Hi,
That error LogBackend: unexpected error: Error: read ECONNRESET may be related to an issue with the download of a new release not finishing because of poor internet connection. This issue was fixed with the balenaOS version 2.114.22.
Could you please confirm whether the network of the device you’re referring to is potentially poor?
Could you please tell us what balenaOS version your device is running? If it’s a version below 2.114.22, could you please try updating the OS version on the device? We would advise testing the new chosen OS version on a test device first to ensure that there are no issues with your application and the OS release.
@Jesper I just wanted to check if upgrading helped?
I saw that “Estelle” replied to you, but considering the response is really generic, and made immediately after creating their account, I thought it might be from a bot. We’ve been seeing more of those lately in Forums, so I wanted to check in to see if you were able to upgrade, first of all, and if you were the one who marked this issue as “Solved”, just in case.
I upgraded the OS from 2.113.18 to 3.0.5 and that did not make any difference on that error.
The errors comes every five minutes or so.
This seems to bee the complete log that repeats every 5 minutes:
Feb 01 23:14:21 88a4885 bc0ddca747f1[1908]: [api] GET /v1/healthy 200 - 8.159 ms
Feb 01 23:14:21 88a4885 balena-supervisor[3062]: [api] GET /v1/healthy 200 - 8.159 ms
Feb 01 23:14:25 88a4885 bc0ddca747f1[1908]: [error] LogBackend: unexpected error: Error: read ECONNRESET
Feb 01 23:14:25 88a4885 balena-supervisor[3062]: [error] LogBackend: unexpected error: Error: read ECONNRESET
Feb 01 23:14:25 88a4885 bc0ddca747f1[1908]: [error] at TLSWrap.onStreamRead (node:internal/stream_base_commons:217:20)
Feb 01 23:14:25 88a4885 balena-supervisor[3062]: [error] at TLSWrap.onStreamRead (node:internal/stream_base_commons:217:20)
Feb 01 23:15:57 88a4885 bc0ddca747f1[1908]: [info] Reported current state to the cloud
Feb 01 23:15:57 88a4885 balena-supervisor[3062]: [info] Reported current state to the cloud
Feb 01 23:15:58 88a4885 healthdog[181229]: try: 1, refid: B9B5DFA9, correction: 0.000140497, skew: 0.005
Feb 01 23:16:44 88a4885 NetworkManager[1824]: <info> [1706829404.5894] device (wlan0): set-hw-addr: set MAC address to 4A:D3:21:7B:34:C4 (scanning)
Feb 01 23:16:44 88a4885 kernel: brcmfmac: brcmf_cfg80211_set_power_mgmt: power save disabled
Feb 01 23:16:44 88a4885 NetworkManager[1824]: <info> [1706829404.6054] device (wlan0): supplicant interface state: inactive -> disconnected
Feb 01 23:16:44 88a4885 NetworkManager[1824]: <info> [1706829404.6056] device (p2p-dev-wlan0): supplicant management interface state: inactive -> disconnected
Feb 01 23:16:44 88a4885 NetworkManager[1824]: <info> [1706829404.6059] device (wlan0): supplicant interface state: disconnected -> inactive
Feb 01 23:16:44 88a4885 NetworkManager[1824]: <info> [1706829404.6060] device (p2p-dev-wlan0): supplicant management interface state: disconnected -> inactive
Feb 01 23:17:58 88a4885 healthdog[181249]: try: 1, refid: B9B5DFA9, correction: 0.000140160, skew: 0.005
Hello @Jesper could you please confirm the supervisor version you are running?
On the other hand, could you please confirm if you are using cellular connectivity on your device?
The cause of these messages is that we’ve enabled a proxy layer to reduce latency between devices in the field and the API. This proxy layer prevents long-running stream requests where no data is passed for more than 20seconds, which older Supervisors are not built to handle.
Hi @mpous
The supervisor version is 15.2.0
They are using ethernet, for network, and the building are connected to the internet using fiber connection.
Hi @mpous, I’m investigating some rpi4s going offline and only coming back only after unplugging/plugging the power cable on. I wanted to rule this option out among other things.
