I just turned on Docker on my main server and installed a couple of apps. As soon as I try and go to the docker app webUI it times out. I have tried the normal different web browser and clearing cache etc. but that doesn't solve the issue. Yesterday I installed Docker on a friends server that I manage and it worked instantly.

 

I am assuming my problem is because I am using a pfSense VM and I have unRAID's default bridges turned off and define my bridges through the unRAID "go" file? Can anyone think of another reason why the apps webUI won't display?

 

Contents of bridges in "go"

# Start Bridges
brctl addbr out0
brctl addbr in0
brctl stp out0 on
brctl stp in0 on
brctl addif out0 eth2
brctl addif in0 eth1
ifconfig eth1 up
ifconfig eth2 up
ifconfig out0 up
ifconfig in0 up

OK, so more data!

 

I can access my docker apps WebUI's from my bare metal machine and iPhone. I cannot access my docker apps WebUI's from any of my Virtual Machines... So I guess my VM bridge (in0) cannot talk to the docker bridge? How do I fix this?

OK, so more data!

 

I can access my docker apps WebUI's from my bare metal machine and iPhone. I cannot access my docker apps WebUI's from any of my Virtual Machines... So I guess my VM bridge (in0) cannot talk to the docker bridge? How do I fix this?

As you know archedraft, I have the same setup as you and have just tested this out on my virtual machine.  I can browse the web from it fine but can't access any docker webgui apps.

 

I'll have a look tomorrow as just got into bed, but suspect out the two of us it won't be me who finds a solution. 

 

 

OK, so more data!

 

I can access my docker apps WebUI's from my bare metal machine and iPhone. I cannot access my docker apps WebUI's from any of my Virtual Machines... So I guess my VM bridge (in0) cannot talk to the docker bridge? How do I fix this?

 

Just something to try, whichever the network settings are for a particular docker, be they host or bridge, try the other.

 

bear in mind if you select host you cannot define a port, so choose one that you know doesn't clash with anything else.

 

that's about all i can suggest, if that's not it, you need someone au fait with network routing.

 

OK, so more data!

 

I can access my docker apps WebUI's from my bare metal machine and iPhone. I cannot access my docker apps WebUI's from any of my Virtual Machines... So I guess my VM bridge (in0) cannot talk to the docker bridge? How do I fix this?

As you know archedraft, I have the same setup as you and have just tested this out on my virtual machine.  I can browse the web from it fine but can't access any docker webgui apps.

 

I'll have a look tomorrow as just got into bed, but suspect out the two of us it won't be me who finds a solution.

Thanks for testing. If both of us are having the same issue it must be because of the user created bridges... Now we just need to understand how the default unRAID bridge handles dockers and VM's.

 

OK, so more data!

 

I can access my docker apps WebUI's from my bare metal machine and iPhone. I cannot access my docker apps WebUI's from any of my Virtual Machines... So I guess my VM bridge (in0) cannot talk to the docker bridge? How do I fix this?

As you know archedraft, I have the same setup as you and have just tested this out on my virtual machine.  I can browse the web from it fine but can't access any docker webgui apps.

 

I'll have a look tomorrow as just got into bed, but suspect out the two of us it won't be me who finds a solution.

Thanks for testing. If both of us are having the same issue it must be because of the user created bridges... Now we just need to understand how the default unRAID bridge handles dockers and VM's.

 

I came across this issue a few weeks ago, now I think about it, confused the hell out of me!

 

Same phenomenon?

I came across this issue a few weeks ago, now I think about it, confused the hell out of me!

 

Same phenomenon?

 

So just to clarify; if you used the user bridges created via the "go" file and turned on a VM, you could not access the docker webUI's, correct?

 

and if you let unRAID create the br0 bridge and turned on a VM, you could access the docker webUI's via the VM, correct?

I came across this issue a few weeks ago, now I think about it, confused the hell out of me!

 

Same phenomenon?

 

So just to clarify; if you used the user bridges created via the "go" file and turned on a VM, you could not access the docker webUI's, correct?

 

and if you let unRAID create the br0 bridge and turned on a VM, you could access the docker webUI's via the VM, correct?

 

Only thing I can confirm is the behaviour as I detailed in the post.

 

I've just edited my go file and am rebooting, I'll fire up a VM and post the results.

