Davidc502- wrt1200ac wrt1900acx wrt3200acm wrt32x builds

I figured I would share my automated script for installing/upgrading dnscrypt 2 after a firmware upgrade.

wget -P /tmp https://davidc502sis.dynamic-dns.net/releases/dnscrypt-proxy.tar.gz ; cd /tmp ; gunzip -d /tmp/dnscrypt-proxy.tar.gz ; tar xvf /tmp/dnscrypt-proxy.tar ; rm -f /tmp/dnscrypt-proxy.tar
cp /tmp/dnscrypt-proxy/dnscrypt-proxy /usr/sbin/ ; chmod 755 /usr/sbin/dnscrypt-proxy ; cp /tmp/dnscrypt-proxy/dnscrypt-proxy.toml /etc/config/ ; cp /tmp/dnscrypt-proxy/init.d/dnscrypt-proxy \
/etc/init.d/ ; chmod 755 /etc/init.d/dnscrypt-proxy ; /etc/init.d/dnscrypt-proxy enable ; /etc/init.d/dnscrypt-proxy start ; rm -rf /tmp/dnscrypt-proxy
1 Like

Hi WiteWulf, can i ask, did you perform the sysupgrade via the Luci GUI with the 'keep settings' ticked or did you do it via putty sysupgrade. I have my router set up with linksys firmware on partition1 and LEDE on partition2 and want to make sure any sysupgrade only affects partition2 so i can always go back to linksys firmware if i need to. I'm also interested in whether all settings are kept across davidc502 upgrades. thanks.

You're very welcome to actually share your findings. I will not look at post #182 in a thread called "Davidc502- wrt1200ac wrt1900acx wrt3200acm wrt32x builds" to check for LuCI issues. Someone bothered to report the VLAN bug in the tracker and I fixed it within 40 minutes.

3 Likes

Understood. It was my assumption someone would have reported the issue, so I'm glad you happened across the post, and now have it fixed.

There's another LuCi-base issue somewhere else with a similar symptom (someone posted). I'll see if I can find it.

EDIT

Take a look at this post as there may be some validity.. I have not confirmed.

Same here, i always reboot to my first partition (Linksys) and flash davids factory image from there and configure everything from the scratch.
My config/setup isn't so complicated so i setup all the interfaces and wifi and than copy most config files from my old etc/config folder (SQM, Firewall, adblock, dhcp aso) to save some time. Always worked great for me but as i said my setup is quite simple...

@davidc502
When can we expect a new build with the fixed/updated luci-base? :slight_smile:
I'm still using r7210 because of the Luci problems...

Currently, it is scheduled for August 4th-ish.

I've read the quoted post three times in a row now and I do not understand it. The information contradicts itself:

"Update again - it seems it gets applied but quickly reverts to 192.168.1.1. If I hit 'save and apply', I can ping the new IP for a few seconds, then it drops off and is back to the original default IP."

and

"I click 'save and apply' in the GUI and find it's not staying at the default IP of 192.168.1.1."

So is it is unable to change 192.168.1.1 or unable to stay on 192.168.1.1? I can't make sense out of that report.

sysupgrade via Luci with keep settings ticked, simple as that.

After a little more poking the only changes I noticed were that some services I'd disabled had been re-enabled (dnscrypt, adblocker, samba and openvpn). That was simple enough to fix.

1 Like

I had the same problem with my DIR860L @18.06 Snapshot Build (im using this device as a smart switch without DHCP).
So i did the following to get the IP change working via Luci... change the IPv4 address and deactivate (ignore) DHCP server and click Save & Apply, close browser (quickly), deactivate the network card and manuell set the new gateway IP, then activate the network card/wifi adapter again and connect via ssh to the new gateway IP and type reboot.
After reboot you can can enter Luci again (with new IP), activate the DHCP server and change the network card/wifi adapter to "Obtain IP address/DNS server address automatically".
You can also change the IP via network config if this way does not work for you...:wink:

Why do you need to quickly close the browser?

Otherwise the IP change is getting reverted...
There is also a text in Luci "Changes invalid" or something like this. I don't remember exactly.

Did you try to just hit the "Apply unchecked" button? That will commit the config change regardless of whether access to the device is lost or not.

I wonder if renaming Apply unchecked to Apply anyway would make the situation clearer? Apparently it is not obvious that the rollback can be overruled by hitting that red button.

I plan to add some "Apply straight away" mode (which will be some kind of dropdown choice in the Save & Apply button) later which will skip the 30s rollback for those people who really know what they do but before I do that I want to figure out the "intuitiveness" issues first.)

2 Likes

If i remember right i clicked this box one time but nothing happened and after i deactivate and reactivate my network the old IP was still present.
All i can say is that i don't had this problem with an older 17.01 Snapshot build.

I plan to add some "Apply straight away" mode (which will be some kind of dropdown choice in the Save & Apply button) later which will skip the 30s rollback for those people who really know what they do but before I do that I want to figure out the "intuitiveness" issues first.)

+1
Imho that sounds like a good Idea!

At least something should have happened, either apply-again
... leading to ...
apply-gone
... or some kind of error.

Do you literally mean "nothing happened" like the box not going away / clicking the button has no effect?

But thanks anyway for the insight. I am afraid I will need to wait for some more detailed reports to figure out what is really going on.

The 17.01 releases didn't have this mechanism.

Im not 100% sure what happened when i clicked the box but i know for sure that my config change (IP) went through but still got reverted after a short period of time (~10-15secs).
It could also be related to my browser as i'm using ungoogled Chromium for all my Web GUI's/Router Devices.

I have the following issue on my WRT1200 with the latest firmware. I got a new laptop which is my first unit, which is able to do at least 866mbit over 5Ghz, my older laptops just all could do 300mbit. It has an Intel 9260 card. The issue now is, that the moment I send files from it to the network, the file copy process breaks down after 1-2 seconds and goes to 0 kb/s and idle out after a long time and give a network error. Doing a ping while this happens, the ping to the router suddenly all goes up to about 133ms. Downspeed from the LAN to the laptops works normally of about 33MB/s (which is also though not close to half of 866mbit, it should be in theroy for half duplex be around at least 420mbit). Windows shows 866mbit and 4 green bars to the WRT1200. Config for the 5Ghz is option htmode 'VHT80'. Setting it to HT40, and the upload works stable for the 300mbit. Anyone else have similar issues or know what could cause it?

Appears there is an issue with some intel cards.

1 Like

@anomeome :frowning:

Tried using your script. All good except last command gives "Permission denied" for me over ssh as root...