antonsamoziv wrote:

I really advise everyone to just have the ovpn conf file(from your vpn provider) at /etc/openvpn and have you /etc/config/openvpn just read from the config. If options get deprecated in the future your vpn provider will update the client configs.

config openvpn '[ ]'

    option enabled 1
    option config /etc/openvpn/[ ].conf

i think it's a good idea.  the ovpn conf file from my vpn provider has a file extension ending with ovpn.  does it mean i simply change the file extension from .conf to .ovpn as shown?  the ovpn file from my vpn provider contains my ca, cert, and key.

config openvpn '[ ]'

    option enabled 1
    option config /etc/openvpn/[ ].ovpn

starcms wrote:

NamesAreAPain wrote:

Speaking of building your own image and WiFi, could you post your version of mac80211.c.
I have tried to both mod the driver myself and tried disabling it at start up via /etc/kernel/debug/ieee80211/phy0(or phy1)/mwlwifi/tx_amsdu but for some reason I still get the spikes.

To disable it at startup (if you don't disable it in the driver or use @david's builds), the commands are

echo 0 > /sys/kernel/debug/ieee80211/phy0/mwlwifi/tx_amsdu
echo 0 > /sys/kernel/debug/ieee80211/phy1/mwlwifi/tx_amsdu

Yeah, I tried that (doing that in an init script just after networking, ignore the /etc, I meant /sys) but I still got ping spikes (when pinging my PC from the router) on my 4.14 build.

davidc502 wrote:

NamesAreAPain wrote:

Speaking of building your own image and WiFi, could you post your version of mac80211.c.
I have tried to both mod the driver myself and tried disabling it at start up via /etc/kernel/debug/ieee80211/phy0(or phy1)/mwlwifi/tx_amsdu but for some reason I still get the spikes.

Line 781 just set amsdu to false.

Line 781 is an if statement

if (priv->tx_amsdu)

Line 782 looks more likely

sta_info->is_amsdu_allowed =

but that is the same place I modded in my 4.9 build that did not seem to work.

NamesAreAPain wrote:

starcms wrote:

To disable it at startup (if you don't disable it in the driver or use @david's builds), the commands are

echo 0 > /sys/kernel/debug/ieee80211/phy0/mwlwifi/tx_amsdu
echo 0 > /sys/kernel/debug/ieee80211/phy1/mwlwifi/tx_amsdu

Yeah, I tried that (doing that in an init script just after networking, ignore the /etc, I meant /sys) but I still got ping spikes (when pinging my PC from the router) on my 4.14 build.

davidc502 wrote:

Line 781 just set amsdu to false.

Line 781 is an if statement

if (priv->tx_amsdu)

Line 782 looks more likely

sta_info->is_amsdu_allowed =

but that is the same place I modded in my 4.9 build that did not seem to work.

That is where I change it as well (set to false).  Have you been able to confirm the build is using your modded wifi driver and not something pulled down from github?

Also, how many clients are connected when getting these spikes? Are you getting much interference where you are?

Line 781 is an if statement

if (priv->tx_amsdu)

Line 782 looks more likely

sta_info->is_amsdu_allowed =

but that is the same place I modded in my 4.9 build that did not seem to work.

EDIT: Nope, still having problems. Ping spikes to 20-50ms every minute or so on 2.4 and massive pings in the 100s range on 5Ghz (though I can only test with my phone and steamlink).

Once I set US and 23 db for power... boom! Permanent Disabled Connection.  Will not come to life, even if i reset and reboot.

....

There must be something linksys has done to the latest batch of refurbished wrt1900ac/v2 units. Don't want to use outdated firmware with security issues. I have a feeling the new kernel does not like the recent certified refurbished ac/v2 hardware.

All is working great but is anyone else having issues with Chromecast ver.2 not working? Keep getting reconnect me messages "Chromecast is connected to blah but can't access the Internet"

Was working fine with a recent march bryanslayer build last week.

Can't seem to figure out whats wrong?

Notsofast wrote:

Once I set US and 23 db for power... boom! Permanent Disabled Connection.  Will not come to life, even if i reset and reboot.

....

