[Plugin] Tailscale

[macmanluke]

after updating to the latest version on one of my machines im getting

 

Quote

 

/usr/local/emhttp/plugins/tailscale/update-settings.sh: line 3: /boot/config/plugins/tailscale/tailscale.cfg: No such file or directory
Taildrop not configured or share not available, disabling
rm: cannot remove '/etc/config/uLinux.conf': No such file or directory
Stopping tailscaled.
Starting tailscaled:  /usr/local/sbin/tailscaled
 

 

Tried removing and reinstalling no luck. Is on 6.12rc2
Other machine is fine (but 6.11.5)

[EDACerton]

4 minutes ago, macmanluke said:

after updating to the latest version on one of my machines im getting

 

 

Tried removing and reinstalling no luck. Is on 6.12rc2
Other machine is fine (but 6.11.5)

Is Tailscale running? Those messages looks fine. It just means that you haven’t changed any settings in the config page. 
 

I’ll work on making that a little less intimidating. 

[macmanluke]

sorry didnt read that one, previous message before the reinstall didnt have starting tailscale line.
 

Seem to have gotten it working now after the reinstall and reauth

 

[ericswpark]

@EDACerton same, just found a `dead.letter` in the root home directory with the contents being the same as the output from @macmanluke's reply. I'm running 2023.04.01.

[mavgink]

Quick question:
I received an email:
---
Output from your job 32
Taildrop not configured or share not available, disabling
Stopping tailscaled.
Starting tailscaled:  /usr/local/sbin/tailscaled
---

Which is fine, but it got send to the "from address" and not the recipient I configured in settings > notifications
Is there an option I missed somewhere?

[Gex2501]

It appears Tailscale doesn't start after being updated. I had to run this to get it back up.

 

/etc/rc.d/rc.tailscale start

 

[EDACerton]

1 hour ago, mavgink said:

Quick question:
I received an email:
---
Output from your job 32
Taildrop not configured or share not available, disabling
Stopping tailscaled.
Starting tailscaled:  /usr/local/sbin/tailscaled
---

Which is fine, but it got send to the "from address" and not the recipient I configured in settings > notifications
Is there an option I missed somewhere?

The Tailscale plugin doesn’t generate any emails on its own, this must have come from something else. 

[EDACerton]

7 hours ago, Gex2501 said:

It appears Tailscale doesn't start after being updated. I had to run this to get it back up.

 

/etc/rc.d/rc.tailscale start

 

That’s odd. Can you run

 

grep tailscaled /var/log/syslog

 

and send me the results? (PM is fine if you don’t want to post publicly). 

Edited April 4, 2023 by EDACerton

[JudasD]

this is a great idea, love it!  Tailscale still works if the array goes offline, no fancy setups!

 

Thank you for all of your efforts.  Very easy to bring the new server online.  Dead simple.

 

JD

 

[Gex2501]

On 4/3/2023 at 9:46 AM, EDACerton said:

That’s odd. Can you run

 

grep tailscaled /var/log/syslog

 

and send me the results? (PM is fine if you don’t want to post publicly). 

Just sent you my log output via PM. I suppose I should note that I'm working on setting up a new unraid server using 6.12.0-rc2

[EDACerton]

Updated to Tailscale 1.38.4.

[Flashman007]

On 4/3/2023 at 7:44 PM, EDACerton said:

The Tailscale plugin doesn’t generate any emails on its own, this must have come from something else. 

I'm getting the same email at every update: 

 

Taildrop not configured or share not available, disabling

Stopping tailscaled.

Starting tailscaled:  /usr/local/sbin/tailscaled

 

Any ideas?

[EDACerton]

8 hours ago, Flashman007 said:

I'm getting the same email at every update: 

 

Taildrop not configured or share not available, disabling

Stopping tailscaled.

Starting tailscaled:  /usr/local/sbin/tailscaled

 

Any ideas?

The messages are fine, that’s normal behavior if you haven’t configured a Taildrop folder in the plugin settings. 
 

As for the emails — the plugin doesn’t send those, are you using an auto-update plugin?

[elcapitano]

Great Plugin.

Can files be Taildroped FROM unraid?

[EDACerton]

10 minutes ago, elcapitano said:

Great Plugin.

Can files be Taildroped FROM unraid?

In theory, you should be able to use

tailscale file cp <files...> <target>

 

[elcapitano]

1 hour ago, EDACerton said:

In theory, you should be able to use

tailscale file cp <files...> <target>

 

Yeah, I saw that, but I don't seem to be able to get it working . .

[EDACerton]

2 hours ago, elcapitano said:

Yeah, I saw that, but I don't seem to be able to get it working . .

Ok. That’s not something that I really do, let me play around with the feature a bit to see what happens when I try. 

[EDACerton]

3 hours ago, elcapitano said:

Yeah, I saw that, but I don't seem to be able to get it working . .

OK, I just tried it, and it seems to be working fine for me...

 

Do you have a path configured for incoming Taildrops in the plugin options? The plugin disables Taildrop if that isn't set.

[TrojanHorsePower]

Switched from the tailscale docker (which worked fine but wanted to try the plugin version) and everything seemed to be working for a while. But when I tried to access my server from my dorm and noticed I couldn't get to 192.168.1.3 (server IP AT HOME) tried the tailscale 100.X.X.X IP and that also does not work. After a bit of troubleshooting, I discovered the tailscaled service on my tower is down? How can I get it up again without having to reboot the server all the time. If there are logs or stuff, I can send them over to help you fix this bug. Does anyone else have this problem that the tailscaled deamon just stops?

