Although I do not have exactly the same error, I think the actual issue is the same:
19-03-16 10:54:20 INFO (MainThread) [hassio.core] Hass.io is up and running 19-03-16 10:57:46 INFO (SyncWorker_7) [hassio.docker.interface] Pull image hassioaddons/node-red-armhf tag 1.4.0. 19-03-16 10:57:48 ERROR (SyncWorker_7) [hassio.docker.interface] Can't install hassioaddons/node-red-armhf:1.4.0 -> 500 Server Error: Internal Server Error ("readlink /mnt/data/docker/overlay2/l: invalid argument"). 19-03-16 10:59:20 INFO (MainThread) [hassio.homeassistant] Updated Home Assistant API token 19-03-16 11:00:29 INFO (SyncWorker_1) [hassio.docker.interface] Pull image hassioaddons/node-red-armhf tag 1.4.0. 19-03-16 11:00:30 ERROR (SyncWorker_1) [hassio.docker.interface] Can't install hassioaddons/node-red-armhf:1.4.0 -> 500 Server Error: Internal Server Error ("readlink /mnt/data/docker/overlay2/l: invalid argument").
If you see “image not found” errors, those are usually a matter of not having the available space. You can check quickly with df -h in a console. Often users are adding space to a VM and not realizing that they also need to configure HassOS to use that additional space. This is better handled when installing Hassio the first time as it expands into the available space at setup.
These 500 error/internal server errors which mention the overlay fs are not the same and are unfortunately best addressed by restoring a snapshot to a newly flashed install as far as I can tell.
EDIT:
500 Server Error: Internal Server Error (“readlink /mnt/data/docker/overlay2/l: invalid argument”)
Since I wrote this a new command has been added which can repair the above situation. Running ha su repair in the custom cli console should resolve it.
19-03-17 14:37:29 INFO (MainThread) [hassio.homeassistant] Update Home Assistant to version 0.90.0b3
19-03-17 14:37:29 INFO (SyncWorker_1) [hassio.docker.interface] Update Docker homeassistant/raspberrypi3-homeassistant:0.89.0b1 to homeassistant/raspberrypi3-homeassistant:0.90.0b3
19-03-17 14:37:29 INFO (SyncWorker_1) [hassio.docker.interface] Pull image homeassistant/raspberrypi3-homeassistant tag 0.90.0b3.
19-03-17 14:37:30 ERROR (SyncWorker_1) [hassio.docker.interface] Can't install homeassistant/raspberrypi3-homeassistant:0.90.0b3 -> 500 Server Error: Internal Server Error ("readlink /var/lib/docker/overlay2/l: invalid argument").
19-03-17 14:37:31 WARNING (MainThread) [hassio.homeassistant] Home Assistant is already running!
19-03-17 14:37:31 INFO (MainThread) [hassio.homeassistant] Successful run Home Assistant 0.90.0b3
Same error here trying to update " SSH & Web Terminal" addon
19-03-20 21:21:47 ERROR (SyncWorker_14) [hassio.docker.interface] Can’t install hassioaddons/ssh-armv7:4.0.2 -> 500 Server Error: Internal Server Error (“readlink /mnt/data/docker/overlay2/l: invalid argument”).
Same here - cannot update to latest version or any add-ons
Anyone have any idea how to fix it or expand the disk space?
I’m also experiencing this issue.
In addition, when I try to update any HA, hassos or addons through the Hass.io UI panel it fails, usually with a 500 Server Error: Internal Server Error, but somtimes it wont report any errors at all. And then after that, I’m unable to access the Hass.io panel at all. It gives me this error when trying to access it Unable to load the panel source: /api/hassio/app/entrypoint.js.