Alright I have made progress 

 

I can now access my dockers through my VM's. I will post how I did this a little later. I just noticed something strange and I think I can clean it up before I post my steps.

Alright I have made progress 

 

I can now access my dockers through my VM's. I will post how I did this a little later. I just noticed something strange and I think I can clean it up before I post my steps.

 

Cool, because I've just buggered up my VM's trying to change the bridge, and now the wife can't watch "The Voice" 

Here are the steps that I used to fix this. I am sure there is a better way but it's working now so I'll leave it be.

 

1. Changes to the bridges in "go"

# Start Bridges
brctl addbr out0
brctl addbr br0
brctl stp out0 on
brctl stp br0 on
brctl addif out0 eth2
brctl addif br0 eth1
ip addr add 192.168.1.205/24 dev br0
ifconfig eth2 up
ifconfig eth1 up
ifconfig out0 up
ifconfig br0 up

Notice that I changed all the in0 to br0 (this fixes a different issue where you keep getting error messages: "Received packet from invalid interface."

 

Also I added the line "ip addr add 192.168.1.205/24 dev br0". This gives br0 & the dockers a static IP address (note this needs to be an unused ip address). So my unRAID machine is 192.168.1.200 and now when I connect to my dockers I use 192.168.1.205:port#.

 

Now I can access my dockers webUI's from my VM's and Bare Metal machines.

 

EDIT: also make sure you change and VM's to br0 instead of in0.

Here are the steps that I used to fix this. I am sure there is a better way but it's working now so I'll leave it be.

 

1. Changes to the bridges in "go"

# Start Bridges
brctl addbr out0
brctl addbr br0
brctl stp out0 on
brctl stp br0 on
brctl addif out0 eth2
brctl addif br0 eth1
ip addr add 192.168.1.205/24 dev br0
ifconfig eth2 up
ifconfig eth1 up
ifconfig out0 up
ifconfig br0 up

Notice that I changed all the in0 to br0 (this fixes a different issue where you keep getting error messages: "Received packet from invalid interface."

 

Also I added the line "ip addr add 192.168.1.205/24 dev br0". This gives br0 & the dockers a static IP address (note this needs to be an unused ip address). So my unRAID machine is 192.168.1.200 and now when I connect to my dockers I use 192.168.1.205:port#.

 

Now I can access my dockers webUI's from my VM's and Bare Metal machines.

 

EDIT: also make sure you change and VM's to br0 instead of in0.

 

So I just changed all my references to in0 to br0, that bit I get.

 

I'm confused by what's going on with

 

ip addr 192.168.1.205/24 dev br0

 

Why does that solve the error messages and why does it change how you access the docker apps.

 

For instance:

 

My unraid IP is 192.168.1.1

 

If I use my desktop PC to access NZBGet I use 192.168.1.1:6789

 

As far as I'm aware the dockers don't use a bridge, so if I added

ip addr add 192.168.1.205/24 dev br0

does that mean I then access NZBGet on 192.168.1.205:6789 or is that the IP Address I'd use only if I was accessing NZBGet from a VM?

 

 

So I just changed all my references to in0 to br0, that bit I get.

 

I'm confused by what's going on with

 

ip addr 192.168.1.205/24 dev br0

 

Why does that solve the error messages

That specific code does not solve the error messages, changing all the in0 to br0 fixes the error messages. (I am assuming unRAID is setup to know how to handle br0 but if you make up something like in0 then it doesn't.

 

why does it change how you access the docker apps.

 

For instance:

 

My unraid IP is 192.168.1.1

 

If I use my desktop PC to access NZBGet I use 192.168.1.1:6789

 

As far as I'm aware the dockers don't use a bridge, so if I added

ip addr add 192.168.1.205/24 dev br0

does that mean I then access NZBGet on 192.168.1.205:6789 or is that the IP Address I'd use only if I was accessing NZBGet from a VM?

Correct your NZBGet would now be 192.168.1.205:6789

So I just changed all my references to in0 to br0, that bit I get.

 

I'm confused by what's going on with

 

ip addr 192.168.1.205/24 dev br0

 

Why does that solve the error messages

That specific code does not solve the error messages, changing all the in0 to br0 fixes the error messages. (I am assuming unRAID is setup to know how to handle br0 but if you make up something like in0 then it doesn't.

 

why does it change how you access the docker apps.

