[GH-ISSUE #1515] Newt on client: too many colons in address #29765

Closed
opened 2026-06-13 09:52:32 -05:00 by GiteaMirror · 8 comments
Owner

Originally created by @Viceman256 on GitHub (Sep 22, 2025).
Original GitHub issue: https://github.com/fosrl/pangolin/issues/1515

Ran fine yesterday. Tried running the command today for my client and it fails to process:

INFO: 2025/09/21 21:43:40 Newt version 1.5.0
INFO: 2025/09/21 21:43:43 Websocket connected
ERROR: 2025/09/21 21:43:43 Failed to split endpoint: address https://[url]:51820: too many colons in address

The domain is entered exactly how it was provided from the portal, as well as the same way it's entered on my other clients. This client specifically is a raspberry pi.

Originally created by @Viceman256 on GitHub (Sep 22, 2025). Original GitHub issue: https://github.com/fosrl/pangolin/issues/1515 Ran fine yesterday. Tried running the command today for my client and it fails to process: INFO: 2025/09/21 21:43:40 Newt version 1.5.0 INFO: 2025/09/21 21:43:43 Websocket connected ERROR: 2025/09/21 21:43:43 Failed to split endpoint: address https://[url]:51820: too many colons in address The domain is entered exactly how it was provided from the portal, as well as the same way it's entered on my other clients. This client specifically is a raspberry pi.
GiteaMirror added the bug label 2026-06-13 09:52:32 -05:00
Author
Owner

@oschwartz10612 commented on GitHub (Sep 22, 2025):

Are you on ipv6 by chance? This could be related to invalid parsing of
ipv6 addresses. I think there is an open pr to fix this - need to look
into merging.

<!-- gh-comment-id:3316587144 --> @oschwartz10612 commented on GitHub (Sep 22, 2025): Are you on ipv6 by chance? This could be related to invalid parsing of ipv6 addresses. I think there is an open pr to fix this - need to look into merging.
Author
Owner

@Viceman256 commented on GitHub (Sep 22, 2025):

Are you on ipv6 by chance? This could be related to invalid parsing of
ipv6 addresses. I think there is an open pr to fix this - need to look
into merging.

Thanks for the super quick response! Both devices do have IPv6 enabled, but I'm unsure how to determine if it's routing through Ipv6. When I ping my server from my pi, it shows an ipv4 address, so I believe it should be routing through ipv4. In gerbil logs, I'm also seeing ipv4 addresses:

INFO: 2025/09/22 02:48:09 Peer [peer] added successfully
INFO: 2025/09/22 02:53:46 Clearing connections for removed peer with WG IP: [ip]
INFO: 2025/09/22 02:53:46 Cleared 0 connections for WG IP: [ip]
INFO: 2025/09/22 02:53:46 Peer [peer] removed successfully
INFO: 2025/09/22 02:53:46 Clearing connections for added peer with WG IP: [ip]
INFO: 2025/09/22 02:53:46 Cleared 0 connections for WG IP: [ip]
INFO: 2025/09/22 02:53:46 Peer [peer] added successfully

ETA:
I just removed IPv6 from my VPS and it still seems to present the same issue. Thanks for the idea though!

I also created a new site just to test that, presented the same issue on the client. I also uninstalled newt on the client and just reinstalled, same behavior.

<!-- gh-comment-id:3316595300 --> @Viceman256 commented on GitHub (Sep 22, 2025): > Are you on ipv6 by chance? This could be related to invalid parsing of > ipv6 addresses. I think there is an open pr to fix this - need to look > into merging. Thanks for the super quick response! Both devices do have IPv6 enabled, but I'm unsure how to determine if it's routing through Ipv6. When I ping my server from my pi, it shows an ipv4 address, so I believe it should be routing through ipv4. In gerbil logs, I'm also seeing ipv4 addresses: INFO: 2025/09/22 02:48:09 Peer [peer] added successfully INFO: 2025/09/22 02:53:46 Clearing connections for removed peer with WG IP: [ip] INFO: 2025/09/22 02:53:46 Cleared 0 connections for WG IP: [ip] INFO: 2025/09/22 02:53:46 Peer [peer] removed successfully INFO: 2025/09/22 02:53:46 Clearing connections for added peer with WG IP: [ip] INFO: 2025/09/22 02:53:46 Cleared 0 connections for WG IP: [ip] INFO: 2025/09/22 02:53:46 Peer [peer] added successfully ETA: I just removed IPv6 from my VPS and it still seems to present the same issue. Thanks for the idea though! I also created a new site just to test that, presented the same issue on the client. I also uninstalled newt on the client and just reinstalled, same behavior.
Author
Owner

@Viceman256 commented on GitHub (Sep 22, 2025):

Lastly, when I signed in my other clients were working well. I rebooted my VPS during testing when I disabled ipv6, and now all clients are giving me this colon error. Even those I haven't touched since yesterday.

Tested disabling firewall on VPS (Hetzner) and no luck. They can ping the server fine. Docker logs show pangolin is creating a websocket connection successfully too. Newt just for some reason seems to be putting a colon at the end of the addresses.

