unRAID OS version 6.5.0-rc1 available

[Can0n 33]

I do have to say my smaller unraid server with an older i5 is running a lot faster on 6.5.rc1 love it so far!

 

Thanks Limetech love the O/S

[digitalfixer 7]

Reboot did not fix the no bar problem or the nerd pack display.

 

Rolled back to 6.4.1 and all is ok again

 

Kevin.

[pwm 221]

8 hours ago, dlandon said:

Doing that would save a lot of support requests.

Hiding the initial warning about disk UDMA ECC errors would also save a lot of support requests.

[bonienl 1100]

32 minutes ago, pwm said:

Hiding the initial warning about disk UDMA ECC errors would also save a lot of support requests.

 

True, but it requires a big rewrite of the underlying SMART detection and reporting system, while removing the superfluous metric value in the network config file is just a one-liner.

[dlandon 576]

I just updated my main server and now I can't specify br0 on dockers to set the IP address.  How do I do that now?

[pwm 221]

9 minutes ago, bonienl said:

 

True, but it requires a big rewrite of the underlying SMART detection and reporting system, while removing the superfluous metric value in the network config file is just a one-liner.

Does it really?

 

The current code obviously has to make a comparison whenever it sees smart data to see if the ECC count has changed.

 

Wouldn't it be enough to have lack of state in the state data (or possibly a -1 value) mean that you can directly auto-freeze (auto-acknowledge) the first SMART ECC value you get access to. So you don't do any big rewrites but handles it as a "lazy acknowledge" when you notice you don't have any valid watermark value for that specific disk.

[bonienl 1100]

4 minutes ago, dlandon said:

I just updated my main server and now I can't specify br0 on dockers to set the IP address.  How do I do that now?

 

Can you post a screenshot of the Docker settings page (with Docker stopped and advanced on)

[dlandon 576]

3 minutes ago, bonienl said:

 

Can you post a screenshot of the Docker settings page (with Docker stopped and advanced on)

Here you go:

br0 never shows up in the docker network choices.

[bonienl 1100]

3 minutes ago, pwm said:

Does it really?

 

Yes, it does.

 

The current implementation reports a SMART issue for any of the monitored SMART values the first time system is started. That is desired because users need to be warned the first time when there are potential issues. "UDMA CRC errors" follow the same logic, but you are asking for an exception here.

[dlandon 576]

1 minute ago, bonienl said:

 

Yes, it does.

 

The current implementation reports a SMART issue for any of the monitored SMART values the first time system is started. That is desired because users need to be warned the first time when there are potential issues. "UDMA CRC errors" follow the same logic, but you are asking for an exception here.

It only occurs once when updating and is not that big a deal.  I don't think creating an exception is a good idea.

[pwm 221]

1 minute ago, bonienl said:

 

Yes, it does.

 

The current implementation reports a SMART issue for any of the monitored SMART values the first time system is started. That is desired because users need to be warned the first time when there are potential issues. "UDMA CRC errors" follow the same logic, but you are asking for an exception here.

But after the user acknowledges it, the system remembers the watermark value. Isn't that watermark value saved so it survives the next reboot?

 

[bonienl 1100]

6 minutes ago, dlandon said:

br0 never shows up in the docker network choices.

 

Looks like it failed to create 'br0' under Docker. A quick workaround is:

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

For my understanding: did you make any setting modifications or it failed right away after upgrading?

 

[dlandon 576]

2 minutes ago, bonienl said:

 

Looks like it failed to create 'br0' under Docker. A quick workaround is:

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

For my understanding: did you make any setting modifications or it failed right away after upgrading?

 

local-kv.db does not exist.

 

I updated with the array not starting on reboot, made some changes to the network gateway address, then started the array.

[bonienl 1100]

3 minutes ago, pwm said:

But after the user acknowledges it, the system remembers the watermark value. Isn't that watermark value saved so it survives the next reboot?

 

 

Yes, it is stored on the flash device. Any new notification is only generated if the SMART value changes as compared to the last known (=stored) value.

[pwm 221]

1 minute ago, bonienl said:

 

Yes, it is stored on the flash device. Any new notification is only generated if the SMART value changes as compared to the last known (=stored) value.

 

find-watermark-for-drive;

if no-persistant-ecc-for-drive {

    save-as-current-watermark

}

if ecc != saved-value {

    emit-event

    save-as-current-watermark

}

Edited March 2, 2018 by pwm

[bonienl 1100]

5 minutes ago, dlandon said:

local-kv.db does not exist.