There must be something linksys has done to the latest batch of refurbished wrt1900ac/v2 units. Don't want to use outdated firmware with security issues. I have a feeling the new kernel does not like the recent certified refurbished ac/v2 hardware.

If you have a 1900ACSv2, 1200ACSv2, or 3200ACM, the power tables are read from EEPROM and not the firmware.  There is no way to change them.

Hence, changing the country code will result in nothing but problems...

Cantenna wrote:

All is working great but is anyone else having issues with Chromecast ver.2 not working? Keep getting reconnect me messages "Chromecast is connected to blah but can't access the Internet"

Was working fine with a recent march bryanslayer build last week.

Can't seem to figure out whats wrong?

My Chromecast v2 (The new model, but not the 4k model) is working great here on r6565

starcms wrote:

Notsofast wrote:

Once I set US and 23 db for power... boom! Permanent Disabled Connection.  Will not come to life, even if i reset and reboot.

....

There must be something linksys has done to the latest batch of refurbished wrt1900ac/v2 units. Don't want to use outdated firmware with security issues. I have a feeling the new kernel does not like the recent certified refurbished ac/v2 hardware.

If you have a 1900ACSv2, 1200ACSv2, or 3200ACM, the power tables are read from EEPROM and not the firmware.  There is no way to change them.

Hence, changing the country code will result in nothing but problems...

So I got a 1900ACS v1, I can change tx power then? Can't seem to find a recommendation for safe tx power.

Cantenna wrote:

starcms wrote:

If you have a 1900ACSv2, 1200ACSv2, or 3200ACM, the power tables are read from EEPROM and not the firmware.  There is no way to change them.

Hence, changing the country code will result in nothing but problems...

So I got a 1900ACS v1, I can change tx power then? Can't seem to find a recommendation for safe tx power.

Yes, you can.  I keep mine maxed out at 30dBm for both 2.4 and 5ghz.

No idea about your Chromecast.  Did you try resetting it to factory defaults and set it up fresh?  Only thing I can think of.

Edit: And in my post, I meant to say 1200ACv2 (there is no 1200ACS )

starcms wrote:

Cantenna wrote:

So I got a 1900ACS v1, I can change tx power then? Can't seem to find a recommendation for safe tx power.

Yes, you can.  I keep mine maxed out at 30dBm for both 2.4 and 5ghz.

No idea about your Chromecast.  Did you try resetting it to factory defaults and set it up fresh?  Only thing I can think of.

Edit: And in my post, I meant to say 1200ACv2 (there is no 1200ACS )

Thanks mate, yea, tried resetting chromecast as well.

Thanks for the info regarding tx power, I'll give it a go, does the 30dBm for both cause premature wear or can the hardware handle it? (Just curious... will by another router if it dies anyways...)_

Cantenna wrote:

starcms wrote:

Yes, you can.  I keep mine maxed out at 30dBm for both 2.4 and 5ghz.

No idea about your Chromecast.  Did you try resetting it to factory defaults and set it up fresh?  Only thing I can think of.

Edit: And in my post, I meant to say 1200ACv2 (there is no 1200ACS )

Thanks mate, yea, tried resetting chromecast as well.

Thanks for the info regarding tx power, I'll give it a go, does the 30dBm for both cause premature wear or can the hardware handle it? (Just curious... will by another router if it dies anyways...)_

No, no premature wear.  They just keep the defaults at 23dBm to keep interference to other people's wifi at a minimum.  Some might argue about health effects, but unless you have the router duct-taped to your head, you'll be fine lol

starcms wrote:

Cantenna wrote:

Thanks mate, yea, tried resetting chromecast as well.

Thanks for the info regarding tx power, I'll give it a go, does the 30dBm for both cause premature wear or can the hardware handle it? (Just curious... will by another router if it dies anyways...)_

No, no premature wear.  They just keep the defaults at 23dBm to keep interference to other people's wifi at a minimum.  Some might argue about health effects, but unless you have the router duct-taped to your head, you'll be fine lol

So , the max I seem to be able to select is 23dbm, that is, at least in the gui

NamesAreAPain wrote:

EDIT: Nope, still having problems. Ping spikes to 20-50ms every minute or so on 2.4 and massive pings in the 100s range on 5Ghz (though I can only test with my phone and steamlink).

