# Error: too many failed ADR requests

**URL:** <https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962>\
**Category:** LMIC\
**Created:** [June 11, 2018, 3:58pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962 "2018-06-11T15:58:50Z")\
**Posts on this page:** 10\
**Page:** 1

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 11, 2018, 3:58pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/1 "2018-06-11T15:58:51Z")

</div>

Hi all,  
i’m currently testing the ADR of my new node, but it seems not working as i expected. After 10 packages it seems that the ADR process is started, but not responded or recognized from the LMIC library. After a few packages i become the “error: too many failed ADR requests”. See the screenshot from my gateway:

 ![image](https://www.thethingsnetwork.org/forum/uploads/default/original/2X/f/fb7b2894f12e77358ba36866795f6c3296a19e9f.png)

Here is the code for init the LMIC:

> void setup()  
> {  
> uint8\_t mcusr = MCUSR;  
> MCUSR = 0;
> 
> ```
> Serial.begin(115200);
> debugPrintLn(F("Boot")); 
> 
> showBootStatus(mcusr);
> 
> // LMIC init
> os_init();
> // Reset the MAC state. Session and pending data transfers will be discarded.
> LMIC_reset();
> 
> LMIC_setClockError(MAX_CLOCK_ERROR * 1 / 100);
> 
> LMIC_setupChannel(0, 868100000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(1, 868300000, DR_RANGE_MAP(DR_SF12, DR_SF7B), BAND_CENTI); // g-band
> LMIC_setupChannel(2, 868500000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(3, 867100000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(4, 867300000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(5, 867500000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(6, 867700000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(7, 867900000, DR_RANGE_MAP(DR_SF12, DR_SF7), BAND_CENTI); // g-band
> LMIC_setupChannel(8, 868800000, DR_RANGE_MAP(DR_FSK, DR_FSK), BAND_MILLI); // g2-band
> 
> LMIC_setLinkCheckMode(1);
> LMIC_setAdrMode(1);
> LMIC.dn2Dr = DR_SF9;
> debugPrintLn(F("S")); // Setup complete!"
> debugFlush();   
> 
> ```
> 
> }

This is my onEvent method:

> void onEvent (ev\_t ev)  
> {  
> debugPrint(os\_getTime());  
> debugPrint(": ");  
> debugPrintLn(ev);  
> switch(ev)  
> {  
> case EV\_SCAN\_TIMEOUT:  
> //debugPrintLn(F(“EV\_SCAN\_TIMEOUT”));  
> break;  
> case EV\_BEACON\_FOUND:  
> //debugPrintLn(F(“EV\_BEACON\_FOUND”));  
> break;  
> case EV\_BEACON\_MISSED:  
> //debugPrintLn(F(“EV\_BEACON\_MISSED”));  
> break;  
> case EV\_BEACON\_TRACKED:  
> //debugPrintLn(F(“EV\_BEACON\_TRACKED”));  
> break;  
> case EV\_JOINING:  
> //debugPrintLn(F(“EV\_JOINING”));  
> break;  
> case EV\_JOINED:  
> //debugPrintLn(F(“EV\_JOINED”));  
> break;  
> case EV\_RFU1:  
> //debugPrintLn(F(“EV\_RFU1”));  
> break;  
> case EV\_JOIN\_FAILED:  
> //debugPrintLn(F(“EV\_JOIN\_FAILED”));  
> break;  
> case EV\_REJOIN\_FAILED:  
> //debugPrintLn(F(“EV\_REJOIN\_FAILED”));  
> break;  
> case EV\_TXCOMPLETE:  
> debugPrintLn(F(“EV\_TXC”));  
> if (LMIC.txrxFlags & TXRX\_ACK)  
> debugPrintLn(F(“R ACK”)); // Received ack  
> if (LMIC.dataLen)  
> {  
> debugPrintLn(F(“R “));  
> debugPrintLn(LMIC.dataLen);  
> debugPrintLn(F(” bytes”)); // of payload  
> }  
> LMIC\_transmitted = 1;  
> break;  
> case EV\_LOST\_TSYNC:  
> //debugPrintLn(F(“EV\_LOST\_TSYNC”));  
> break;  
> case EV\_RESET:  
> //debugPrintLn(F(“EV\_RESET”));  
> break;  
> case EV\_RXCOMPLETE:  
> // data received in ping slot  
> //debugPrintLn(F(“EV\_RXCOMPLETE”));  
> break;  
> case EV\_LINK\_DEAD:  
> //debugPrintLn(F(“EV\_LINK\_DEAD”));  
> break;  
> case EV\_LINK\_ALIVE:  
> //debugPrintLn(F(“EV\_LINK\_ALIVE”));  
> break;  
> default:  
> //debugPrintLn(F(“Unknown event”));  
> break;  
> }  
> }

Where is the problem? Do i need to set some other flag or use another event. The serial monitor of the Arduino IDE shows only that a “EV\_TXCOMPLETE” is received by the LMIC.

