I am on the latest-dev branch in docker container. https://github.com/Koenkk/zigbee2mqtt/commit/93e5636f0c51f5fd403ea1250db9456bfe7b5fcc.
On latest firmware from https://github.com/koenkk/z-stack-firmware.
I have pulled latest-dev from docker hub and restarted zigbee2mqtt with same config using docker-compose.
zigbee2mqtt fails to start.
zigbee2mqtt:info 7/30/2019, 2:54:27 PM Starting zigbee2mqtt version 1.5.1 (commit #unknown)
zigbee2mqtt:info 7/30/2019, 2:54:27 PM Starting zigbee-shepherd
zigbee2mqtt:info 7/30/2019, 2:54:29 PM Error while starting zigbee-shepherd, attempting to fix... (takes 60 seconds) (Error: Cluster with name 'manuSpecificCluster' does not exist)
zigbee2mqtt:info 7/30/2019, 2:55:29 PM Starting zigbee-shepherd
zigbee2mqtt:error 7/30/2019, 2:55:31 PM Error while starting zigbee-shepherd! (Error: dev exists, unregister it first.)
zigbee2mqtt:error 7/30/2019, 2:55:31 PM Press the reset button on the stick (the one closest to the USB) and start again
zigbee2mqtt:error 7/30/2019, 2:55:31 PM Failed to start
{"message":"dev exists, unregister it first.","stack":"Error: dev exists, unregister it first.\n at /app/node_modules/zigbee-herdsman/dist/lib/shepherd.js:449:19\n at Promise.apply (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:1185:26)\n at Promise.promise.promiseDispatch (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:808:41)\n at /app/node_modules/zigbee-herdsman/node_modules/q/q.js:1411:14\n at runSingle (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:137:13)\n at flush (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:125:13)\n at process._tickCallback (internal/process/next_tick.js:61:11)"}
zigbee2mqtt:error 7/30/2019, 2:55:31 PM Exiting...
I have logged into container and recursively grepped manuSpecificCluster. It only exists in the database.db file. For WXKG01LM - Xiaomi MiJia wireless switch.
Interesting that if I run latest everything starts fine.
Normal start
Just run latest dev docker container.
zigbee2mqtt version: 1.5.1
CC253X firmware version: 20190619
Exactly the same happens in my environment with Aqara sensors WSDCGQ11LM.
I cannot change to dev branch at the moment.
Thanks, fixed in latest dev
yes, working!
that comment was a little too quick:
zigbee2mqtt is working and stays working, but this message occurs in the log:
7/31/2019, 8:37:12 PM - error: Error: Cluster with ID '64636' does not exist
at getClusterByID (/app/node_modules/zigbee-herdsman/dist/zcl/index.js:30:15)
at Object.getClusterLegacy (/app/node_modules/zigbee-herdsman/dist/zcl/index.js:40:19)
at /app/node_modules/zigbee-herdsman/dist/lib/components/af.js:342:29
at Object._.forEach (/app/node_modules/zigbee-herdsman/node_modules/busyman/index.js:411:27)
at Object.af.zclClustersReq (/app/node_modules/zigbee-herdsman/dist/lib/components/af.js:341:7)
at /app/node_modules/zigbee-herdsman/dist/lib/components/event_handlers.js:252:32
at _fulfilled (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:854:54)
at /app/node_modules/zigbee-herdsman/node_modules/q/q.js:883:30
at Promise.promise.promiseDispatch (/app/node_modules/zigbee-herdsman/node_modules/q/q.js:816:13)
at /app/node_modules/zigbee-herdsman/node_modules/q/q.js:877:14
edit: it happens when powercycling my ikea repeaters (I allways do that when restarting my docker image)
Runs fine after last commit. Thank you.
@prein2 should also be fixed (wait until new docker image is up).
Most helpful comment
Thanks, fixed in latest dev