No issues including these two, but would like to hear other opinions... positive or negative??

I just checked around on the openwrt forum and seems to me that cryptodev correct functionality can be check using
openssl speed -evp aes128 -elapsed -engine cryptodev
and **Theorically** since openvpn uses openssl, if it works in openssl it should work with openvpn as well. But BrainSlayer had to enable it specifically for openvpn in his image to make it work so since I have no idea what the output of that command means I was asking for confirmations. I mean it's a pity if openvpn had the chance to use HW acceleration but it doesn't.

Thanks for reply anjd regards

(Last edited by puppinoo on 14 Jun 2017, 15:49)

Absolutely -- Wifi driver is still in pre-release form, but works really well now.  Besides wifi needing additional honing, I would consider the build stable.

Hi again,
<Further reading>
So if I got it correctly for kernel operations cryptodev is enabled by default but since openvpn use userspace to work cryptodev support could need to be enabled manually.
Of course my lack of knowledge doesn't let me have an educated opinion so I seek advice.

Regards

(Last edited by puppinoo on 14 Jun 2017, 16:19)

I probably won't ever use a WRT1x00 as a client bridge, but it might be a good idea to included them, due to the 'catch22' of not being able to install the packages after an upgrade.
Both packages should not interfere and stay disabled until the router is configured as a client bridge.

Checking the current build, there's no advantage (acceleration)  using cryptodev vs the default engine. This tells me cryptodev is missing critical configurations.

Thanks for checking.
I hope it will be enabled (in a performance-wiseuseful way) soon (maybe it's already there, I mean working, and it's just a matter of correct configuration.

Thank you, @starcms. I am using `r4323`, and I also tried to switch between `wpad` and `wpad-mini`  in previous releases. The issue remains.  I opened an issue on the wifi driver's git repo.

I also notice the issue comes with the kernel error message like:

Option or all 3 of them? I meet issues with "ff" before mostly.
Ok... I have 1900v1 among others as well, and have 40Mhz at 2.4 working.
Here is my config:

        option txpower '30'
        option hwmode '11g'
        option noscan '1'
        option bursting '1'
        option htmode 'HT40'
        option channel '4'
        option frag '2340'
        option rts '2340'
        option country 'BM'
        option distance '300'

But - i still use older r3716 release there, so maybe newer one bring this issue.

You must live in the middle of nowhere to be able to even consider using 2.4GHz at 40MHz...

@T-Troll, what is the bursting and noscan options for?

I following LEDE project instruction: Try source vars, clean-all, build-ca etc.

But unable to execute build-ca.

By find / -name ca.crt, I find some cert already located at /tmp/syscfg/openvpn.

For your build, those cert already created and I don't need to rebuild it. Right?

noscan force 40mHz enabled even for heavy-loaded environments ^_^ (yes, it violates the regs).
In fact, i have about 20 visible networks in 2.4
bursting is a part of WMM support, increase stream transfer performance (CTS become per-time, not per-packet, if i remember correctly).

PS: And country code BM - for legal 30dBa in 2.4. Bermuda is nice country.

Hi all
I'm writing for some help, because I don't know what to do
I'm on a 1900acs with "Lede leviathan VI r3716-cd0f990 / LuCI Master (git-17.068.73516-4c10d29)"

I changed the lan default IP to 192.168.2.1 so my lan has 192.168.2.0 net
Trying to reach my modem at 192.168.1.1:80 I was redirected to the Lede-Luci login page.
Checking on this, I noticed that my router it's still using the 192.168.1.1 ip on a vnic
Following the ifconfig's relevant details :
lede# ifconfig
br-lan    Link encap:Ethernet  HWaddr
          inet addr:192.168.2.1 Bcast:192.168.2.255  Mask:255.255.255.0
          inet6 addr: xxxxxxxxxxxxx/64 Scope:Link
          inet6 addr: xxxxxxxxxxxxxx/60 Scope:Global
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2626 errors:0 dropped:0 overruns:0 frame:0
          TX packets:3814 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:244524 (238.7 KiB)  TX bytes:4005021 (3.8 MiB)

eth0      Link encap:Ethernet  HWaddr
          inet6 addr: xxxxxxxxxxxxxxxxxxxxxxx/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:156 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:532
          RX bytes:0 (0.0 B)  TX bytes:29280 (28.5 KiB)
          Interrupt:35

eth0.1    Link encap:Ethernet  HWaddr
          inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:139 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:26998 (26.3 KiB)

Due to this, i'm not able to reach my modem's web gui
Can someone please help me to fix this?

Thanks in advance

p.s. i would like to upgrade my firmware to the latest shelby release, which is the better way to do so?
it's better to switch on the second firmware (still stock) and flash from there, so the first partition (lede) get overwritten, keeping untouched the stock partition?
Or is it possible to upgrade using the "lede-mvebu-linksys-wrt1900acs-squashfs-sysupgrade.bin" from the flash firmware web ui's page?
I go to read again the FAQs

Many thanks (one more),
Kom

(Last edited by komodo_1 on 19 Jun 2017, 14:08)

@komodo_1

Since both the modem and sub interface eth0.1 have the same IP, that's the problem. Can you change the sub interface IP OR modem IP?