Kind regards  
Stefan

---

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 11, 2018, 7:34pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/2 "2018-06-11T19:34:25Z")

</div>

Here some debug statements from the LMiC, setting debug level to 2.

> 9329234: engineUpdate, opmode=0x908  
> 9329311: Uplink data pending  
> 9329373: Considering band 0, which is available at 10930440  
> 9329610: Considering band 1, which is available at 9482879  
> 9329941: Considering band 2, which is available at 4950416  
> 9330273: Considering band 3, which is available at 0  
> 9330584: No channel found in band 3  
> 9330774: Considering band 0, which is available at 10930440  
> 9331111: Considering band 1, which is available at 9482879  
> 9331443: Considering band 2, which is available at 4950416  
> 9331788: No channel found in band 2  
> 9331978: Considering band 0, which is available at 10930440  
> 9332315: Considering band 1, which is available at 9482879  
> 9332648: Airtime available at 9482879 (channel duty limit)  
> 9332979: Airtime available at 9576587 (global duty limit)  
> 9333305: Uplink delayed until 9576587  
> 9333524: Scheduled job 0x449, cb 0x2007 at 9576462
> 
> 9576572: Running job 0x449, cb 0x2007, deadline 9576462  
> 9576691: engineUpdate, opmode=0x108  
> 9576768: Uplink data pending  
> 9576903: Airtime available at 9482879 (previously determined)  
> 9577252: Airtime available at 9576587 (global duty limit)  
> 9577578: Ready for uplink  
> 9578410: Updating info for TX at 9576767, airtime will be 2896. Setting available time for band 1 to 2357133312  
> 9578769: TXMODE, freq=868300000, len=14, SF=7, BW=125, CR=4/5, IH=0  
> 9581294: irq: dio: 0x0 flags: 0x8  
> 9581386: Scheduled job 0x449, cb 0xef3 ASAP  
> 9582107: Running job 0x449, cb 0xef3, deadline 0  
> 9582223: Scheduled job 0x449, cb 0xc45 at 9643155  
> 9643485: Running job 0x449, cb 0xc45, deadline 9643155  
> 9643669: RXMODE\_SINGLE, freq=868300000, SF=7, BW=125, CR=4/5, IH=0  
> 9646830: irq: dio: 0x0 flags: 0x40  
> 9646951: Scheduled job 0x449, cb 0x20cd ASAP  
> 9647675: Running job 0x449, cb 0x20cd, deadline 0  
> 9648280: Received downlink, window=RX1, port=-1, ack=1  
> 9648387: 10
> 
> 9648469: engineUpdate, opmode=0x800
> 
> 9650217: engineUpdate, opmode=0x808  
> 9650293: Uplink data pending  
> 9650356: Considering band 0, which is available at 10930440  
> 9650593: Considering band 1, which is available at 9866367  
> 9650925: Considering band 2, which is available at 4950416  
> 9651257: Considering band 3, which is available at 0  
> 9651567: No channel found in band 3  
> 9651757: Considering band 0, which is available at 10930440  
> 9652095: Considering band 1, which is available at 9866367  
> 9652427: Considering band 2, which is available at 4950416  
> 9652771: No channel found in band 2  
> 9652961: Considering band 0, which is available at 10930440  
> 9653298: Considering band 1, which is available at 9866367  
> 9653632: Airtime available at 9866367 (channel duty limit)  
> 9653962: Uplink delayed until 9866367  
> 9654180: Scheduled job 0x449, cb 0x2007 at 9866242
> 
> 9866336: Running job 0x449, cb 0x2007, deadline 9866242  
> 9866454: engineUpdate, opmode=0x8  
> 9866527: Uplink data pending  
> 9866655: Airtime available at 9866367 (previously determined)  
> 9867003: Ready for uplink  
> 9867835: Updating info for TX at 9866526, airtime will be 3216. Setting available time for band 1 to 1969094656  
> 9868194: TXMODE, freq=868500000, len=16, SF=7, BW=125, CR=4/5, IH=0  
> 9871348: irq: dio: 0x0 flags: 0x8  
> 9871440: Scheduled job 0x449, cb 0xef3 ASAP  
> 9872161: Running job 0x449, cb 0xef3, deadline 0  
> 9872277: Scheduled job 0x449, cb 0xc45 at 9933209  
> 9933544: Running job 0x449, cb 0xc45, deadline 9933209  
> 9933727: RXMODE\_SINGLE, freq=868500000, SF=7, BW=125, CR=4/5, IH=0  
> 9935632: irq: dio: 0x1 flags: 0x80  
> 9935725: Scheduled job 0x449, cb 0x20cd ASAP  
> 9936448: Running job 0x449, cb 0x20cd, deadline 0  
> 9936565: Scheduled job 0x449, cb 0xc71 at 9995453  
> 9995936: Running job 0x449, cb 0xc71, deadline 9995453  
> 9996120: RXMODE\_SINGLE, freq=869525000, SF=9, BW=125, CR=4/5, IH=0  
> 10000040: irq: dio: 0x1 flags: 0x80  
> 10000136: Scheduled job 0x449, cb 0x20e4 ASAP  
> 10000861: Running job 0x449, cb 0x20e4, deadline 0  
> 10000969: 10
> 
> 10001058: engineUpdate, opmode=0x900
> 
> 10002793: engineUpdate, opmode=0x908  
> 10002873: Uplink data pending  
> 10002938: Considering band 0, which is available at 10930440  
> 10003186: Considering band 1, which is available at 10188126  
> 10003529: Considering band 2, which is available at 4950416  
> 10003867: Considering band 3, which is available at 0  
> 10004182: No channel found in band 3  
> 10004379: Considering band 0, which is available at 10930440  
> 10004722: Considering band 1, which is available at 10188126  
> 10005065: Considering band 2, which is available at 4950416  
> 10005415: No channel found in band 2  
> 10005610: Considering band 0, which is available at 10930440  
> 10005953: Considering band 1, which is available at 10188126  
> 10006308: Airtime available at 10188126 (channel duty limit)  
> 10006639: Airtime available at 10264935 (global duty limit)  
> 10006977: Uplink delayed until 10264935  
> 10007207: Scheduled job 0x449, cb 0x2007 at 10264810
> 
> 10265398: Running job 0x449, cb 0x2007, deadline 10264810  
> 10265523: engineUpdate, opmode=0x108  
> 10265602: Uplink data pending  
> 10265752: Airtime available at 10188126 (previously determined)  
> 10266113: Airtime available at 10264935 (global duty limit)  
> 10266450: Ready for uplink  
> 10267288: Updating info for TX at 10265601, airtime will be 2896. Setting available time for band 1 to 255918080  
> 10267653: TXMODE, freq=868100000, len=14, SF=7, BW=125, CR=4/5, IH=0  
> 10270184: irq: dio: 0x0 flags: 0x8  
> 10270278: Scheduled job 0x449, cb 0xef3 ASAP  
> 10271003: Running job 0x449, cb 0xef3, deadline 0  
> 10271121: Scheduled job 0x449, cb 0xc45 at 10332045  
> 10332390: Running job 0x449, cb 0xc45, deadline 10332045  
> 10332579: RXMODE\_SINGLE, freq=868100000, SF=7, BW=125, CR=4/5, IH=0  
> 10335118: irq: dio: 0x0 flags: 0x40  
> 10335239: Scheduled job 0x449, cb 0x20cd ASAP  
> 10335965: Running job 0x449, cb 0x20cd, deadline 0  
> 10336562: Received downlink, window=RX1, port=-1, ack=1  
> 10336673: 10
> 
> 10336761: engineUpdate, opmode=0x800

