<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
On 4/10/19 14:03, <a class="moz-txt-link-abbreviated" href="mailto:dropkick23@gmail.com">dropkick23@gmail.com</a> wrote:<br>
<blockquote type="cite"
cite="mid:CAHAtVgp8bUGK10cV0nEOcuExwd--3kefomDyFfzz68Ug8Dv-8A@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">
<div dir="ltr">On Sat, 6 Apr 2019 16:21:50 -0700<br>
Paul Theodoropoulos <<a
href="mailto:paul@anastrophe.com" target="_blank"
moz-do-not-send="true">paul@anastrophe.com</a>>
wrote:<br>
<br>
> gpsd needs to start before ntpd when using shm
refclock.<br>
<div><br>
</div>
<div>That is completely opposite from my experience. I
have two systemd service files one to start gpsd and
one to start ntpd. the ntpd service file is from
ntpsec distribution. gpsd service file is from gpsd
distribution and edited to remove reference to
gpsd.socket and change After=chronyd.service to
ntpd.service.</div>
<br>
<div>New system fresh install of Arch Linux ARM, GPSD
and NTPD.</div>
<div>System is raspi 3+</div>
<div>Linux tock2 4.19.34-1-ARCH #1 SMP PREEMPT Wed Apr
10 00:14:00 UTC 2019 armv7l GNU/Linux<br>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
<br>
I guess I'd assume then that it's a difference between Raspbian and
Arch - unless maybe it's something in the /boot/cmdline.txt or
config.txt hardware setup?<br>
<br>
root@RaspiPi3B+ NTPsec: ~ # cat /boot/config.txt<br>
#dtparam=sd_overclock=100<br>
dtoverlay=pps-gpio,gpiopin=4<br>
dtoverlay=pi3-disable-bt<br>
#dtoverlay=pi3-miniuart-bt<br>
#dtparam=pwr_led_trigger=none<br>
gpu_mem=32<br>
enable_uart=1<br>
init_uart_baud=115200<br>
boot_delay=0<br>
disable_splash=1<br>
temp_limit=70<br>
disable_pvt=1<br>
<br>
<br>
root@RaspiPi3B+ NTPsec: ~ # cat /boot/cmdline.txt<br>
root=/dev/mmcblk0p2 rootfstype=f2fs elevator=deadline
fsck.repair=yes rootwait nohz=off ipv6.disable=1
dwc_otg.lpm_enable=0 dwc_otg.fiq_fsm_enable=0 dwc_otg.fiq_enable=0
dwc_otg.nak_holdoff_enable=0<br>
<br>
But none of that looks particularly like it'd affect device
permissions or such.<br>
<pre class="moz-signature" cols="72">--
Paul Theodoropoulos
<a class="moz-txt-link-abbreviated" href="http://www.anastrophe.com">www.anastrophe.com</a></pre>
</body>
</html>