I’ve tried logging into my RPi using ssh and update that way. But that’s unsuccessul as well.
core-ssh:~# hassio homeassistant restart
Post http://hassio/homeassistant/restart: dial tcp: lookup hassio on [::1]:53: dial udp [::1]:53: connect: cannot assign requested address
I thought this might be to do with a bad SD card or corruption. However, I did created a fresh SD install on a different RPi using Hass.io (32 bit). This booted fine, but after restoring my latest snapshot backup, all of the above errors and inability to update HA, hassos and addons have returned with the restore of my last snapshot.
All this seems to have been happening for about the last 5-7 days.
Regarding the 500 error/internal server errors, I don’t have any snapshots available. If I take a snapshot now that I’ve uninstalled node-red and now trying to re-install it (but with this error), will the snapshot be valid?
df -h shows that there’s 5.8G 56% free space. So seems unlikely that the disk is full?
Although it’s far from ideal, I can confirm restoring a recent snapshot fixed the 500 error for me.
Did you do a restore or a wipe & restore? Doing the latter seems to cause the same errors to occur on a fresh install.
I did a regular restore.
It really seems like this is a problem introduced in 0.88.2. I, like others with the same issue, am running Hassio and any attempt to update or install an add-on ultimately results in the readline error. For example when attempting to install configurator I get this message but it’s the same one I get attempting to upgrade:
Can't install homeassistant/armv7-addon-configurator:2.1 -> 500 Server Error: Internal Server Error ("readlink /mnt/data/docker/overlay2/l: invalid argument").
I’ve been getting a version of this error on every attempted update since 0.88.2 and am now stuck since I don’t have a recent backup. Everything seems to work okay, just can’t install or upgrade anything. I’m sure there are lots of novice users getting the same error but don’t know how/where to look in the logs to find out why the update isn’t working.
I don’t really think “wipe and restore” is a reasonable option for this so is there a way to report this as a bug against Hassio? We’re scaring away novice users by ignoring problems that occur in the easiest way of running HA (Hassio). Most of us using Hassio are doing so because we don’t know how or can’t invest the time in learning the underlying OS issues with running a HA system. We just want it to work and apparently many of us can’t use the over-the-air system update.
I don’t think so. There are similar issues out there going back to 0.6x or earlier.
Fair enough, it just seemed like a lot of folks stopped being able to update around then. Either way, I’m one of at least several people (and I’m guessing a fair number more who don’t know how to troubleshoot) with a basically vanilla installation on Hassio that can’t update any further and can’t fix the issue without wiping my system and starting from scratch.
Having same issue after update to 0.90.2 ha did not come back up.
I’m on RPI-B.
I still can console.
I can’t even downgrade
hassio home assistant update --version “0.90.1”
getting Error: Unknown Error, see logs
What do you see in the logs when you run the following after attempting that update?
hassio supervisor logs
I’m curious if it’s also that readline error.
I tried and I was able to downgrade to 0.90.0 with
hassio home assistant update --version “0.90.0”
HA came up and I was able to access GUI again.
Then upgraded to 0.90.2 again. No issues now.
Unfortunately I did not check the supervisor log last night.
Its all gone now.
Guys,
Not sure if this is still helpful, but I was having the same issue as you and using the Pi via WiFi.
My error was:
19-04-04 18:43:23 ERROR (SyncWorker_3) [hassio.docker.interface] Can’t install hassioaddons/grafana-armv7:2.0.2 -> 500 Server Error: Internal Server Error (“Get https://registry-1.docker.io/v2/: dial tcp: lookup registry-1.docker.io on [ipv6]:53: read udp [something (ipv6?]:43589->[something??]:53: i/o timeout”).
Just connected a cable and the update worked. Just testing now the addons as well.
Don’t think that this is a reasonable solution but from the error I would say that the Docker is expecting to receive the files by Cable and not by WiFi. I would say that this is because it may interrupt the connection with possible Sensors?
Thanks everyone.
My pi is connected via ethernet so unfortunately that’s not my issue. Someone commented on the Github related issue that it seems like a bad Docker DB but no guidance on how to fix it.
I’m running into the same issue here. Restoring also won’t work. Anyone have a workaround?