Hass.io Access Crashing Overnight, Every Night

Hass.io is crashing within 12 hours of running. Access to the Front End, Samba, and SSH all go down. With SSH reporting:“ssh: Could not resolve hostname hassio.local: nodename nor servname provided, or not known”. Removing and reconnecting the power cord brings everything back online. The RPi shows power and network lights running despite Hassio being inaccessible.

It does not appear to be receiving my MQTT data from temperature/humidity sensors while it is down.

I am running HASS.IO on a new, quality SD card and with a quality power cord on a RPi3 with DuckDNS and Let’s Encrypt (via DuckDNS Add-on).

I have tried the following with no luck:

Config Check runs without issue.
Disable ffmepg cameras (heard this has been an issue in the past).
Deleted the database file.
Successfully updated to Version 58.1.

What are you using to power it (brand and model ideally)?

Vilros Adapter (Switching Adaptor)
Model: FJ-SW7260502500DU
50/60Hz 0.4A Max
Output: 5V —2500mA

Having the same issue (I’m the op of the reddit post :))

It seems to me, from my observations and your description that it’s a network connectivity issue.

All my timed and motion sensor automations are still working, as do the temp/humidity wired ones (dht11)

I don’t have mqtt enabled but other than that, I am experiencing the same issues as OP.

Hi

I have had the same thing, hanging every 12 or so hours, I have just used the Snapshot version to go back to before I installed MQTT

So I am back on V57.2, will give that a few days and update to latest version, I will let that run a few days and if ok will install MQTT

Regards, Dave

After being stumped trying to figure out what the cause was of the same symptoms that you are experiencing, I decided to test my Pi’s network connectivity on its own without hassio. I saved the image of my hassio SD card on my PC and burnt a fresh Raspbian Stretch Lite.

After adding my wifi credentials, I popped in the card and booted up the Pi. The next step I took was to SSH into it and ran a ping test. My first test was to ping google.com 60 times at 30-second intervals. The first 40 or so pinged without any issues. Out of the next 20 packets, the Pi dropped 15!

My conclusion—faulty wifi hardware. I got a replacement board, burned that backup hassio image I made, and I’ve been running without a single problem ever since!

Note: I see you have a Vilros adapter. My Raspberry Pi 3 was a Vilros kit from Amazon. Yours, too?

Yes, same kit. I just started a fresh install of Hass.io. If the problem re-appears, I suppose I will have to get a new Pi. Thanks for the tip!

Hi Jut an update, since I restored back to V57.2 it been running fine all week, so I dont feel that it is a hardware problem.

Re Dave

I did a fresh install of the latest version of Hass.io on Tuesday night and have slowly added components throughout the week. So far it has been running fine, but I have not added the TP-Link Router Device Tracker component. I believe this may have been a contributing factor.

Front End decided to hang on me this morning immediately after I left the house after almost a week of no issues. Based on the Logbook and History, it must have something to do with Bluetooth Add-On. Has that add-on been an issue in the past?

Is it just the frontend that froze, or did Samba and SSH go down as well? Also, can you reproduce the hang by leaving your house again (or turning your phone’s Bluetooth off)?

I have experienced the same symptoms since running HASSIO on a RPI3. Hang ups of front end after a day, a couple of days at most.
Most times a hard reboot was needed to gain access again. Once the front end came back alive after waiting a night, cannot explain how or why.

Things I tried, without success, where:

  • fresh install off HASSIO image and rebuild from scratch
  • tried another fresh RPI3
  • went from SD Card to booting from an USB stick

Most times reboots of the Home Assistant container resulted in components not working properly. Often, some add-ons were stopped after a reboot of the RPI3, even when set to auto-start in the config. Just plain old strange behavior.

Finally I made an image of the USB stick and wrote that image to an older SSD harddisk which I then connected to the USB using a SATA to USB convertor. I have yet to experience a hang up or reboot issues…

So it seems that the quality of the storage medium you use, is very important for a proper working HASSIO installation. Before, I did not experience this using HASSBIAN (running from the same SD Card).

Tested the Bluetooth theory this week and there were a couple crashes during entry/exit of my home zone, but not consistent enough to assign it to that component. Starting to give up on Hass.io.

A couple of crashes seem to have occurred when motion would have been detected by a Z-Wave sensor after it was “sleeping” for nearly 8 hours. While I know this component and device are not the root cause (issue existed prior to install) it made hold some clues to the problem.

Do you have another Raspberry Pi you can install your hassio build/config on? I know well the frustration of not finding the cause of many crashes after weeks of problems, and it feels so good now that it’s completely fixed for me (by replacing the Pi). If it’s not easy for you to work out, I understand, but I would like to implore you to at least consider ruling out network hardware failures of your specific Pi as the root cause of your issues.

I connected the pi through Ethernet, never had this issue since.
Not a fix, but a work around :slight_smile:

What steps do I need to take to switch over to the Ethernet connection? Will I have to edit the Resin Config file on the SD card?

Yep, I had to shut down, get the card to a pc and edit the network files.
I also enabled the WiFi on a different IP just in case.
Don’t remember by heart what I did exactly because i am terribly new to this and I was researching online.
I’ll see if I kept any backups though.

In case you are still interested, was fiddling with my sdcard today, so here is the network configuration i set up:

3 files inside the system-connections folder:

file 1 resin-eth:

[connection]
id=my-ethernet
type=ethernet
interface-name=eth0
permissions=
secondaries=
autoconnect-priority=1

[ethernet]
mac-address-blacklist=

[ipv4]
address1=x.x.x.x/24,x.x.x.x
dns=8.8.8.8;8.8.4.4;
dns-search=
method=manual

[ipv6]
addr-gen-mode=stable-privacy
dns-search=
method=auto

file 2 resin-wifi:

[connection]
id=resin-wifi
type=wifi
autoconnect-priority=10

[wifi]
hidden=true
mode=infrastructure
ssid=MAIN_WIFI_NETWORK

[wifi-security]
auth-alg=open
key-mgmt=wpa-psk
psk=WIFI_PASS

[ipv4]
address1=x.x.x.x/24,x.x.x.x
dns=8.8.8.8;8.8.4.4;
dns-search=
method=manual

[ipv6]
addr-gen-mode=stable-privacy
method=auto

Note: The address1 here is different than the ethernet one, as they connect simultaneously.

and file 3 resin-wifi2:

[connection]
id=resin-wifi2
type=wifi
autoconnect-priority=20

[wifi]
hidden=true
mode=infrastructure
ssid=BACKUP_WIFI_NETWORK

[wifi-security]
auth-alg=open
key-mgmt=wpa-psk
psk=WIFI_PASS

[ipv4]
address1=x.x.x.x/24,x.x.x.x
dns=8.8.8.8;8.8.4.4;
dns-search=
method=manual

[ipv6]
addr-gen-mode=stable-privacy
method=auto

With same address as the other wifi connection as they wont ever be connected at the same time.

The issue is completely resolved by switching to an ethernet connection and has been functioning correctly for nearly a month.

Had the same issue. Crash every night . Switched over to ethernet no problems for a week .

Simply mount sd card and clear down resin-sample or resin-wifi. Hass.io will default to ethernet them simply fix the DHCP ip address on Internet router.