sorry
Dear Tom, I support all your comments on the required changes in the binding!
Although OOM has been fixed (thanks to both of you @J-N-K and @Trinitus01), this is neither the cause of the problem nor sufficient as fix. As indicated in multiple posts it is most probably due to some change in the protocol on the amazon server. Now only the side effects are fixed, not the original problem.
Also the webservice functionality of the binding e.g. lastVoiceCommand needs to be fixed. Just like the many more suggestions you made in previous posts
Further functionality way beyond the current error fix is highly appreciated, as your very constructive posts:
and
Please provide your changes to the general public by PR, you seem to have the most advanced updates done for the functionality of the binding. The community will appreciate your efforts!
2 Likes
I have had same memory issues this afternoon relating to the Amazon binding, had to remove the binding from the addons.cfg and then restart oh to stop it failing on a pi 3b+.
Stable now but not ideal.
Just installed it and will report tomorrow if it fixed the leak.
I am on openhab 2.5.3-1
the amazonechocontrol binding 2.5.3 is installed
I am located in germany but do not see the error message in my logs.
Does that mean a specific subset of the binding triggers the problem ? Which one ?
I use paper Ui for setup all. Rules are set only to send text to voice.
Thatās interesting. During my investigations I only had the account configured, nothing else. So maybe you are lucky because your connection is still running and did not reconnect since Friday.
Hi everyone,
I am in Ontario, Canada (EDT) and I am using alexa.amazon.ca and I have had the same issue to. The OOM issues got to the point were I had to reboot my OH 2 times a day until I added these options in my openhab config file:
EXTRA_JAVA_OPTS="-Xms400m -Xmx650m"
Added these last night, so far no OOM errors. But the websocket error happens every 60 seconds.
Here is my system spec:
Release = Linux Mint 19.3 Tricia
Kernel = Linux 5.3.0-53-generic
Platform = Dell Inc. OptiPlex 9020M/0Y5DDC, BIOS A02 08/11/2014
Uptime = 0 day(s). 8:16:17
CPU Usage = 3.02% avg over 4 cpu(s) (4 core(s) x 1 socket(s))
CPU Load = 1m: 0.12, 5m: 0.09, 15m: 0.09
Memory = Free: 5.85GB (76%), Used: 1.82GB (24%), Total: 7.68GB
Swap = Free: 1.99GB (100%), Used: 0.00GB (0%), Total: 1.99GB
Root = Free: 101.57GB (91%), Used: 9.31GB (9%), Total: 116.86GB
Updates = 9 apt updates available.
Sessions = 1 session(s)
Processes = 151 running processes of 32768 maximum processes
Welcome to __ _____ ____ _
____ ____ ___ ____ / / / / | / __ )(_)___ _____
/ __ \/ __ \/ _ \/ __ \/ /_/ / /| | / __ / / __ `/ __ \
/ /_/ / /_/ / __/ / / / __ / ___ |/ /_/ / / /_/ / / / /
\____/ .___/\___/_/ /_/_/ /_/_/ |_/_____/_/\__,_/_/ /_/
/_/
openHAB 2.5.5-1 (Release Build)
So if there was a change on the Amazon side it seems to be in many regions.
I rebooted several times since friday, at least 5 times today.
I have set the log level for the binding to ERROR several months ago, so I was not aware of any problems and was astonished reading about it, as I have no memory problems.
I am using the binding and my system is running with normal speed.
The only thing unusual I noticed was having OOM warnings while installing a binding (but it was installed) and losing my connection to openHAB cloud two times without successful reconnecting, only after reboot.
At the moment everything is up and running with a stable memory usage of approx. 600 to 650 MB.
Last reboot 14 hours ago.
The lastvoicecommand problem as been resolved like 3-4 months ago on iobroker .
Maybe someone can have a check . It was related to timing with the ws connection. Canāt say more than that.
1 Like
Happening here as well 
Driving me crazy! None of the āLastVoiceCommandā or TTS features are working. And itās causing a memory leak on my system which causes openHAB to lock-up and shut down. I completely purged openHAB from my machine and reinstalled and then loaded a backup .zip file, reconfigured everything, and it still is not working at all. Must be an error on the Amazon server side. Is this something I can call Amazon customer support to ask about? I canāt even imagine trying to explain what openHAB is, let alone the error weāre having, to their technical support peopleā¦
Can you help me figure out which config file to add that code to? Iām having trouble locating the main openhab.config file in the current release of openhab2 :-/
I made a few changes to my system recently and thought that it must be one of these giving me OOMs so thanks for everyone sharing - Ive installed the new JAR and yes still getting the errors messages in the log as expected but is anyone noticing other things still not right?
My Zwave network seems to have become unresponsive, timers as well - i really rely on this binding so reluctant to remove it if I dont have to but is the smaller part of the system to the rest
Run through the same fix here as getting OOM issues and web socket errors. Hoping a permanent fix comes out soon itās oddly frustrating when you rely entirely on your overly convoluted and highly complex home automation system to run your daily life lol!!!
If you reboot so often (or restart openHAB), you are probably not seeing the OOM because it takes some time to leak enough memory to make it appear.
Same problem here. Please help. WAF is going down rapidly.
1 Like
Hi, my openHAB is working for approx. 24 hours now. My Alexa routines and TTS were running without problems this morning.
Memory usage 534 MB, workload approx. 7%.
RPi 3B
OH 2.5.5
Up for 16h since reboot after installing new .jar.
No sign of memory leak.
EchoControl TTS working
Web socket errors suppressed by setting log level to WARN
Thanks @J-N-K
Hiā¦ did you change anything in your configuration?
What system do you use?
My system (pi4 4gb) only runs stable without the binding, when i install the binding again, the memory increases steadily until a total crash of OH2, it takes 5 to 6 hours.
BR
Tom