 

For instance:

 

My unraid IP is 192.168.1.1

 

If I use my desktop PC to access NZBGet I use 192.168.1.1:6789

 

As far as I'm aware the dockers don't use a bridge, so if I added

ip addr add 192.168.1.205/24 dev br0

does that mean I then access NZBGet on 192.168.1.205:6789 or is that the IP Address I'd use only if I was accessing NZBGet from a VM?

Correct your NZBGet would now be 192.168.1.205:6789

 

OK, experimented before your reply, didn't add the ip addr line and was still getting the error messages in my log. 

 

So do you enter your unraid gui with 192.168.1.200 or 192.168.1.205 ?

 

This may take me a while to get my head around!

OK, experimented before your reply, didn't add the ip addr line and was still getting the error messages in my log. 

 

So do you enter your unraid gui with 192.168.1.200 or 192.168.1.205 ?

 

This may take me a while to get my head around!

Strangely enough, I can enter my unRAID gui from either 192.168.1.200 or 192.168.1.205

 

The dockers can only be accessed through 192.168.1.205:port#

is br0 enable in the Unraid settings?  Mine isn't perhaps that's the issue as I am still seeing the error messages and am still able to use my apps with the default unraid IP address..

nope I still have unRAID's default bridge disabled in the web GUI. Did you change all your KVM configs to use br0? If not maybe that's why you are still getting the "Received packet from invalid interface."

Ignore me, I'm a pillock!

 

I'd missed the add from

 

ip addr add 192.168.1.101/24 dev br0

 

My go file now looks like this:

 

#Ethernet Bridge Setup

brctl addbr br0
brctl addbr out0

brctl stp br0 on
brctl stp out0 on

brctl addif br0 eth1
brctl addif out0 eth2

ip addr add 192.168.1.101/24 dev br0
ip addr add 192.168.1.102/24 dev out0

ifconfig eth1 up
ifconfig eth2 up

ifconfig br0 up
ifconfig out0 up

 

My default Unraid IP address is 192.168.1.1

 

Interestingly I can now access the docker apps from both 192.168.1.1:PORT and 192.168.1.101:PORT

 

In the top right corner of my Unraid GUI my IP address is showing as 192.168.1.101

 

My error messages have disappeared like you've described.

 

I'm still completely confused by what is actually happening behind the scenes, not that it matters.

 

archedraft I'm seriously considering putting a tag to my username reading

 

"UNsophisticated beta tester" 

Awesome glad you got it up and running. Also you shouldn't need the "ip addr add 192.168.1.102/24 dev out0" in your "go" file. I appreciate you testing the pfSense VM with me, you honestly saved me a bunch of time and tests by confirming things up front.

Awesome glad you got it up and running. Also you shouldn't need the "ip addr add 192.168.1.102/24 dev out0" in your "go" file. I appreciate you testing the pfSense VM with me, you honestly saved me a bunch of time and tests by confirming things up front.

 

Mate you've helped me so much, and been so very very patient, that I'm very humbled by your thanks, they're not necessary at all.

Awesome glad you got it up and running. Also you shouldn't need the "ip addr add 192.168.1.102/24 dev out0" in your "go" file. I appreciate you testing the pfSense VM with me, you honestly saved me a bunch of time and tests by confirming things up front.

 

Ran into a few issues today after a few reboots and finding my VMs couldn't get an IP address from DHCP.

 

Quick question, in your network settings at the top what MAC address is your machine using? Eth0, Eth1 or Eth2

 

I've currently got Eth1 & Eth2 bound to br0 and out0 respectively and Eth0 was my default Unraid IP, I'm now finding that Unraid has claimed the Eth1 interfaces mac address, which is what I think may be the problem.

Eth2 is for the modem. Eth1 goes from pfSense to my 8 port switch. UnRAID uses eth0. Both eth0 and eth1 have to have an Ethernet cable or I do not have access to the Internet on my VMs.

My unRAID also shows my eth1 MAC address as default but it needs both eth1 and eth0 for some reason.

My unRAID also shows my eth1 MAC address as default but it needs both eth1 and eth0 for some reason.

 

Ok, thanks, still trying to get things working my end.

 

Is your eth0 IP allocated by DHCP (with or without a reserved address) or set in the Unraid settings?

I gave both unRAID eth0 and br0 eth1 static IP address in pfsense that are outside the range of my DHCP on pfsense.