It seems that the message is received in RX1 window but not interpreted or used.

---

<div class="post-metadata">

**Author:** ![dub](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/dub/32/14129_2.png) [@dub](https://www.thethingsnetwork.org/forum/u/dub)\
**Post date:** [June 13, 2018, 10:06am UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/3 "2018-06-13T10:06:21Z")

</div>

@biermi

Looks like TTN gave up on your node and refuses to send any more ADR requests. However, from your log it appears to be that your node is already running failry efficiently (SF7 TX and RX),

Here is what you can do to debug the thing:

1. Set ACKREQ on each uplink (this will cause adr optimize on a first uplink after join). You can keep ACKREQ bit set on every uplink.
2. Clear log on TTN console and restart the device.
3. You should see that first ACK following your first uplink will contain ADR data if that was needed.
4. If 3 is not the case you’ll have to wait for 20+ uplinks.

Repeat 1 to 3 and post snapshot of TTN console log for first ACK,

The above will reveal if TTN ever sends ADR data or not.

If it sends it then we need to look at LMIC mac processing.

---

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 13, 2018, 10:47am UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/4 "2018-06-13T10:47:38Z")

</div>

How can i set the ACKREQ on each uplink with LMIC? There is no operation for that, i can only activate:

> LMIC\_setLinkCheckMode(1);  
> LMIC\_setAdrMode(1);

as operation for the lib, that i have done. I can later post the join and first uplink messages from the TTN console of my gateway.

---

<div class="post-metadata">

**Author:** ![dub](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/dub/32/14129_2.png) [@dub](https://www.thethingsnetwork.org/forum/u/dub)\
**Post date:** [June 13, 2018, 12:29pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/5 "2018-06-13T12:29:36Z")

</div>

LMIC\_setLinCheckMode(1) is the right way to do so.