If you are going to do:

echo 0 > /sys/kernel/debug/ieee80211/phy0/mwlwifi/tx_amsdu
echo 0 > /sys/kernel/debug/ieee80211/phy1/mwlwifi/tx_amsdu

(not necessary in @david's build since he has AMSDU disabled in the driver), you need to make sure the commands/your script runs the commands BEFORE the wifi interfaces come up (or I guess running them and then restarting the wifi interfaces would work as well).  I know you had said you ran your script AFTER networking.
That's your issue. 

Sorry I neglected to say that before.

Under testing I was able to, but randomly. It was nice at one time to boost my signal on the 5.8ghz to be right up there with 2.4ghz. Usually 5.8ghz  is 10 db lower. Though is don't see any improvement beyond 27 db when I do have a larger scale to choose from. Either I get 23db as my max, or 24db, or 27db or 30db as my max. It's a crap shoot when getting 5.8 to function with changes, and i must mix region and fiddle with max gains on both to enable 5.8ghz. Very time consuming at least, and you cannot change wifi on either band, or boom! Back to square one.

Seems that the older build by David worked for me as for setup but had some issues, that i did not keep for reference as they were not important, just a quick test and painless so to speak. These new builds are a wakeup call for me on seeing such issues.

Have installed proper factory firmware for my router, no longer using ACS/v2 firmware., though i never had an issue with ACS/v1 firmware with wifi gain and performance on my original test runs earlier when breaking in.

Notsofast wrote:

Under testing I was able to, but randomly. It was nice at one time to boost my signal on the 5.8ghz to be right up there with 2.4ghz. Usually 5.8ghz  is 10 db lower. Though is don't see any improvement beyond 27 db when I do have a larger scale to choose from. Either I get 23db as my max, or 24db, or 27db or 30db as my max. It's a crap shoot when getting 5.8 to function with changes, and i must mix region and fiddle with max gains on both to enable 5.8ghz. Very time consuming at least, and you cannot change wifi on either band, or boom! Back to square one.

Seems that the older build by David worked for me as for setup but had some issues, that i did not keep for reference as they were not important, just a quick test and painless so to speak. These new builds are a wakeup call for me on seeing such issues.

Have installed proper factory firmware for my router, no longer using ACS/v2 firmware., though i never had an issue with ACS/v1 firmware with wifi gain and performance on my original test runs earlier when breaking in.

You have an ACSv2!  Your power tables are written to and read from EEPROM and cannot be modified.  Therefore, changing the transmit power level (in LuCi or via SSH) will have ZERO effect.  Changing the country/region will cause your wifi to shut down until you reflash.  It's not that complicated.

And it's 5ghz, not 5.8...

Edit: Actually it's 5.0 thru 5.8GHz depending on the channel to be completely accurate.

My Chromecast v2 didn't work with previus version, today at nitght I will test with r6565.

Is possible compile the latest DNScrypt to OpenWRT?
The latest version support Cloudflare over TLS.

Here the link https://github.com/jedisct1/dnscrypt-proxy/releases.

Thanks.

I would agree with you 100%, but testing shows I can do country and power gain changes and make them stick. But... odd things happen depending on build, and how I setup or make changes in wifi after the fact of having both  bands function.

Initially I heard of ACS/v2 having locked down gain tables, and nobody talked about AC/v2 having them, so i thought I was immune, thus the odd testing results that peeps in the know won't think of trying. If i had not found some results showing I can over ride the locked table, i would not have posted or expected i could with future builds.

Cantenna wrote:

starcms wrote:

No, no premature wear.  They just keep the defaults at 23dBm to keep interference to other people's wifi at a minimum.  Some might argue about health effects, but unless you have the router duct-taped to your head, you'll be fine lol

So , the max I seem to be able to select is 23dbm, that is, at least in the gui

For 2.4GHz, you should be able to select 30dBm on all channels.  For 5GHz, you can only select it on some (the upper four definitely, which I recommend anyway since they don't have DFS).

My Chromecast v2 didn't work with previus version, today at nitght I will test with r6565.