Cliffy

I am having very similar issues as well. I have tried all RC versions of 6.6 and they all drop my network connection some time in the evening. I believe it's happening when backups are being copied to my unraid share from my pc, but I'm just guessing here. I cannot ping the unraid from my pc. I do have a monitor/keyboard connected to unraid. The console looks normal with no error messages. I log in and try to ping anything on my network and it fails to reply until a reboot then it's back to working normal with network connectivity. When reverting back to 6.5.3 I have no connectivity issues.

FIXED IT! I have 1:1 Nat working. I have 6 publicly facing WAN IPs and not 8 as I thought. I was trying to give the broadcast IP to my apache docker. Hence why it was working on the local network and not over the internet. Thanks for all your help CHBMB and your suggestions... it got me thinking.

Even if I could, I would be routing using the IP from my router as I cannot specify which IP and port to forward.

It's not supplied by my ISP. I'm using a Cisco RV082 router. It has pretty basic functions.

I'm not sure I follow this. So I shouldn't be using br0, but instead use host. Then change ports to 81 and 444? Then forward the ports 80 to 81 to my unraid box from my router? MY router does not allow me to forward one port to a different port.

As far as I know I have it all up and running. My apache docker works and serves the pages I want when I browse the local IP. I'm using br0 because I don't know of any other way (or how to) give my docker an IP address other than the IP of my unraid server. If I don't use br0, my apache docker will have the same IP as my unraid server which is a problem because it uses both ports 80 and 443. Maybe I'm going about this all wrong I don't know. I know my basic way around networking, but not so much with vlans, bridging, etc...

It may be worth mentioning that I am using UnRAID v6.4.0-rc7a

Yes, I am using br0 for my docker and its using the macvlan driver.

I own a range of 8 public static IP addresses. Example: 1.1.1.178-185 They are all routed to my router by my ISP. I use different IPs for different services behind my router and route them using 1:1 NAT. For example let's say that the IP 1.1.1.178 is a public IP and it's the IP of my router. 1.1.1.180 is also a public IP which I have setup a 1:1 NAT to a local linux webserver on my network. This works fine. Lets say 1.1.1.181 is the IP I want to forward to my new apache docker on my network. When I setup 1:1 nat on this public IP (1.1.1.181) to the local network IP of my docker, the docker can no longer access the internet. I'm guessing there is a gateway / DNS / routing issue perhaps with the br0 docker network?

Thanks CHBMB... that all makes sense. However I would like to assign a different IP other than that of my router since I have other IPs available. I have other websites running on other IPs as well on an Ubuntu server with 1:1 NAT without any issues for years. I was hoping it was as simple as setting up a 1:1 Nat with the local IP of my apache docker, but as I mentioned it breaks the connection when I enable 1:1 Nat on my router. The docker can no longer access the internet.

I have a range of static public IP addresses. I pointed the subdomain to one of the public IP addresses other than the IP address of my router. If I forward the ports I will then lose access to my router remotely. The port forwarding is only forwarding the requests from the router IP and not the subdomain IP.

Thanks... I would like to use the standard ports 80 & 443. I do not want to specify a port when accessing the website.

I'm having an issue relating to this docker however it's a network connectivity issue. I am using br0 as the network type with it's own local IP address. I want to be able to access the apache server from subdomain.mydomain.com. Everything seems to be working fine when I test the site within the network, but I cannot connect when outside the network (ie: accessing from my phone). When I enable 1:1 nat on our cisco router (RV082) to map the external public ip address to the local ip address of the docker it breaks the connection. When I exec into the docker I can no longer ping or access external domain names or ip addresses whereas I can when the 1:1 nat on the router is removed. Could it be an issue with the router firmware not supporting it? Is this the suggested method of accessing the apache server with my own subdomain? Thanks for your help.

I am having the same issues as well with windows 10 and credentials. I have private shares and never had any issues in the past connecting to them until v6.3 I have tried the suggestions, restarting workstation service, etc... and no go.

I deleted my docker.img however still having the same problem. UPDATE: After disabling and enabling Docker, it now started and the Docker tab is now displayed.

I'm having the same problem and error message in the log file. I noticed I was having problems with OwnCloud not working so I tried to Start/Stop it, however it would not start. I decided to reboot UnRaid Server and that's when the Docker settings tab disappeared and I can no longer start docker. I am using Unraid v6.0.1

Excellent... thanks Squid. Your method of a subfolder worked fine other than the double listing. I would say it was more of an issue with dockerman. It looks to me that the templates are sorted by date hence the reason why new user templates would appear in a second section. I took your update.

Well.... that sorta worked. I now only have the one templates-user listed, however I lost my Docker2XML section/folder. Probably because it's not listed in the template repositories and lost them on some sort of docker templates refresh. No big deal. I only had 2 test dockers in there.

Found them and fixed it! They are located on the flash drive under config\plugins\dockerMan As you mentioned, Docker2XML was located in the templates-user directory. I moved it into the templates folder.

ahhh! Where are these templates stored on the unRaid system? Am I able to move these templates into one group again from the console?

I currently have two listings of [templates-users] when I go to add container. Each listing consists of different containers that I have installed in the past. What would cause some templates to be stored in one location verses the other location? See attachment...

Thanks Squid... sorry I didn't notice the other thread mentioning this issue.

Today I added the following to the template repositories: https://github.com/sparklyballs/beta-repo/tree/master/Sparkly-Beta After I click the "save" button I receive the following error: Warning: unlink(/tmp/tmp-1637217078.tar.gz): No such file or directory in /usr/local/emhttp/plugins/dynamix.docker.manager/dockerClient.php on line 157 I tried this on 2 servers and they both resulted in the same error. Now when I click to add a container no docker templates appear in the drop down list except for my user defined templates. I then continued to try and revert back to the list of repositories I had before. I removed all the template repositories and clicked save then added https://github.com/gfjardim/docker-containers/tree/templates and click save, but now I receive the same error as above everytime.

wow... awesome! That was quick!

I found this: