Strange. Same errors. I’m going to manually run each command and see what happens.
I just looked and the file on the server is updated.
Lines 98 to 100
$SUDO sed -i -e ‘s/ssid=ArchiveSW/ssid=’$accessPoint’/’ /etc/hostapd/hostapd.txt
$SUDO sed -i -e ‘s/channel=11/channel=’$accessChannel’/’ /etc/hostapd/hostapd.txt
$SUDO sed -i -e ‘s/wpa_passphrase=WeatherFlow/wpa_passphrase=’$accessPass’/’ /etc/hostapd/hostapd.txt
That is what I see in the script. Here is the result of manually executing those lines.
pi@raspberrypi:~ $ cd /etc/hostapd
pi@raspberrypi:/etc/hostapd $ sudo wget -N http://fsoft.com/archivesw/script/hostapd.txt
--2019-03-08 21:33:11-- http://fsoft.com/archivesw/script/hostapd.txt
Resolving fsoft.com (fsoft.com)... 162.253.126.169
Connecting to fsoft.com (fsoft.com)|162.253.126.169|:80... connected.
HTTP request sent, awaiting response... 301 Moved Permanently
Location: https://fsoft.com/archivesw/script/hostapd.txt [following]
--2019-03-08 21:33:11-- https://fsoft.com/archivesw/script/hostapd.txt
Connecting to fsoft.com (fsoft.com)|162.253.126.169|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 221 [text/plain]
Saving to: ‘hostapd.txt’
hostapd.txt 100%[==================================>] 221 --.-KB/s in 0s
2019-03-08 21:33:11 (86.1 MB/s) - ‘hostapd.txt’ saved [221/221]
pi@raspberrypi:/etc/hostapd $ sudo sed -i -e 's/ssid=ArchiveSW/ssid='$accessPoint'/' /etc/hostapd/hostapd.txtpi@raspberrypi:/etc/hostapd $ sudo sed -i -e 's/channel=11/channel='$accessChannel'/' /etc/hostapd/hostapd.txt
pi@raspberrypi:/etc/hostapd $ sudo sed -i -e 's/wpa_passphrase=WeatherFlow/wpa_passphrase='$accessPass'/' /etc/hostapd/hostapd.txtpi@raspberrypi:/etc/hostapd $ sudo mv hostapd.txt hostapd.conf
pi@raspberrypi:/etc/hostapd $ ls
hostapd.conf ifupdown.sh
pi@raspberrypi:/etc/hostapd $
EDIT: I still get an error on starting hostapd.
pi@raspberrypi:/etc/hostapd $ sudo systemctl start hostapd
Job for hostapd.service failed because the control process exited with error code.
See "systemctl status hostapd.service" and "journalctl -xe" for details.
pi@raspberrypi:/etc/hostapd $
Did you put those lines in manually or from the script? If manually, you need to substitute the proper variables.
Also, you have two commands on the first line.
I missed the variables. I’ll give it another go.
There are two other steps that have to be done before you can do anything with hostapd.
As long as hostapd.txt is in the folder, the script should run.
Other than the name resolution issue it worked that time with no other errors. I powered off my mobile RPi and the mobile hub never even gave me the red light.
Now to get ArchiveSW back up and running. I lost access to it on my Mobile unit and was able to get the Access Point working so I didn’t have to reconfigure the WF hub. Should I be able to just run the ArchiveSW install script again to get it back or is there something else I should look for first?
You should be good to go with running the install script. It should not do any damage to the database and everything should install normally.
Also, please run this script and send me the output. It will become part of the accesspoint script. It outputs your interface data.
bash -e <(wget -qO - http://fsoft.com/archivesw/script/test.sh)
I ran the install script on my mobile unit and still get a “refused to connect” error. Not sure what happened or what is not running. I can ping it, ssh into it, and VNC to it but the web server isn’t responding from localhost or externally.
I’ll run the test.sh script on both the test unit and mobile unit tomorrow. I need to hit the sack.