 

That's weird. Docker creates this file immediately after it is started (if not present). It holds all docker network assignments.

 

If you do a "rc.docker restart" from CLI, still this db file isn't there?

 

[dlandon 576]

1 minute ago, bonienl said:

 

That's weird. Docker creates this file immediately after it is started (if not present). It holds all docker network assignments.

 

If you do a "rc.docker restart" from CLI, still this db file isn't there?

 

Sorry, I did it with the docker stopped.  If the docker is started it is there.

 

I rolled back and now 6.4.1 has the same problem.  What do I need to do to get out of this problem on either version?

[bonienl 1100]

10 minutes ago, dlandon said:

I updated with the array not starting on reboot, made some changes to the network gateway address, then started the array.

 

Can you post your docker.cfg file, it is under /config on your flash device.

[bonienl 1100]

4 minutes ago, dlandon said:

I rolled back and now 6.4.1 has the same problem.  What do I need to do to get out of this problem on either version?

 

With docker started

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

 

[bonienl 1100]

16 minutes ago, pwm said:

 

if no-persistant-ecc-for-drive {

 

 

LOL

If only it was just that easy 

[dlandon 576]

13 minutes ago, bonienl said:

 

Can you post your docker.cfg file, it is under /config on your flash device.

I decided to stay on 6.5.0 and sort through this issue.

10 minutes ago, bonienl said:

rm /var/lib/docker/network/files/local-kv.db /etc/rc.d/rc.docker restart

This did not resolve the problem.

 

docker.cfg:

DOCKER_ENABLED="yes"
DOCKER_IMAGE_FILE="/mnt/cache/system/docker/docker.img"
DOCKER_IMAGE_SIZE="15"
DOCKER_APP_CONFIG_PATH="/mnt/cache/appdata/"
DOCKER_APP_UNRAID_PATH=""
DOCKER_AUTHORING_MODE="no"
DOCKER_LOG_ROTATION="yes"
DOCKER_LOG_SIZE="10m"
DOCKER_LOG_FILES="1"
DOCKER_CUSTOM_NETWORKS="eth1"
DOCKER_USER_NETWORKS="remove"

Appreciate the help.  I need to get my ownCloud working.

[bonienl 1100]

Try to add br0 manually

docker network create -d macvlan --subnet=192.168.1.0/24 --gateway=192.168.1.1 -o parent=br0 br0

 

[dlandon 576]

4 minutes ago, bonienl said:

Try to add br0 manually

docker network create -d macvlan --subnet=192.168.1.0/24 --gateway=192.168.1.1 -o parent=br0 br0

 

That did it.  Is this a bug, or is it my setup?

[bonienl 1100]

2 minutes ago, dlandon said:

That did it.  Is this a bug, or is it my setup?

 

I am not sure what went wrong for you. For me it is working and can't reproduce this error.

 

root@vesta:~# docker network ls
NETWORK ID          NAME                DRIVER              SCOPE
24d7006fca22        br0                 macvlan             local
a55a2930f25f        br0.5               macvlan             local
09e57f65a472        bridge              bridge              local
2e0d5b0ce86b        host                host                local
5b5cd0378f3a        none                null                local

 

[dlandon 576]

9 minutes ago, bonienl said:

 

I am not sure what went wrong for you. For me it is working and can't reproduce this error.

 

root@vesta:~# docker network ls
NETWORK ID          NAME                DRIVER              SCOPE
24d7006fca22        br0                 macvlan             local
a55a2930f25f        br0.5               macvlan             local
09e57f65a472        bridge              bridge              local
2e0d5b0ce86b        host                host                local
5b5cd0378f3a        none                null                local

 

On my test server.

docker.cfg:

DOCKER_ENABLED="yes"
DOCKER_IMAGE_FILE="/mnt/cache/system/docker/docker.img"
DOCKER_IMAGE_SIZE="15"
DOCKER_APP_CONFIG_PATH="/mnt/cache/appdata/"
DOCKER_APP_UNRAID_PATH=""
DOCKER_AUTHORING_MODE="no"
DOCKER_LOG_ROTATION="yes"
DOCKER_LOG_SIZE="10m"
DOCKER_LOG_FILES="1"
DOCKER_CUSTOM_NETWORKS="eth1"
DOCKER_USER_NETWORKS="remove"

docker network ls:

NETWORK ID          NAME                DRIVER              SCOPE
98c2722408d7        bridge              bridge              local
469c4c9736b7        host                host                local
0772cc6efdc5        none                null                local

Should the DOCKER_CUSTOM_NETWORKS be br0 and not eth1?