As per processing of the state after join it’ll set LMIC.adrAckReq to some value.  
Here is code snippet from lmic.c

```
static void stateJustJoined (void) {
    LMIC.seqnoDn = LMIC.seqnoUp = 0;
    LMIC.rejoinCnt = 0;
    LMIC.dnConf = LMIC.adrChanged = LMIC.ladrAns = LMIC.devsAns = 0;
    LMIC.moreData = LMIC.dn2Ans = LMIC.snchAns = LMIC.dutyCapAns = 0;
    LMIC.pingSetAns = 0;
    LMIC.upRepeat = 0;
    LMIC.adrAckReq = LINK_CHECK_INIT;
    LMIC.dn2Dr = DR_DNW2;
    LMIC.dn2Freq = FREQ_DNW2;
    LMIC.bcnChnl = CHNL_BCN;
    LMIC.ping.freq = FREQ_PING;
    LMIC.ping.dr = DR_PING;
}
```

---

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 13, 2018, 12:42pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/6 "2018-06-13T12:42:32Z")

</div>

Ok, post the screenshots of the first results later here.

---

<div class="post-metadata">

**Author:** ![dub](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/dub/32/14129_2.png) [@dub](https://www.thethingsnetwork.org/forum/u/dub)\
**Post date:** [June 13, 2018, 12:59pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/7 "2018-06-13T12:59:32Z")

</div>

When I said that ADR should work on LMIC I actually know that there is a code there. I haven’t tried it but based on what I am seeing in “decodeFrame” function of lmic.c processing of ADR response is there:

```
            if( (LMIC.ladrAns & 0x7F) == (MCMD_LADR_ANS_POWACK | MCMD_LADR_ANS_CHACK | MCMD_LADR_ANS_DRACK) ) {
                // Nothing went wrong - use settings
                LMIC.upRepeat = uprpt;
                setDrTxpow(DRCHG_NWKCMD, dr, pow2dBm(p1));
            }

```

Add some debug print and you should be able to resolve it no time.  
First thing that you need to find out is if ADR portion of the lora mac message is there.

---

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 13, 2018, 3:09pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/8 "2018-06-13T15:09:06Z")

</div>

It seems it is working now, without any code changes on the node. The only change, so far i know, is that the backend software is changed as here described [ADR Not Working on AU915 (Device stuck on SF12)](https://www.thethingsnetwork.org/forum/t/adr-not-working-on-au915-device-stuck-on-sf12/14891). Maybe it was the same problem or similar which is now fixed by @htdvisser.  
 ![image](https://www.thethingsnetwork.org/forum/uploads/default/original/2X/a/a279296f00ffa0f2c4675b3514ea9ef5df6f379c.png)  
On the screen shot is shown that the ADR requests starts after 12 uplink packages and ends after 3.

---

<div class="post-metadata">

**Author:** ![dub](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/dub/32/14129_2.png) [@dub](https://www.thethingsnetwork.org/forum/u/dub)\
**Post date:** [June 13, 2018, 11:27pm UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/9 "2018-06-13T23:27:18Z")

</div>

@biermi

How do you know that any of the messages sent to device between 16:58:45 and 17:00:38 contain ADR data ?

Have you actually seen ADR sent from message arriving to device and being processed there in decodeFrame() function ?

The reason I am asking is the fact that device was on SF7 and it stayed there so device was reasonably optimized already. However that brings another question into focus and this is at what SF it joined the network and how it did get to SF7 for start.

---

<div class="post-metadata">

**Author:** ![biermi](https://www.thethingsnetwork.org/forum/user_avatar/www.thethingsnetwork.org/biermi/32/38364_2.png) [@biermi](https://www.thethingsnetwork.org/forum/u/biermi)\
**Post date:** [June 14, 2018, 6:35am UTC](https://www.thethingsnetwork.org/forum/t/error-too-many-failed-adr-requests/14962/10 "2018-06-14T06:35:11Z")

</div>

I have looked in the console at the packages which are send to the node, they showing, similar to the first screenshot in this thread, mac commands, which will be send to the node. In the serial monitor of the node was shown that packages are received but no additional logging within the decodeFrame() function. And there are only 3 packages send to the node, in the tests before every uplink package followed by a downlink package starting after 12 uplink packages received by the gateway.

The node was joining with SF7 at the beginning, the console of the gateway shows the joining request was received with SF7. I have not set any SF in the code, it must be came from the LMiC lib himself. My assumption was similar to yours, i thought it will start with SF12 for joining and lowers than the SF to the optimal value. But it starts with SF7. I’m using the LMiC from @matthijskooijman [https://github.com/matthijskooijman/arduino-lmic](https://github.com/matthijskooijman/arduino-lmic).