RE: FAILURE response when reading RebootCount in GeneralDiagnosticsCluster

chiba — Sat, 07 Feb 2026 00:16:15 GMT

Thank you, your help.

RE: FAILURE response when reading RebootCount in GeneralDiagnosticsCluster

Edvin — Fri, 06 Feb 2026 09:21:01 GMT

Hello,

It seems like it is in fact a bug. When the last fabric is removed, the factory reset is essentially performed. Since we remove all of the data, we also remove the reboot count variable - but we do not reboot ourselves due to KConfig, and we don't re-initialize the setting. Since the setting does not exist, the error is returned.

The solution is to either reboot the board once (the reboot count will be re-initialized on the reboot), or change the implementation of the last fabric removal:

diff --git a/samples/matter/common/src/app/fabric_table_delegate.h b/samples/matter/common/src/app/fabric_table_delegate.h
index 5e914cd4cf..b9dc2d8327 100644
--- a/samples/matter/common/src/app/fabric_table_delegate.h
+++ b/samples/matter/common/src/app/fabric_table_delegate.h
@@ -60,6 +60,8 @@ private:
                                chip::DeviceLayer::PersistedStorage::KeyValueStoreMgrImpl().DoFactoryReset();
                                /* Erase Network credentials and disconnect */
                                chip::DeviceLayer::ConnectivityMgr().ErasePersistentInfo();
+                               /* Make sure Reboot Count is set to 1 */
+                               chip::DeviceLayer::ConfigurationMgr().StoreRebootCount(1);
#ifdef CONFIG_CHIP_WIFI
                                chip::DeviceLayer::WiFiManager::Instance().Disconnect();
                                chip::DeviceLayer::ConnectivityMgr().ClearWiFiStationProvision();

The above change works for me (the person looking into this from the Matter Team), and I don't see the error anymore. Though the reboot count is reset to 1. The user might read it before ErasePersistentInfo() and set it again to the same value if they prefer.

There is a task started internally to fix this in a future release, but I do not have details on exactly how it will be implemented.

Best regards,

Edvin

RE: FAILURE response when reading RebootCount in GeneralDiagnosticsCluster

Edvin — Tue, 27 Jan 2026 09:48:00 GMT

Hello,

Sorry for the late reply. I need to run this by our Matter team. I will get back to you as soon as I know more.

Best regards,

Edvin