-
-
Notifications
You must be signed in to change notification settings - Fork 1.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
PIA not able to retrieve OpenVPN config bundles #1548
Comments
Right now, only https://www.privateinternetaccess.com/openvpn/openvpn.zip seems to be present. |
All of the configuration files appear to be present but without the "-nextgen" suffix. |
@haugene ping |
Well then that can be worked around with the appropriate Environment variable. It also seems that the file names have changed as well, i.e. "CA Toronto" is now "ca_toronto". That will be a breaking change for anyone else using PIA. I'm not sure who maintains the Unraid config, but they will probably need to update the template, as well, but I can easily work around that for myself. |
This works for a k8s deployment as well. Setting the env variable "PIA_OPENVPN_CONFIG_BUNDLE" to any of the above configuration endpoints removing the "-nextgen" suffix and updating the ovpn config names which I found in the log file. |
This just started for me as well and setting the environment variable didn't work for me. I am new to this, any way to see the log files? Maybe I am missing something |
Thanks for help from above. Here is an example if anyone struggle like me in the first time:
The |
Thanks @YujiShen ! That was it! |
Thanks @YujiShen !
That way the script will choose a random server. |
Also if you are using Unraid like I am, for the |
In addition to doing this, I had to remove the LOCAL_NETWORK env variable. |
Guys, Thanks in advance. I am still struggling with this issue. My containers are running on ubuntu 18.04 with docker 19.03. Also removed the LOCAL_NETWORK variable, and though I am no longer getting the unzip errors now I am getting authentication errors, log below. What am I missing: Provider PIA has a custom setup script, executing it |
Same exact issue. This is getting real frustrating. |
@jsorondo @trwolff04 I think you are using 2.x image, these are next gen server config, and PIA seems deprecate old server by this change. Try |
Actually I'm an idiot and used the wrong PIA password. Back in business. |
FIxed thanks. I had a third container still holding the old image. Stopped container, deleted and purged old image. Issue resolved |
I had to do this to get mine to work on OMV4. Thanks a bunch, everyone! Y'all some lifesavers! |
Not working for me :/ - keeps saying 'inactivity timeout' then exiting. |
Can you post your output from the logs as well as variables? |
`
environment: Everything else is default. |
@MartenvanWezel try |
That was also my train of thought. Thank you @YujiShen |
Nope, it can find 'Netherlands' it seems, if I lowercase it it says: `
|
@MartenvanWezel Are you using 3.x image? Or 2.x? The netherlands.ovpn is inside that zip file. If container is using correct URL, it should find it. |
@MartenvanWezel Which image version are you using? I verified the profile names are lowercase as stated by @YujiShen . |
Had this issue, was fixed by @YujiShen most upvoted post here. |
@jsorondo - haugene-latest. I'll try a manual update. |
Update: I think I'm now on a better version, but it's possibly trying ipv6 and failing? `
` |
@YujiShen This seems to be the main problem why build 2.14 refuses to work. Is there any way to configure it so 2.14 will be able to use the new files? Would setting OVPN_CONFIG to custom and providing it with a manual file do it any good? Would port forwarding still work? |
Jumping in here. I merged #1552 yesterday which should fix this issue. It has been on the master branch for ~16 hours and just merged it to dev as well. @superkrups20056 You can mount the new configs into the old image and it would connect to VPN. But the port forwarding will not work, there is a new script for that and it's only in the new image. There are some bugs reported on the new script but for my use it seems to work fine. Have not gotten the time to go into detail on it yet. Brushing up the repo on many fronts these days and the PIA port-updater script is in the pipeline 👍 |
@haugene I had some issues and had to reinstall the docker and am now getting this issue. Any ideas?
|
@11jwolfe2 Have you pulled the latest image? It's hard to see as you haven't provided the complete log?
