zigbee stack reset coordinator

[Litvinus]

hi!

when resetting the coordinator (zb_bdb_reset_via_local_action(0)), an error occurs, the program hangs. It looks like the "zb_address_unlock" function. The same when using the command "bdb factory_reset" in the cli example. I use ver 4.1.0. 

[Elyxoft]

Hi, Thank's for the answer.

I use soft device S140 for ble part and with a button I form a network as a router with distribute security:

"

  NRF_LOG_INFO("Set role as router for ditribute network");
  zb_set_network_router_role((1l << ZIGBEE_CHANNEL));
  zb_enable_distributed();
  zb_zdo_setup_network_as_distributed();
  NRF_LOG_INFO("Start network formation");
  bdb_start_top_level_commissioning(ZB_BDB_NETWORK_FORMATION);

"

Then with another button I try to leave by using :

"

zb_buf_get_out_delayed(zb_bdb_reset_via_local_action);

"

And I hang with call stack:

Now I try the workarround for zed but I am not one.

regards,

Sébastien

[Elyxoft]

Hi,

Workaround do not work for router.

Have you another one ?

regards,

Sébastien

[Marte Myrvold]

Hi Sébastien,

We do not have a workaround for routers or coordinators yet. I will update you if such a workaround is found.

Best regards,

Marte

[Atok]

Hello Marte

We also have this exact same problem in a custom application with nRF SDK 4.1.0

This is a severe issue, as being able to factory reset a zigbee device is an essential operation.

Please try to give us a workaround for Routers and Coordinators as well.

Best regards

[Atok]

We seem to have identified a workaround for Routers and Coordinators.

No side effects have been discovered so far.

  // WORKAROUND zb_bdb_reset_via_local_action fails due to a bug in zboss if this is not done. Reason unclear.
  zb_address_ieee_ref_t addr_ref;
  if (zb_address_by_short(ZB_PIBCACHE_NETWORK_ADDRESS(), ZB_FALSE, ZB_FALSE, &addr_ref) == RET_OK) {
    zb_address_delete(addr_ref);
  }

  zb_bdb_reset_via_local_action(0);
  zf_system_request_reboot(false, ZF_SYSTEM_CONFIG_STORE_REBOOT_DELAY);

Best regards

Markus