Good day.
I am currently developing with two NRF52840 DK boards, and I also have two NRF Dongles. On a dongle a sniffer for ZigBee. I am working with SDK 3.2. This is my first project with ZigBee, so I don't have much know-how yet.
My goal is to implement the coordinator as like CLI agent. So that it subscribes to the multi-sensor and receives data. Is so to speak an A to B communication.
With "zb_zdo_match_desc_req()" I get the short address and the endpoint of the sensor. Then with "zb_zdo_ieee_addr_req()" I get the long address (EUI64) of the sensor. After that the "zb_zdo_bind_req()" request starts (Between coordinator and sensor) and in the callback function ("p_resp->status") this is confirmed successfully with ZB_ZDP_STATUS_SUCCESS. However, I only connect to the ZB_ZCL_CLUSTER_ID_TEMP_MEASUREMENT cluster.
My questions:
1) Is it necessary to connect both clusters (temperature/pressure) to subscribe to the multisensor? If so, is the bind request simply done again with the ZB_ZCL_CLUSTER_ID_PRESSURE_MEASUREMENT?
2) Is subscribing the sensor in the first step as below, correct?
// Step 8:
// Subscribe on Sensor
static void subscribe_device(zb_uint8_t param)
{
zb_buf_t * p_buf = ZB_BUF_FROM_REF(param);
configure_reporting_req_t req;
tsn_ctx_t * p_tsn_cli;
zb_uint8_t * p_cmd_ptr;
zb_ret_t zb_err_code;
//zb_bool_t subscribe;
zb_ieee_addr_t sensor_ieee_adr;
NRF_LOG_INFO("----------------------------------------");
NRF_LOG_INFO("Step 8: subscribe_device");
req.profile_id = ZB_AF_HA_PROFILE_ID;
req.cluster_id = ZB_ZCL_CLUSTER_ID_TEMP_MEASUREMENT; //ZB_ZCL_CLUSTER_ID_PRESSURE_MEASUREMENT; //
req.attr_id = ZB_ZCL_ATTR_TEMP_MEASUREMENT_VALUE_ID;
req.attr_type = ZB_ZCL_ATTR_TYPE_S16; //ZB_ZCL_ATTR_TYPE_16BIT;
req.interval_min = ZIGBEE_CONFIGURE_REPORT_DEFAULT_MIN_INTERVAL;
req.interval_max = ZIGBEE_CONFIGURE_REPORT_DEFAULT_MAX_INTERVAL;
req.remote_node.addr_short = m_device_ctx.bulb_params.short_address;
req.remote_addr_mode = ZB_APS_ADDR_MODE_16_ENDP_PRESENT; //= parse_address(addr_buf, &req.remote_node, ADDR_ANY);
req.remote_ep = m_device_ctx.bulb_params.endpoint;
p_buf = ZB_GET_OUT_BUF();
NRF_LOG_INFO("Configure Reporting");
p_tsn_cli = get_free_tsn_ctx();
if (!p_tsn_cli)
{
NRF_LOG_ERROR("Too many configure reporting requests");
ZB_FREE_BUF(p_buf);
return;
}
// Configure new tsn context.
p_tsn_cli->taken = true;
//p_tsn_cli->p_cli = p_cli;
p_tsn_cli->tsn = ZCL_CTX().seq_number;
ZB_ZCL_GENERAL_INIT_CONFIGURE_REPORTING_SRV_REQ(p_buf,
p_cmd_ptr,
ZB_ZCL_ENABLE_DEFAULT_RESPONSE);
ZB_ZCL_GENERAL_ADD_SEND_REPORT_CONFIGURE_REPORTING_REQ(p_cmd_ptr,
req.attr_id,
req.attr_type,
req.interval_min,
req.interval_max,
ZIGBEE_CONFIGURE_REPORT_DEFAULT_VALUE_CHANGE);
ZB_ZCL_GENERAL_SEND_CONFIGURE_REPORTING_REQ(p_buf,
p_cmd_ptr,
req.remote_node,
req.remote_addr_mode,
req.remote_ep,
m_device_ctx.co_params.endpoint, //CONTROLL_ENDPOINT
req.profile_id,
req.cluster_id,
NULL);
NRF_LOG_INFO("Report sended Waiting for Values");
//zb_err_code = ZB_SCHEDULE_ALARM(zb_subscribe_timeout, p_tsn_cli->tsn, ZIGBEE_CLI_CONFIGURE_REPORT_RESP_TIMEOUT * ZB_TIME_ONE_SECOND);
if (zb_err_code != RET_OK)
{
NRF_LOG_ERROR("Unable to schedule timeout timer");
invalidate_ctx(p_tsn_cli);
}
NRF_LOG_INFO("End");
}
3) Does it even need "cmd_zb_subscribe_unsubscribe_cb" and what is that unsubscribe for?
My problems:
1) The "ep_handler_report" is not called, but I see in Wireshark that a Report Attributes has been sent?
Is there still something to change at the sensor, because there I have left everything as is.
My ep_handler_report Handler:
static zb_uint8_t ep_handler_report(zb_uint8_t param)
{
NRF_LOG_INFO("Ep Handler Report");
zb_buf_t * p_zcl_cmd_buf = (zb_buf_t *)ZB_BUF_FROM_REF(param);
zb_zcl_parsed_hdr_t * p_cmd_info = ZB_GET_BUF_PARAM(p_zcl_cmd_buf, zb_zcl_parsed_hdr_t);
tsn_ctx_t * p_tsn_ctx;
if (p_cmd_info->cmd_id == ZB_ZCL_CMD_REPORT_ATTRIB)
{
NRF_LOG_INFO("print_attr");
print_attr(p_cmd_info, param);
ZB_FREE_BUF_BY_REF(param);
return ZB_TRUE;
}
else if (p_cmd_info->cmd_id == ZB_ZCL_CMD_CONFIG_REPORT_RESP)
{
// Find command context by ZCL sequence number.
p_tsn_ctx = get_ctx_by_tsn(p_cmd_info->seq_number);
if (p_tsn_ctx != NULL)
{
NRF_LOG_WARNING("Unsubscribe cb");
cmd_zb_subscribe_unsubscribe_cb(p_tsn_ctx, param);
return ZB_TRUE;
}
}
NRF_LOG_INFO("False Handler report");
ZB_FREE_BUF_BY_REF(param);
return ZB_FALSE;
}
/* Register callback for handling ZCL commands. */
ZB_AF_SET_ENDPOINT_HANDLER(m_device_ctx.bulb_params.endpoint, ep_handler_report);
The other functions as well as the output of the data I took from the CLI as well as from other discussions.
With the function "print_attr()" the data is output (Same function as with CLI), what am I doing wrong here with subscribe?
2) After a while the network crashes, any idea what this is? The Connected LED on the sensor goes out.
I would also be glad if you could give me a quick rundown on the process for subscribing to devices. If you want the whole code, I can share it with pleasure.
Thanks in advance. 
With kind regards
Jonas T
