK

Could you also confirm a few details on your side?

1. Configuration – Did you change any sdkconfig / menuconfig options, or use any custom project configuration?
2. Firmware – Is the firmware you tested built locally from source, or are you using a prebuilt binary (e.g. from ESP-AT release packages)? If local, please share the branch/commit and build steps if possible.

With those details, we can align the environment and narrow down what might be different. Thanks!