NTP sync fails after reboot randomly

[DHLarson]

Still running 384.17 on AC86U for my network router. Having occasional reliability issues with the router failing to time sync after reboot, causing all kinds of havoc. Primary address is pool.ntp.org Just tried adding a hardcoded IP address as a secondary server to a NIST Stratum A destination which seemed to solve it for the moment. I'm considering the benefits of using a list of ip addresses as primary (I assume I can do that with comma delineation) to remove DNS resolution as a possible glitch but wonder if there is something else I'm doing wrong. Thoughts?

 

[dave14305]

Make sure your WAN DNS servers are reliable, and look at the syslog to see when the WAN comes up and when ntpd starts.

I don’t think the Busybox ntp client accepts comma-delimited entries, so assume you can only use 1 IP per field.

Let us know if you have any special DNS setup on your router or any VPN client running. Also report the value on the Tools / Other Settings page “Wan: Use local caching DNS server as system resolver (default: No)”

386.2 made some changes to how ntp deals with DNS timeouts/delays.

 

[DHLarson]

Make sure your WAN DNS servers are reliable, and look at the syslog to see when the WAN comes up and when ntpd starts.

I don’t think the Busybox ntp client accepts comma-delimited entries, so assume you can only use 1 IP per field.

Let us know if you have any special DNS setup on your router or any VPN client running. Also report the value on the Tools / Other Settings page “Wan: Use local caching DNS server as system resolver (default: No)”

386.2 made some changes to how ntp deals with DNS timeouts/delays.

Current WAN DNS servers are prime and secondary Comcast Multicast Servers - could point them toward other servers perhaps. Suggestions?

Using a NordVPN client for a single LAN address with a locked-in DNS server within the pipe - don't think that would apply here since it doesn't initialize until NTP syncs the router.

I do have Cloudflare DNS over TLS primary and secondary servers set up but not sure if that is attempted for use before or after the rooter has fully booted.

Local Caching as System Resolver is set to NO

I don't see any log messages for WAN up or NTPD - do I need to reset logging level which is currently at NOTICE

 

[Jack Yaz]

while this doesn't solve the underlying cause, it can alleviate the symptoms: https://www.snbforums.com/threads/ntp-boot-watchdog-script.71069/

 

[DHLarson]

while this doesn't solve the underlying cause, it can alleviate the symptoms: https://www.snbforums.com/threads/ntp-boot-watchdog-script.71069/

I'll give it a try Jack - thanks!

 

[borris]

I'm having exact same problems with 382.2_4 on ax86u and ax88u. Only 1 connected at a time. Just wanted test a seperate router to see if it was hardware problem. Tried downgrading to 382.1 and still same problems. I haven't sat there rebooting before so have never seen this problem so couldn't say when it started doing this.

I have wireguard, diversion, scribe, ntpmerlin, skynet, dnscrypt running. They work on initial install but after a reboot time is out of sync resulting in no internet.

Any help appreciated

Watchdog script just keeps restarting and doesn't fix the problem.

Also the undiv stats page, skynet, ntpmerlin and scribe gui disappear

 

[borris]

Thinking now it's dnsmasq? Getting skip event restart dnsmasq. Seems to be in sync there abouts maybe couple seconds out randomly sometimes if it syncs up.

That goes away if I disable diversion?