If you're not going to use the stock firmware any more, run the sysupgrade -F *.img from ssh, and blow it away.  Otherwise, boot over to the stock firmware and flash from there to blow away the leviathan lede install. 

Would like to hear someone chime in.. but can you flash the .bin image from the stock firmware to upgrade a lede on the other partition?

@kwauhk2000

Looks like the current package has issues generating the certs. Have you checked around to see if others are having the same issue? If not, you may want to put in a bug report for the issue in Flyspray.

A very recent post of someone else having an issue  >>  https://forum.openwrt.org/viewtopic.php?id=70004

(Last edited by davidc502 on 16 Jun 2017, 14:01)

@komodo_1 Here you can find how you can flash davidc502sis.dynamic-dns.net/faq/
As I know you can do both. flashing from your lede and overwrite the stockfirmware or go back to stock and do it from there. I had 2 lede's at my router on both partitions and it worked fine. You can always flash back to stock if you want. In other words it's on you if you prefer stock/lede or lede/lede.

to your ip adress problems.
Where did you change the lan ip? Luci? or /etc/config/network?
Did you tried a reboot?
Maybe you could try to change the ip address again if something was messed up the last time? Just guessing

EDIT I was to slow with my reply

(Last edited by raspioff on 16 Jun 2017, 14:19)

I just want to confirm this. I have the same problems with my wrt1900acsv2 as client or clientbridge and thx for opening an issue on git. If you find any solution please let us know Thx.

Hi david, thanks for your reply.

I confirm that the problem with the modem gui reachability it's due to the modem and the router using the same ip (x.x.1.1)
The weird thing it's that my router's lan interface it's configured  to use the 192.168.2.0 net and not the 192.168.1.0 one

config interface 'lan'
    option type 'bridge'
    option ifname 'eth0.1'
    option proto 'static'
    option netmask '255.255.255.0'
    option ipaddr '192.168.2.1'
    option delegate '0'

I don't know if I can act on the modem's lan ip to change it, because i'm not at home actually and with this strange behavior i cant reach the modem via ui to test this, that's why i was testing the ip change on openwrt.

br-lan    Link encap:Ethernet  HWaddr
          inet addr:x.x.2.1 Bcast: x.x.2.255  Mask:255.255.255.0
          inet6 addr: xxxxxxxxxxxxxxxxxxxxx/64 Scope:Link
          inet6 addr: xxxxxxxxxxxxxxxxxxxxx/60 Scope:Global
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
[...snip...]
eth0.1    Link encap:Ethernet  HWaddr
          inet addr:x.x.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

openvpn is disabled, i tried many times to completely disable ipv6 but it's still active on br-lan

I'm not able to understand what's starting the x.x.1.1 ip on the eth0.1 interface and where is it configured.

I did a grep to find it and some files still contain that ip, but i can't edit the most of them (not writable partition i think)
i.e. /bin/config_generate, /lib/preinit/00_preinit.conf, /rom/etc/config/openvpn, /rom/etc/config/openvpn_recipes, ecc.

I'm actually using this env variables
   boot_part_ready=3
   boot_part=2

Thanks again for your suggestions,
K

(Last edited by komodo_1 on 19 Jun 2017, 14:09)

Sorry raspioff, i read only now your reply.
I changed the lan ip from luci, and this is corectly reflected on /etc/config/network
I rebooted many times but that class it's still there.
I actually solved my problem changing the modem internal ip, but now I need to understand where this 192.168.1.x on eth0.1 comes from

I read the faq just after the post, thank you for sharing your experience and for the clarifications

In LuCI, when you go to Network --> Interfaces --> LAN  --> Physical Settings, are you sure you have bridge interfaces checked and below that have eth0.1 and both your wireless networks checked/selected (and only those checked, not eth0 or anything else)?

When I run ifconfig, it doesn't show any IP for eth0.1 (because it's a part of br-lan), only for br-lan,  It seems for some reason your bridge interfaces isn't working correctly.  However, the snippet of /etc/config/network you posted is correct.:

root@WRT1200AC:~# ifconfig
br-lan    Link encap:Ethernet  HWaddr XXXXXXXXXXX
          inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: XXXXXXXXXXXX/64 Scope:Link
          inet6 addr: XXXXXXXXXXXX::1/64 Scope:Global
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:8382958 errors:0 dropped:0 overruns:0 frame:0
          TX packets:9454098 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:4201857205 (3.9 GiB)  TX bytes:16955034735 (15.7 GiB)

eth0      Link encap:Ethernet  HWaddr XXXXXXXXXXX
          inet6 addr: XXXXXXXXXXXX/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1257231 errors:0 dropped:0 overruns:0 frame:0
          TX packets:148935 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:532
          RX bytes:161392210 (153.9 MiB)  TX bytes:126078789 (120.2 MiB)
          Interrupt:36

eth0.1    Link encap:Ethernet  HWaddr XXXXXXXXXXX
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:43900 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:7338552 (6.9 MiB)

(Last edited by starcms on 17 Jun 2017, 08:47)

Hi starcms,
thanks for interesting
Yes, i confirm the physical settings are correct

I'll keep seek for it, but maybe after the update will be resolved

raspioff did you ever try to use a bin file to update, from the other partition (stock/lede or lede/lede), an already running lede installation?

