Known Issues in 1.6.0 Release

With ~25 of my devices on the one network, a burst like that can be a lot of traffic.
If it is only a burst, then base don my experience, I would expect that it will repeat at some interval.

It is 100% certain that ZT was what killed my network. I’m turning the service back on for a couple of remote users now. WIll keep an eye on things.

Yeah it seems to be a burst on startup. Smaller burst on network join.

Not just start up. This device has been on line for 20 minutes now. Not being used at all.
image

Then a roughly a minute later
image

Sunday is when it was released to chocolatey. Look at my network history. The graph only updates every 8 hours on the month view. so it will not show normal for a while now that I have shut down the service globally.

Ok @jbusch. Looks like there’s a feedback loop triggered sometimes on Windows. Working on a fix. Will be in the 1.6.1 windows release.

2 Likes

Thank you. In the mean time, I have used chocolatey to go back to 1.4.6 and pinned the version so they do not upgrade again.

I’m having a new issue with 1.6.1 on Ubuntu.

It seems to be working at first but then it hangs when joining or leaving numerous networks.

sudo zerotier-cli listnetworks
Error connecting to the ZeroTier service: timed out

The REST backend in the daemon is single threaded so if it’s still processing a request/many requests that’s possible. Don’t do a bunch at once

Yeah, i’ve tested 1.6.0 on 3 Win10 machines now and its def hosed.

You guys forgot to sign ZTTAP300 i believe. Can post anything you need for help.

No it’s signed. It’s the same exact same copy of the driver & signature going back to 2015 and should still be grandfathered in under Windows’ driver signing rules. Something seems to have changed in Windows driver signing verification. It’ll load at startup, but only a certain number of times, then Windows doesn’t see the signature as valid anymore. Rebooting fixes it temporarily.

We’re jumping through the Seven Circles of Hell hoops of the new Windows driver signing process. Works internally on a test we did. Updated signature will be in the 1.6.1 Windows release.

Well in my case 1.6.1 always does that and 1.4.6 never does that, so it doesn’t really sound right.

I expect nothing less from Microsoft my friend, its a crap-shoot every update.

Downgrading the version got me back rolling, looking forward to DNS function baked in vs the script i’ve made work.

Thanks for the response.

1 Like

zerotier-cli bond commands are missing / non functional … compiled on both linux and FreeBSD, same issue both sides… only documentation on multipath is https://zerotier.atlassian.net/wiki/spaces/SD/pages/568459265/Multipath … very little information anywhere else on it… so I have to assume it is the correct info…

root@OPNsense:~ # zerotier-cli bond enable
ZeroTier One version 1.6.0 build 0 (platform 7 arch 2)
Copyright © 2020 ZeroTier, Inc.
Licensed under the ZeroTier BSL 1.1 (see LICENSE.txt)
Usage: zerotier-cli [-switches] <command/path> []

am I missing something fundamental on bonding here?

the bonding CLI commands didn’t make it into 1.6.0. Coming on 1.6.1

is there any pre-release?, i’m happy to test … it’s my primary reason for Zerotier in the first place :smiley:

Grant, are you able to PM me? i found the issue.

We’ve already resolved the issue internally. Trying to track down a macOS bug and then new Windows and macOS binaries will be up.

It looks like the .inf was just set with the wrong encoding? But alright, i took the INF from 1.4.6 as a workaround.

Interesting… what was it set to? I’ll have to double check the installer to make sure it’s not changing that somehow.

EDIT: It’s showing UTF-8 everywhere for me. 1.4.6, 1.6.0, internal 1.6.1, so without more info I’m not sure that’s it.

the latest 16.1 build OSX still doesn’t pick up IP or route. Futher details 1.6.0 no ips assigned on macos 10.13

There seems to be something different about how ZT learns about internal (behind NAT) nodes in 1.6.0. My FreeBSD nodes are learning lots of external routes but nothing locally behind the NAT, which used to happen without any interference. I don’t mind adding a local.conf file to hint it, so long as encryption isn’t deactivated.

Is this a known issue?