https://devzone.nordicsemi.com/f/nordic-q-a/100658/ot_radio_workq-crash-at-application-start-after-updating-to-ncs-2-4-0-nrf5340-ssedHi, I recently updated my Matter application to NCS 2.4, and have been seeing crashes in ot_radio_workqueue twice upon application start. It doesn&#39;t happen frequently, but when the device gets in this state the only way to get it fixed is erasingen-USTelligent Community 13Mon, 26 Jun 2023 13:44:38 GMThttps://devzone.nordicsemi.com/thread/433000?ContentTypeID=1Mon, 26 Jun 2023 13:44:38 GMT137ad170-7792-4731-bb38-c0d22fbe4515:56a9edd5-033a-4db7-911d-bdc258fe8f92abollaert<p>Resolved the issue in the mean time, I had a race condition in my own code that was the cause of this.</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/430773?ContentTypeID=1Tue, 13 Jun 2023 12:56:01 GMT137ad170-7792-4731-bb38-c0d22fbe4515:49cbce73-3fe8-45a3-9023-732ea5fd5011abollaert<p>OK, I was already building the app core image with reset on fatal error set to n, going to add it to the rpmsg image as well and wait the issue to crop up again so I can take a backtrace.</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/430736?ContentTypeID=1Tue, 13 Jun 2023 11:58:41 GMT137ad170-7792-4731-bb38-c0d22fbe4515:5afdae4f-4e35-4ef0-82c5-289fa6c022c6AHaug<p>Hi,</p> <p>Disregard the suggestion about the stack size for now, as the input from the Matter team in my previous comment is more relevant for solving the issue.</p> <p>Kind regards,<br />Andreas</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/430730?ContentTypeID=1Tue, 13 Jun 2023 11:48:34 GMT137ad170-7792-4731-bb38-c0d22fbe4515:b02f636d-bdfa-47e8-a97e-cb0c17dd5175abollaert<p>I did notice while digging through the samples that I didn&#39;t have this config option:</p> <p><pre class="ui-code" data-mode="text">CONFIG_IEEE802154_NRF5_RX_STACK_SIZE=1024</pre></p> <p>Is assume this is the stack you mean?</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/430666?ContentTypeID=1Tue, 13 Jun 2023 09:01:19 GMT137ad170-7792-4731-bb38-c0d22fbe4515:77da08ee-8a7b-44f0-97e9-6c7d7cf215d5AHaug<p>Hi again,</p> <p>Could you build with -Dmultiprotocol_rpmsg_CONFIG_RESET_ON_FATAL_ERROR=n -DCONFIG_RESET_ON_FATAL_ERROR=n and report if this results in the same behaviour?</p> <p>If it crashes, could you then collect gdb startrace from the net core and post it here? Please ensure that multiprotocol_rpmsg image is built with&nbsp;CONFIG_RESET_ON_FATAL_ERROR=n`</p> <p>Kind regards,<br />Andreas</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/430522?ContentTypeID=1Mon, 12 Jun 2023 13:45:11 GMT137ad170-7792-4731-bb38-c0d22fbe4515:6942ff84-0b98-4c44-829b-c63dd0540c6bAHaug<p>Hi,</p> <p>I&#39;ll bring this up to the Matter team for some pointers and I&#39;ll get back to you for further investigation of the issue you&#39;re describing.</p> <p>PS: One thought is that the issue is caused by the openthread radio workqueue stack beeing too small. Could you try increasing that stacksize while waiting for some input?</p> <p>Kind regards,<br />Andreas</p><div style="clear:both;"></div>