Thank again for cheking on it
Good week-end to all

(Last edited by komodo_1 on 17 Jun 2017, 16:14)

eth0 and eth1 swapped on WRT1900ACSV2?

I was reviewing my config and about to make some changes, when I noticed that my etho and eth1 seemed to be swapped from the documentation and many (everyone elses) online references for the wrt1900acsV2.

On my setup,  eth0 is on LAN and eth1 is on WAN on my setup, and  WAN is on VLAN2 eth 1.2

I am currently running reboot snapshot r4088-7296767, and I did not preserve settings on that upgrade and redid all my configs.

I have read somewhere that LEDE made a change and swapped eth0 and eth1, but seems like this was done a while ago and I'd already have it?

https://git.lede-project.org/?p=source. … da518a2eb8

Or something else I have to do to get these to match convention?

Or maybe I'm confused.

Now on all models, Eth0 (and Eth0.1 using default switch settings) is LAN.  Eth1 (and Eth1.2 using default switch settings) is WAN, just as the page you linked says, and apparently exactly how your router is configured from what you said.

Before the LEDE swap, Eth1 was LAN and Eth0 was WAN on your model (as is still shown on the openwrt wrt wiki page).

(Last edited by starcms on 18 Jun 2017, 00:27)

Love your AP names; always great to meet another TWD fan

Edit: I believe I found your issue.

From the picture you posted, I noticed that you have the interface "ModemTI" on the same VLAN as WAN and WAN6 (eth1.2).  Is ModemTI the static interface you created set to 192.168.1.1 to be able to access your modem?    If so, ModemTI should be set to be on a LAN VLAN (eth0.X), not on the WAN VLAN (eth1.2) as it is now.  So you would need to create a separate LAN VLAN for ModemTI (such as eth0.3) (also, as it is right now, you possibly have a giant hole in your firewall). This is most likely where your problems are coming from and why ifconfig is showing eth0.1 has an IP of 192.168.1.1 (even though ifconfig shouldn't show any IP for eth0.1 because it is part of a bridge, br-lan).  You are basically confusing the router because of the incorrect switch settings and for some reason it is generating that 192.168.1.1 IP on eth0.1 even though eth0.1 is part of br-lan with a static IP of 192.168.2.1.

A MUCH, MUCH simpler solution would be to delete the ModemTI interface completely, and set the LAN interface back to 192.168.1.1 (basically restore everything in LuCI --> Network --> Interfaces and LuCI --> Network --> Switch (if you changed any Switch settings) to the defaults or install @david's latest build and choose not to keep settings).  Then go to LuCI --> Services --> uHTTPd, and simply change the HTTP IP from 192.168.1.1:80 to 192.168.1.2:80 (or any IP with 192.168.1.X, where X is not 1) and HTTPS IP from 192.168.1.1:443 to 192.168.1.2:443.  Then you can access your modem from 192.168.1.1, and your router/LuCi from 192.168.1.2.  (To access the router via SSH, you would still use 192.168.1.1)   Just incase you are curious, uHTTPd is the webserver on LEDE that allows you to access the router/LuCi from your browser.

Or, since you can change the modem's IP, if you wish to keep the router's LuCI IP and SSH IP to both be the same to avoid confusion, instead of changing uHTTPd settings on the router, as I just described above in BOLD, change the modem's IP to be 192.168.1.2.  Then you can access the modem at 192.168.1.2, and access the router via LuCI and SSH at 192.168.1.1

(Last edited by starcms on 18 Jun 2017, 01:39)

@david, nope, only the full .img can be flashed from the stock Linksys firmware. 

But just curious, what would be the point of "upgrading" LEDE on the other partition (using the .bin instead of the .img)?  All the settings and packages that had been installed on the LEDE build on the other partition would still be lost, even if you could flash the .bin from the stock Linksys firmware -- just like if you flash the .bin from LuCi but choose not to keep settings.  Because for instance if you have LEDE on both partitions, and you flash the .bin from LuCi without choosing keep settings, the build on the other partition that you are "upgrading" doesn't retain any added packages or settings.  It has everything stock, just as if you had flashed the .img.

(Last edited by starcms on 18 Jun 2017, 01:36)