Il 18/11/18 21:14, Patrick Shanahan ha scritto:
* Marco Calistri
[11-18-18 16:24]: Il 18/11/18 16:56, Carlos E. R. ha scritto:
On 18/11/2018 14.51, Marco Calistri wrote:
I was relying on ntpd service since ever until I read about "chronyd" as the substitute of ntpd adopted by new opensuse releases, then I switched to chronyd.
Despite this it seems ntp package being still necessary (?) or could be removed by the system?
To verify if my system is synchronized I use the command "timedatectl":
marco@linux-turion64:~> timedatectl Local time: dom 2018-11-18 11:40:48 -02 Universal time: dom 2018-11-18 13:40:48 UTC RTC time: dom 2018-11-18 11:40:48 Time zone: America/Sao_Paulo (-02, -0200) System clock synchronized: yes NTP service: inactive RTC in local TZ: yes
Interesting command! I did not know about it. But it would be rather more interesting if it displayed milliseconds, as it does it is not sufficient.
It also, now that I think, does not display the network time at all, in my case from ntp, so I do not know how well my clock is synced - or not.
Hello Carlos,
My Linux box is not a 24/24 server then I decided to switch to chronyd: https://www.thegeekdiary.com/centos-rhel-7-chrony-vs-ntp-differences-between...
being a server is not a requirement for ntpd vs chronyd, it is a choice. one may choose either for any connected box providing any service.
I was referring specifically to the fact that my Linux computer is not connected to Internet all day long: https://chrony.tuxfamily.org/faq.html#_how_does_code_chrony_code_compare_to_...
Nov 18 18:05:22 wahoo systemd[1]: Starting NTP Server Daemon... Nov 18 18:05:22 wahoo ntpd[20983]: ntpd 4.2.8p10@1.3728-o Fri May 26 14:07:29 UTC 2017 (1): Starting Nov 18 18:05:22 wahoo ntpd[20983]: Command line: /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -c /etc/ntp.conf Nov 18 18:05:22 wahoo ntpd[20984]: proto: precision = 0.084 usec (-23) Nov 18 18:05:22 wahoo ntpd[20984]: switching logging to file /var/log/ntp Nov 18 18:05:22 wahoo start-ntpd[20973]: Starting network time protocol daemon (NTPD) Nov 18 18:05:22 wahoo systemd[1]: Started NTP Server Daemon. Hint: Some lines were ellipsized, use -l to show in full. Local time: Sun 2018-11-18 18:11:15 EST Universal time: Sun 2018-11-18 23:11:15 UTC RTC time: Sun 2018-11-18 23:11:15 Time zone: America/Indiana/Indianapolis (EST, -0500) Network time on: yes NTP synchronized: yes RTC in local TZ: no ntp_gettime() returns code 0 (OK) time df9c7013.f22003ac Sun, Nov 18 2018 18:11:15.945, (.945801052), maximum error 169215 us, estimated error 1875 us, TAI offset 0 ntp_adjtime() returns code 0 (OK) modes 0x0 (), offset -836.944 us, frequency -8.556 ppm, interval 1 s, maximum error 169215 us, estimated error 1875 us, status 0x2001 (PLL,NANO), time constant 6, precision 0.001 us, tolerance 500 ppm,
in this particular case the server happens to be running ntpd but could just as well be running chronyd. the time precision reported is the same.
I personally dislike using two daemons doing practically the same job, if I can just use one. Regards, -- Marco Calistri Build: openSUSE Tumbleweed 20181116 Kernel: 4.19.1-1-default - Cinnamon 3.8.9