Not sure where to place this so just doing it here.

 

The newish start stop procedure via emHTTP for docker is klunky. To recap the old method was a button to enable the service automatically and a clear STOP button.

 

The new method merges these into one so in order to stop docker you change the autostart to stop then apply.

 

This is not nice or intuitive it should be two seperate configuration options.

 

 

I thought the same until I clicked on the containers icon.  There lies the secret menu. Like you, I preferred it just one click away.

Not sure where to place this so just doing it here.

 

The newish start stop procedure via emHTTP for docker is klunky. To recap the old method was a button to enable the service automatically and a clear STOP button.

 

The new method merges these into one so in order to stop docker you change the autostart to stop then apply.

 

This is not nice or intuitive it should be two seperate configuration options.

Not sure I agree!  Note that the option is not to AutoStart docker but whether it is enabled or not.  It seems relatively intuitive to change that setting and hit Apply as you would elsewhere in the GUI.  Setting Enabled to No and hitting apply does the stop.  I would have thought to a new user that is all they need to know/do.

 

If you are suggesting that you want separate Start/Stop buttons, then perhaps the Use Case for these needs to be better explained?  I am not saying that there is not one, just that it needs to be more than suggesting the current UI is 'clunky' which is a subjective impression anyway.

 

I thought that it was a deliberate decision to make stopping a container a two-click action.

It just doesnt follow what is typical for a webgui but i freely admit its kinda how we do it now. But consider every other service control gui I can think off has a page that lets you start, stop or restart a service. Most also define autostart on this single page as well whereas In unRAID you have to go into 5+ pages to do this docker being just one of them.

 

Or put it this way in emHTTP the only way to stop docker is to actually tell unRAID never to start docker again.... and then to start it you are telling it to start docker automatically for ever more. This is not the Linux way, these two things are separate decisions and actions.

 

A few random screenshots on how other projects present it

 

Not sure where to place this so just doing it here.

 

The newish start stop procedure via emHTTP for docker is klunky. To recap the old method was a button to enable the service automatically and a clear STOP button.

 

The new method merges these into one so in order to stop docker you change the autostart to stop then apply.

 

This is not nice or intuitive it should be two seperate configuration options.

We changed it to match how all other services on unraid are enabled or disabled. See SMB, NFS, afp, etc.  We did it to conform to the rest of the system.

I can see that but IMHO this is a step backwards. Rather the other services should be brought forwards.

I can see that but IMHO this is a step backwards. Rather the other services should be brought forwards.

Our current approach consolidates two controls into one (start/stop, and autostart).  I think that's a step forward in simplicity, not backward.

I can see where you are coming from but I cite my original problem with this design:

 

Or put it this way in emHTTP the only way to stop docker is to actually tell unRAID never to start docker again.... and then to start it you are telling it to start docker automatically for ever more. This is not the Linux way, these two things are separate decisions and actions.

 

This is a reason every single Linux and windows server ever installed does it the normal way.

 

I am sure with some effort i could find some clever kernel people making a proper arguement for this in some mailing list or simimair but as an exmaple the current way requires two mouse clicks to stop the service and the old way required one. This alone makes it less slick.

 

IMHO it is needlessly atypical and inflexible or rather...  different for different sake.

I can see where you are coming from but I cite my original problem with this design:

 

Or put it this way in emHTTP the only way to stop docker is to actually tell unRAID never to start docker again.... and then to start it you are telling it to start docker automatically for ever more. This is not the Linux way, these two things are separate decisions and actions.

 

This is a reason every single Linux and windows server ever installed does it the normal way.

 

I am sure with some effort i could find some clever kernel people making a proper arguement for this in some mailing list or simimair but as an exmaple the current way requires two mouse clicks to stop the service and the old way required one. This alone makes it less slick.

 

IMHO it is needlessly atypical and inflexible or rather...  different for different sake.

I don't see anyone else really complaining about this though. It just seems like an odd thing to nit pick on given that its been this way throughout the entire product since before v5.

 

I really don't think this is a big deal, however we could review this for change at some point in the future. Just not expecting us to do this before we move to 6.0 final.

Its not the end of the world but if i didn't raise it there is definitely no chance it would be changed.

 

The whole service thing need reworkled anyway e.g. there are daemons you cant control at all either the old or new way.

 

My opinion is still to move to a model people know and understand because it is the norm.

 

Perhaps we should change this ticket name and give it a late milestone for consideration in detail when things arent so hectic.

Admittedly I didn't dig into this much, but from 1st glance (and at that point gave up) I assumed there was no longer a stop button.

The only toggle seemed to be specifically for autostart yes/no, so that wasn't it.

I clicked on the docker name itself and something came up, but didn't see a stop.

So at that point I thought, ehhh, whatever and did something else.

 

So as a regular joe user I think it is a step backwards at least at 1st glance.

 

Knowing all this now, what do I have to click/do to stop a docker and not disable it from auto starting on next reboot?

That is interesting as it over an above what I have raised here. i.e. what you are looking at is how to stop a container and what I raised is how to stop all of docker etc

Admittedly I didn't dig into this much, but from 1st glance (and at that point gave up) I assumed there was no longer a stop button.

The only toggle seemed to be specifically for autostart yes/no, so that wasn't it.

I clicked on the docker name itself and something came up, but didn't see a stop.

So at that point I thought, ehhh, whatever and did something else.

 

So as a regular joe user I think it is a step backwards at least at 1st glance.

 

Knowing all this now, what do I have to click/do to stop a docker and not disable it from auto starting on next reboot?

In the Dashboard or on the Docker page, click on the docker's icon and you get a popup and one of the choices is start/stop.

Admittedly I didn't dig into this much, but from 1st glance (and at that point gave up) I assumed there was no longer a stop button.

The only toggle seemed to be specifically for autostart yes/no, so that wasn't it.

I clicked on the docker name itself and something came up, but didn't see a stop.

So at that point I thought, ehhh, whatever and did something else.

 

So as a regular joe user I think it is a step backwards at least at 1st glance.

 

Knowing all this now, what do I have to click/do to stop a docker and not disable it from auto starting on next reboot?

In the Dashboard or on the Docker page, click on the docker's icon and you get a popup and one of the choices is start/stop.

 

Got it, thanks!..

I think I clicked on the name and not the icon.

Back in the "bad old days" (when we were still using plugins ), we could back up our cache drive with rsync using a script (thanks dirtysanchez) dropped in the appropriate cron folder that stopped the services, rsync'd the drive, and restarted the services.  How can I do the same thing in this new world of dockers?

 

#Stop services
/etc/rc.d/rc.sabnzbd stop
/etc/rc.d/rc.sickbeard stop
/etc/rc.d/rc.couchpotato stop

#Backup cache via rsync
date >/var/log/cache_backup.log
/usr/bin/rsync -avrtH --delete /mnt/cache/appdata/ /mnt/user/System/unRAID_cache/appdata >>/var/log/cache_backup.log
/usr/bin/rsync -avrtH --delete /mnt/cache/docker.img /mnt/user/System/unRAID_cache >>/var/log/cache_backup.log

#Start services
/etc/rc.d/rc.sabnzbd start
/etc/rc.d/rc.sickbeard start
/etc/rc.d/rc.couchpotato start

See here

Thanks trurl!