<!-- gh-comment-id:3316617106 --> @Viceman256 commented on GitHub (Sep 22, 2025): Lastly, when I signed in my other clients were working well. I rebooted my VPS during testing when I disabled ipv6, and now all clients are giving me this colon error. Even those I haven't touched since yesterday. Tested disabling firewall on VPS (Hetzner) and no luck. They can ping the server fine. Docker logs show pangolin is creating a websocket connection successfully too. Newt just for some reason seems to be putting a colon at the end of the addresses.
Author
Owner

@AstralDestiny commented on GitHub (Sep 23, 2025):

https://github.com/fosrl/olm/issues/8 This one @oschwartz10612 ?

<!-- gh-comment-id:3323990108 --> @AstralDestiny commented on GitHub (Sep 23, 2025): https://github.com/fosrl/olm/issues/8 This one @oschwartz10612 ?
Author
Owner

@oschwartz10612 commented on GitHub (Sep 23, 2025):

Yeah @AstralDestiny thats what I am thinking not sure need to spend some
time on this lol.

<!-- gh-comment-id:3324048285 --> @oschwartz10612 commented on GitHub (Sep 23, 2025): Yeah @AstralDestiny thats what I am thinking not sure need to spend some time on this lol.
Author
Owner

@heroichaiku commented on GitHub (Nov 30, 2025):

I'll say in advance that I've been using pangolin/newt for all of 5 minutes so this might not be useful.

I had the same issue seeing "too many colons" error log message. How I managed to fix this was an error in a configuration file.

What I had before in /opt/pangolin/config/config.yml for gerbil: configuration.

gerbil:
  base_endpoint: "https://pangolin.galaxy.lan"

and what fixed it. (remove the protocol form the end_point.

gerbil:
  base_endpoint: "pangolin.galaxy.lan"

Replace this with your own dashboard URL and 🤞 this is the solution for anyone who needs it.

<!-- gh-comment-id:3592617464 --> @heroichaiku commented on GitHub (Nov 30, 2025): I'll say in advance that I've been using pangolin/newt for all of 5 minutes so this might not be useful. I had the same issue seeing "too many colons" error log message. How I managed to fix this was an error in a configuration file. What I had before in `/opt/pangolin/config/config.yml` for `gerbil:` configuration. ```yml gerbil: base_endpoint: "https://pangolin.galaxy.lan" ``` and what fixed it. (remove the protocol form the end_point. ```yml gerbil: base_endpoint: "pangolin.galaxy.lan" ``` Replace this with your own dashboard URL and 🤞 this is the solution for anyone who needs it.
Author
Owner

@dawiddyd commented on GitHub (Dec 25, 2025):

I have the same issue :( My VPS provides only IPv6 address and because of that bug, I can't connect to the server from newt.

and what fixed it. (remove the protocol form the end_point.
gerbil:
base_endpoint: "pangolin.galaxy.lan"
Replace this with your own dashboard URL and 🤞 this is the solution for anyone who needs it.

Unfortunately, this does not fix my issue. I'm still having issues:

INFO: 2025/12/25 18:27:55 Connecting to endpoint: pangolin.my.domain
WARN: 2025/12/25 18:28:25 Initial reliable ping failed, but continuing: all 5 ping attempts failed, last error: failed to read ICMP packet: i/o timeout
WARN: 2025/12/25 18:28:30 Ping attempt 1 failed: failed to read ICMP packet: i/o timeout
WARN: 2025/12/25 18:28:35 Ping attempt 2 failed: failed to read ICMP packet: i/o timeout

Edit

I managed to make this working! I just put IPv6 address in brackets

gerbil:
  base_endpoint: "[IPv6 here]"
<!-- gh-comment-id:3691660761 --> @dawiddyd commented on GitHub (Dec 25, 2025): I have the same issue :( My VPS provides only IPv6 address and because of that bug, I can't connect to the server from newt. > and what fixed it. (remove the protocol form the end_point. > gerbil: base_endpoint: "pangolin.galaxy.lan" Replace this with your own dashboard URL and 🤞 this is the solution for anyone who needs it. Unfortunately, this does not fix my issue. I'm still having issues: ``` INFO: 2025/12/25 18:27:55 Connecting to endpoint: pangolin.my.domain WARN: 2025/12/25 18:28:25 Initial reliable ping failed, but continuing: all 5 ping attempts failed, last error: failed to read ICMP packet: i/o timeout WARN: 2025/12/25 18:28:30 Ping attempt 1 failed: failed to read ICMP packet: i/o timeout WARN: 2025/12/25 18:28:35 Ping attempt 2 failed: failed to read ICMP packet: i/o timeout ``` Edit I managed to make this working! I just put IPv6 address in brackets ``` gerbil: base_endpoint: "[IPv6 here]" ```
Author
Owner

@oschwartz10612 commented on GitHub (Jan 20, 2026):

I think we fixed this in the last newt release but please reopen if still the case.

<!-- gh-comment-id:3770626432 --> @oschwartz10612 commented on GitHub (Jan 20, 2026): I think we fixed this in the last newt release but please reopen if still the case.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/pangolin#29765