 

 

 

 

[EDACerton]

/etc/rc.d/rc.tailscale restart

Should start the service. 
 

Can you also post diagnostics? I want to see if there’s a reason why it didn’t start for you. 

[TrojanHorsePower]

21 minutes ago, EDACerton said:

/etc/rc.d/rc.tailscale restart

Should start the service. 
 

Can you also post diagnostics? I want to see if there’s a reason why it didn’t start for you. 

 

Worked wonderfully, thank you, can access it easier without having to remote into a different machine. I have added the diagnostics. For now, I shall do a duct tape solution and use the userscript plugin to daily restart the service.

tower-diagnostics-20230418-2016.zip

[EDACerton]

7 hours ago, TrojanHorsePower said:

 

Worked wonderfully, thank you, can access it easier without having to remote into a different machine. I have added the diagnostics. For now, I shall do a duct tape solution and use the userscript plugin to daily restart the service.

tower-diagnostics-20230418-2016.zip 156.8 kB · 0 downloads

OK, I looked at your diagnostics.... the issue seems to be that your server can't resolve the address for the Tailscale control servers when it starts, which causes it to shut down.
 

Is there anything with your setup that could make the array a dependency for internet access (e.g., Pihole running in a docker container)?
 

Spoiler

Quote

Apr 12 03:47:43 Tower tailscaled: 2023/04/12 03:47:27 trying bootstrapDNS("derp9d.tailscale.com", "209.177.156.94") for "controlplane.tailscale.com" ...
Apr 12 03:47:43 Tower tailscaled: 2023/04/12 03:47:29 logtail: dial "log.tailscale.io:443" failed: dial tcp: lookup log.tailscale.io: i/o timeout (in 30.001s), trying bootstrap...
Apr 12 03:47:43 Tower tailscaled: 2023/04/12 03:47:30 bootstrapDNS("derp9d.tailscale.com", "209.177.156.94") for "controlplane.tailscale.com" error: Get "https://derp9d.tailscale.com/bootstrap-dns?q=controlplane.tailscale.com": context deadline exceeded

 

Other things notice this too, so it's not just Tailscale:

Quote

Apr 12 03:45:54 Tower root: Installing dynamix.unraid.net.plg 2023.03.09.1140 with Unraid API 2.57.0
Apr 12 03:45:54 Tower root: plugin: running: anonymous
Apr 12 03:45:54 Tower root: Checking DNS...
Apr 12 03:46:34 Tower root: ⚠️ Warning: Your DNS server (1.1.1.1) is unable to resolve 'mothership.unraid.net'
Apr 12 03:46:34 Tower root:   Recommend navigating to Settings -> Network Settings and changing your DNS server to 8.8.8.8

[TrojanHorsePower]

3 minutes ago, EDACerton said:

OK, I looked at your diagnostics.... the issue seems to be that your server can't resolve the address for the Tailscale control servers when it starts, which causes it to shut down.
 

Is there anything with your setup that could make the array a dependency for internet access (e.g., Pihole running in a docker container)?
 

  Hide contents

 

 

 

 

Other things notice this too, so it's not just Tailscale:

 

 

No the array does not need to be running for internet access, It's just a media server with some self-hosted applications. Weird that the server is having trouble doing DNS lookups, while the other devices on the network never experience this. 

 

  

[Gofer]

I just started having a similar issue. Local access stops entirely, everything still works through the tailscale plugin and the tailnet IP. I disable docker and it works instantly.

 

 

sweetums-diagnostics-20230420-1931.zip

[EDACerton]

1 hour ago, Gofer said:

I just started having a similar issue. Local access stops entirely, everything still works through the tailscale plugin and the tailnet IP. I disable docker and it works instantly.

 

 

sweetums-diagnostics-20230420-1931.zip 273.09 kB · 0 downloads

I think there's something weird with your network config. Looking at your boot logs, it seems like you're pulling a DHCP address from br0... but there's also a static entry in your network.cfg for eth119 with the same IP that's being given to br0 by dhcp:

 

Spoiler

Quote

IFNAME[2]="eth119"

PROTOCOL[2]="ipv4"

USE_DHCP[2]="no"

IPADDR[2]="192.168.5.3"

NETMASK[2]="255.255.255.0"

GATEWAY[2]="192.168.5.1"

 

Quote

Apr 20 07:03:57 Sweetums rc.inet1: polling up to 60 sec for DHCP server on interface br0
Apr 20 07:03:57 Sweetums rc.inet1: timeout 60 dhcpcd -w -q -t 10 -h Sweetums -4 br0
Apr 20 07:03:57 Sweetums  dhcpcd[27690]: dhcpcd-9.4.1 starting
Apr 20 07:03:57 Sweetums  dhcpcd[27693]: DUID 00:04:3f:c3:0f:7c:b1:c9:4a:53:a7:cb:4f:03:cb:38:85:27
Apr 20 07:03:57 Sweetums  dhcpcd[27693]: br0: soliciting a DHCP lease
Apr 20 07:03:57 Sweetums  dhcpcd[27693]: br0: offered 192.168.5.3 from 192.168.5.20
Apr 20 07:03:57 Sweetums  dhcpcd[27693]: br0: probing address 192.168.5.3/24

 

 

I would suggest removing the config for eth119, it seems like that's stale but is confusing things.