Thanks for your help. We will get those pesky bugs worked out.
Hankster,
I wanted to get back to you on these errors. They are all from chromium-browser. The current version is 72, however the files for the RPi have not been updated as the current version of Chromium for the RPi is 65.0.3325.181.
Hopefully we will see an update this year or real soon now as the industry goes with small teams.
The matchbox-window-manager error has been corrected.
Reinstalled over my old install and everything went fine and it is now working correctly. Thanks @GaryFunk for all of your hard work!
It looks good. I’m pleased to know you have it working.
Starting dnsmasq causes all connections to be reset. That’s what the script seems to stop. I have moved that to the end of the script and added a 15 second delay before rebooting the RPi.
Writing these installation scripts is almost as involved as writing the main applications.
Here is the interface test script run on my Mobile RPi. I got the access point running manually and have rerun the ArchiveSW install script again but I still cannot connect to the Panel web interface. I keep getting a rejected error in the browser, both locally and externally.
pi@Mobile_ArchiveSW:~ $ bash -e <(wget -qO - http://fsoft.com/archivesw/script/test.sh)
3 network interfaces
wlan0 = 10.0.0.124
dynamic = 2601:1c1:4200:8d00:12f9:2910:9268:7d3c
wlan1 = 192.168.4.1
2 network interfaces
wlan0 = 10.0.0.124
wlan1 = 192.168.4.1
pi@Mobile_ArchiveSW:~ $
On the test RPi I get this:
pi@raspberrypi:~ $ bash -e <(wget -qO - http://fsoft.com/archivesw/script/test.sh)
3 network interfaces
wlan0 = 192.168.4.1
wlan1 = 10.0.0.120
dynamic = 2601:1c1:4200:8d00:b76f:fc38:796c:11ec
2 network interfaces
wlan1 = 10.0.0.120
wlan0 = 192.168.4.1
pi@raspberrypi:~ $
Both are RPi 3B+ and have different USB WiFi dongles but each uses the same Realtech chip.
EDIT: On the Mobile unit I have settings to use the onboard WiFi as the uplink on wlan0 and on the test unit I didn’t specify. It turns out that the onboard WiFi gets wlan1 if I don’t specify any different.
I also get a similar error on both devices as @hankster did.
--disable-quic --enable-tcp-fast-open --disable-gpu-compositing --ppapi-flash-path=/usr/lib/chromium-browser/libpepflashplayer.so --ppapi-flash-args=enable_stagevideo_auto=0 --ppapi-flash-version=
Fontconfig warning: "/etc/fonts/fonts.conf", line 160: blank doesn't take any effect anymore. please remove it from your fonts.conf
[1302:1302:0308/223617.105883:ERROR:gpu_process_transport_factory.cc(1029)] Lost UI shared context.
[1302:1341:0308/223618.207161:ERROR:in_progress_cache_impl.cc(93)] Could not read download entries from file because there was a read failure.
[1302:1417:0308/223618.223194:ERROR:bus.cc(394)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
ATTENTION: default value of option force_s3tc_enable overridden by environment.
[1426:1426:0308/223620.133160:ERROR:sandbox_linux.cc(375)] InitializeSandbox() called with multiple threads in process gpu-process.
pi@raspberrypi:~$
This pops up in a terminal window right after boot up.
That’s very interesting. You are using IPv6 on your local network and the script catches that.
The errors are from the chromium-browser because it hasn’t been updated. We just have to live with that.
Thank you for running the test.
OK, noob question. What address do I need to use to access the panel using a computer on the same network?
for the panel the same address as the server configuration except the port is 8088
@GaryFunk Is there a reason that ArchiveSW is showing different rain numbers then the WF app is? Today’s rain totals are showing 1.120" for ArchiveSW vs .62" for the app/web site.
Maybe because WeatherFlow is adjusting the data. You would have to ask David if your data is being adjusted after it has been sent to the servers.