This is the revision where the URL to PIA configs are fixed. Pia changed their configs, called the new ones "nextgen" and now that the old ones are gone they've renamed it again - removing the "nextgen". The latest version has this fixed. As for the mknod error, are you mounting the tun device? Note that after upgrading to 3.x versions the If you're still having issues after this please post the docker setup you're running (docker run command or docker-compose file) and the complete logs. Then it will be easier to spot any potential issues 😄 |
Right container rev. But it seems to try using ipv6 and that doesn't work. (not sure why, but I never tried to set up ipv6 inside my own LAN). How do I force ipv4?
|
@MartenvanWezel You can try to disable IPv6 by using the option @trowgundam and the others that had the config issue, this is now resolved right? The name of the config files have changed but that is up to PIA to decide. I don't think I will put it on this project to normalize config names. Servers can be removed by the providers and the container config will then be invalid and a manual change is needed. This is kind of the same case to my mind. We could implement some case insensitive lookup of the files. I'm open to that, but then I urge you to create a separate issue to propose that. |
Yes, the latest pulls of the docker resolve the config file issue without needing to set the environment variable. It is able to download the Zips from PIA just fine. The maintainer of the Unraid template has resolved the other issue (with the name of the OpenVPN files changing), so anyone using that should be good now as well, if they've updated their dockers dockers. |
Thank you for the responses, the "-e PIA_OPENVPN_CONFIG_BUNDLE=openvpn " line was the missing new configuration line that solved it |
@mugiwararegex You don't need that line anymore. It was a workaround until it was fixed in the image. It is now set as the default value and you can remove it (if you pull the newest version). |
I am running latest-armhf (6059b231e17a31edd265d9534ba351a6d83361a0e35726bc71b8fbb9bec338e2), Is this not updated correctly? I was running 3.02 but figured id change to latest just to make sure. I tried commenting out the mount /dev/net/tun sections in my kubernetes helm file, but no luck.
Also when i run kubectl logs all i get is the quoted text from above. |
@11jwolfe2 I'll delete those tags right away. latest-armhf is no longer updated. The :latest tag is now multiarch and can be run from ARM. So just use :latest tag and hopefully that should work better. |
@haugene so when i add latest to my tag, it pulls this version
I do not believe this is right because I see on docker hub this is what I should have.
Any idea why I am not pulling the correct image. I am getting this error still...
|
@11jwolfe2 If you skip the LOCAL_NETWORK does it start then? Are you running with --net=host? EDIT: I first thought I knew what the sha digests were referring to. Then after posting it I saw that it didn't match what I posted either 😅 Anyways. The Docker hub currently lists e3634cc99d64 as the digest of the amd64 image. After pulling the image locally and running Docker inspect wee see that it is not the image ID and neither the repo digest that I thought it was. Too tired and don't have time to look at this now. But I wonder where it's hiding. What you can do though are the labels when you run docker inspect on the image you've pulled. It should give the revision. But that should also be printed when you run the container so I don't understand that being your complete logs. Have you run The docker inspect on my image.
|
@haugene since im running the docker container in a kubernetes cluster the best way i know to get logs is kubectl logs. This time I tried to get as much as possible and got this.
I get this error with or without Local_Netowrk once again here is my config yaml for kubernetes.
I appreciate the help |
It just doesn't make sense to me. The logs seem to do other stuff than your deploy.yaml suggests. But your logs say:
Which comes from this code in the container:
Which points to LOCAL_NETWORK being set, it shouldn't print that otherwise. Can you do a |
Sorry for the late reply I really wanted to try and figure this out myself but I havent had any luck. So as for the local network thing. I must have added the wrong log. I tried with and without the local network config and keep having the same issue. Here is the correct log.
as for the yaml output here is it.
I feel like the real issue is coming from this here...
am i missing a configuration parameter? |
SOLVED!!!!!! For anyone using the helm chart made by @bananaspliff located here You have to get rid of the following section. It is not required anymore. After I removed this. Everything worked perfectly.
|
That did the magic, Please note the log when you see the error. Countries are changed to lower cases now |
Describe the problem
PIA will not connect anymore. The docker gets itself into a loop trying to download the OpenVPN config files from PIA. This seems to be because the Zip files it is trying to download, the following:
All of these now return a 404 Error. The docker just goes into a continuous loop of trying to download one of these files.
Add your docker run command
I use Unraid, and here is the command the UI executes to create the docker, formatted for easier readability:
Logs
This repeats just with different Temp File names over and over.
Host system:
Unraid 6.8.3 (if it makes a difference I am using the LinuxServer.io version with Nvidia support for dockers).
The text was updated successfully, but these errors were encountered: