NTP Not synced correctly

K

koops

New Member
Nov 25, 2025
13
0
1
NTP doesn't sync correctly with only one of my computers in my cluster, and I was wondering how I could change the NTP time or maybe sync it correctly? Here is the status of it, its about 5 hours off.
Yes, my system is called butt.
● chrony.service - chrony, an NTP client/server
Loaded: loaded (/usr/lib/systemd/system/chrony.service; enabled; preset: enabled)
Active: active (running) since Tue 2025-11-25 03:16:25 EST; 14min ago
Invocation: d2a434615cfb4edab7d501a059c95a94
Docs: man:chronyd(8)
man:chronyc(1)
man:chrony.conf(5)
Process: 694 ExecStart=/usr/sbin/chronyd $DAEMON_OPTS (code=exited, status=0/SUCCESS)
Main PID: 776 (chronyd)
Tasks: 2 (limit: 9192)
Memory: 5.1M (peak: 5.9M)
CPU: 37ms
CGroup: /system.slice/chrony.service
├─776 /usr/sbin/chronyd -F 1
└─777 /usr/sbin/chronyd -F 1

Nov 25 03:16:24 butt systemd[1]: Starting chrony.service - chrony, an NTP client/server...
Nov 25 03:16:25 butt chronyd[776]: chronyd version 4.6.1 starting (+CMDMON +NTP +REFCLOCK +RTC +PRIVDROP +SCFILTER +SIGND +ASYNCDNS +NTS +SECHASH +IPV6 -DEBU>
Nov 25 03:16:25 butt chronyd[776]: Loaded 0 symmetric keys
Nov 25 03:16:25 butt chronyd[776]: Using leap second list /usr/share/zoneinfo/leap-seconds.list
Nov 25 03:16:25 butt chronyd[776]: Loaded seccomp filter (level 1)
Nov 25 03:16:25 butt systemd[1]: Started chrony.service - chrony, an NTP client/server.
~
 
BobhWasatch said:
What is the output of "chronyc sources", "timedatectl", "ip a" and "ip route"? Do you have the CMOS clock set to local time or UTC (preferred)?
Click to expand...
In order,
1.
MS Name/IP address Stratum Poll Reach LastRx Last sample
===============================================================================
2.
Local time: Tue 2025-11-25 04:37:56 EST
Universal time: Tue 2025-11-25 09:37:56 UTC
RTC time: Tue 2025-11-25 09:37:57
Time zone: America/New_York (EST, -0500)
System clock synchronized: no
NTP service: active
RTC in local TZ: no
3.
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host noprefixroute
valid_lft forever preferred_lft forever
2: nic0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master vmbr0 state UP group default qlen 1000
link/ether a4:1f:72:8c:e0:a6 brd ff:ff:ff:ff:ff:ff
altname enxa41f728ce0a6
3: vmbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether a4:1f:72:8c:e0:a6 brd ff:ff:ff:ff:ff:ff
inet 192.168.5.10/16 scope global vmbr0
valid_lft forever preferred_lft forever
inet6 fe80::a61f:72ff:fe8c:e0a6/64 scope link proto kernel_ll
valid_lft forever preferred_lft forever
4.
default via 192.168.5.189 dev vmbr0 proto kernel onlink
192.168.0.0/16 dev vmbr0 proto kernel scope link src 192.168.5.10
5. Local time.
 
BobhWasatch said:
It would seem that:
  1. Your CMOS clock is way off (that's where the system gets its initial time).
  2. You have no servers or pools listed in /etc/chrony/chrony.conf, or else your DNS is not able to resolve them.
Click to expand...
I might've been wrong about where my CMOS gets the time from, because in BIOS its correct, but in the server its not, and for my other computers in the cluster it has the same chrony config and is correct, it might be my DNS, ill check.
 
I am sorry for being a little dense, its my first time ever using proxmox, let alone linux in a server kind of way. the local time on the server specifically is messed up for some reason, not on the cmos, and I checked all the configs on the other computers and it was all the same, it just wanted to be different for some reason, is there anyway to change the local clock by 5 hours? or add different NTP servers for the one that doesnt work?
 
When it first boots the system will set the time from the BIOS clock (aka CMOS). That clock should be set to UTC, NOT to your local time. That would explain the 5 hours as you are in GMT-5. The other systems probably have the same error in their BIOS but you don't notice because NTP corrects it.

If you have a line like pool 2.pool.ntp.org in /etc/chrony/chrony.conf, try pinging it. If your DNS is not working, you'll get some error like "temporary failure in name resolution", in which case you need to check /etc/resolv.conf. Chaning to a different NTP server won't fix anything if DNS is broken unless you put in an IP address directly, which you should only do if you have a local NTP server.

If you do not have a "pool" or "server" line in /etc/chrony.conf then you need to add one. The first few lines on this machine where I type (not my PVE but plain Debian 13) are:

Code: 
# Welcome to the chrony configuration file. See chrony.conf(5) for more
# information about usable directives.

# Use Debian vendor zone.
pool 2.debian.pool.ntp.org iburst

# Use time sources from DHCP.
sourcedir /run/chrony-dhcp

# Use NTP sources found in /etc/chrony/sources.d.
sourcedir /etc/chrony/sources.d

ETA: You can use the "date" command to set the time but it won't survive a reboot. It needs to be fixed properly.
 
BobhWasatch said:
When it first boots the system will set the time from the BIOS clock (aka CMOS). That clock should be set to UTC, NOT to your local time. That would explain the 5 hours as you are in GMT-5. The other systems probably have the same error in their BIOS but you don't notice because NTP corrects it.

If you have a line like pool 2.pool.ntp.org in /etc/chrony/chrony.conf, try pinging it. If your DNS is not working, you'll get some error like "temporary failure in name resolution", in which case you need to check /etc/resolv.conf. Chaning to a different NTP server won't fix anything if DNS is broken unless you put in an IP address directly, which you should only do if you have a local NTP server.

If you do not have a "pool" or "server" line in /etc/chrony.conf then you need to add one. The first few lines on this machine where I type (not my PVE but plain Debian 13) are:

Code: 
# Welcome to the chrony configuration file. See chrony.conf(5) for more
# information about usable directives.

# Use Debian vendor zone.
pool 2.debian.pool.ntp.org iburst

# Use time sources from DHCP.
sourcedir /run/chrony-dhcp

# Use NTP sources found in /etc/chrony/sources.d.
sourcedir /etc/chrony/sources.d

ETA: You can use the "date" command to set the time but it won't survive a reboot. It needs to be fixed properly.
Click to expand...
okay, it was the DNS server, im just confused on how it didnt work on this pc but not the others. I was set to UTC for the BIOS settings. After setting the DNS to cloudflare or google then what?
 
Code: 
Local time: Tue 2025-11-25 04:37:56 EST
Universal time: Tue 2025-11-25 09:37:56 UTC
RTC time: Tue 2025-11-25 09:37:57
Time zone: America/New_York (EST, -0500)
System clock synchronized: no
NTP service: active
RTC in local TZ: no

This says that your time zone is EST (America/New_York). Is that correct? Notice that the offset is -5 hours.
 
BobhWasatch said:
Code: 
Local time: Tue 2025-11-25 04:37:56 EST
Universal time: Tue 2025-11-25 09:37:56 UTC
RTC time: Tue 2025-11-25 09:37:57
Time zone: America/New_York (EST, -0500)
System clock synchronized: no
NTP service: active
RTC in local TZ: no

This says that your time zone is EST (America/New_York). Is that correct? Notice that the offset is -5 hours.
Click to expand...
Yes my timezone is in new york, should i change it to something thats the same as the UTC?
 
If New York is correct please leave it alone. So we are down to either you don't have a proper server in chrony.conf or your network isn't working right:
  • DNS can't find the server (you claim this is fixed)
  • Chrony has no server listed (did you check this?)
  • Chrony can't reach the server (can you ping it?)
  • Chrony fails to sync (output of "chronyc sources" will show this)
  • There is perhaps a firewall blocking UDP port 123 somewhere between your machine and the NTP server.
 
The DNS is still not being able to resolve the address, i checked the Chrony server list, there was a server there, Chrony doesn't sync still, maybe there is a firewall but theres no firewall enabled on my proxmox, switch, or router back to my isp.
 
If DNS isn't working then chrony won't be able to get the IP address of the server. Can you ping the DNS server that you put in /etc/resolv.conf? Can you ping google.com? Can you ping 8.8.8.8?

Your problem is DNS and/or your network settings, not NTP.
 
What are you talking about? An override of what IP's? According to what you posted earlier you have vmbr0 at 192.168.5.10/16 with gateway 192.168.5.189. This is, um, unusual in two ways, but it is legal.

One unusual thing is that 192.168 is usually a /24, not /16, the other is that the gateway is usually either the first or last IP in the block, not some random IP in the middle (it is normally the router for the subnet). Maybe compare with the working machines and make sure you have a correct gateway and network size.
 
BobhWasatch said:
What are you talking about? An override of what IP's? According to what you posted earlier you have vmbr0 at 192.168.5.10/16 with gateway 192.168.5.189. This is, um, unusual in two ways, but it is legal.

One unusual thing is that 192.168 is usually a /24, not /16, the other is that the gateway is usually either the first or last IP in the block, not some random IP in the middle (it is normally the router for the subnet). Maybe compare with the working machines and make sure you have a correct gateway and network size.
Click to expand...
Thats just how my network was setup for some strange reason, I have no clue for why, but I mean maybe a weird thing with the ISP, I checked with the other machines and they work perfect, there is no complaints with them.
 
You must log in or register to reply here.
Top Bottom