The absolute latest (beta 01-02-2016-r28647) appears to brick and not be recoverable from the recovery images in http://ww2.dd-wrt.de/phpBB2/viewtopic.php?t=283784 - the router downloads the image off of tftp, but it stays in the crashloop (otoh, I didn't pay enough attention to what exact version of stock I had on originally, though it the upgrade obviously did attempt to work). I haven't tried serial or anything (and am unlikely to bother).
Joined: 26 Nov 2011 Posts: 37 Location: Sofia, Bulgaria
Posted: Sun Jan 03, 2016 2:59 Post subject:
Per Yngve Berg wrote:
VLAN2 is the WAN port and shall be assigned to Bridge None.
Why do you have port 2,3 and 4 in vlan2 with the wan port?
For IPTV Set top boxes. They used to connect with an ethernet switch, to remove the switch and connect them straight to the router, I setup a vlan with the WAN interface for the 3 STB's. Works like a charm without a switch _________________ TP-Link Archer C9 AC1900 v1: DD-WRT v3.0-r29621
ISP: Blizoo 50/4 MBits
The absolute latest (beta 01-02-2016-r28647) appears to brick and not be recoverable from the recovery images in http://ww2.dd-wrt.de/phpBB2/viewtopic.php?t=283784 - the router downloads the image off of tftp, but it stays in the crashloop (otoh, I didn't pay enough attention to what exact version of stock I had on originally, though it the upgrade obviously did attempt to work). I haven't tried serial or anything (and am unlikely to bother).
I have same problem, your post should be on the top!
loaded 28493 and am getting what appears to be a bogus generic address of 00:90:4C:0E:51:23 attached to eth2(the 5g wireless Nic).
Bought an identical c9 model and it also came up with this address. where is DD-WRT getting this from?
anybody else have this issue? have not seen this on any of the multitude of routers I have configured for DD-WRT. Unfortunately, flashing back to stock on this unit is not easy. _________________ Please state what make and model router plus the build number and type of DD-WRT you are using. Screen prints and a network diagram can are also helpful. Before you create a new post, use the search function. Chances are your issue has happened to someone else.
loaded 28493 and am getting what appears to be a bogus generic address of 00:90:4C:0E:51:23 attached to eth2(the 5g wireless Nic).
Bought an identical c9 model and it also came up with this address. where is DD-WRT getting this from?
That's the mac address from the cfe, being filled in by the cfe into nvram when you make a long reset.
Stop doing those long resets!
Flashing stock firmware will fix it, you can also set the nvram mac variables manually. _________________ Kernel panic: Aiee, killing interrupt handler!
what's the command to set the mac address manually?
i tried nvram set et0macaddr={the mac address printed on my router}
nvram commit
and it corrected eth0 and eth1 but eth2 remained with the 00:90:4C:0E:51:23 address
tried nvram set eth2macaddr={the eth0 mac address incremented by 2} and it set eth2 to the same address as eth0 and I ended up with a mysterious eth3 interface still set to 00:90:4C:0E:51:23
only way to get back to the way the original way was to issue an erase nvram command
thanks again _________________ Please state what make and model router plus the build number and type of DD-WRT you are using. Screen prints and a network diagram can are also helpful. Before you create a new post, use the search function. Chances are your issue has happened to someone else.
along with eth2, br0 and br0:0 are set to this address of 00:90:4C:0E:51:23
doing a nvram show|grep mac reveals a couple of nvram variables that seem to wrong
0:macaddr and 1:macaddr
still seem to have values from the range of 00:90:4C
i fixed the erroneous bridge interface addresses going into setup/networking/bridge _________________ Please state what make and model router plus the build number and type of DD-WRT you are using. Screen prints and a network diagram can are also helpful. Before you create a new post, use the search function. Chances are your issue has happened to someone else.
seems like the non intuitive fix to the problem of eth2 being set wrong is to run the commands
nvram set 1:maccaddr={the address printer on the router incremented by 2} because the actual address is eth0, the 2.4 ghz lan card is eth1, and the 5 ghz wlan interface is eth2.
nvram commit
reboot
whats odd is there is a 0:maccaddr variable which is still set to some bogus address but doesn't seem to affect anything _________________ Please state what make and model router plus the build number and type of DD-WRT you are using. Screen prints and a network diagram can are also helpful. Before you create a new post, use the search function. Chances are your issue has happened to someone else.
I have a TP-LINK AC1900 C9 v1. I used the stock GUI to load betas/2015/12-23-2015-r28586/tplink_archer-c9v1/factory-to-ddwrt.bin. DD-WRT booted and got a WAN IP and I was able to access the Internet. The only problem is that the router rebooted every few minutes. I tried to load an older DD-WRT webflash but the router rebooted again while it was in progress. Now it starts up with all lights blue and nothing works.
I want to get back to a working DD-WRT. I'm able to get into recovery mode and serve archerc9v1_tp_recovery.bin from a TFTP server. After the file transfer is done, nothing happens. The single light with the down and up arrows remains on and the router doesn't restart or anything. I read in other posts that it should reboot itself immediately after getting the recovery bin.
Now what?
I could really use some help here. I had to submit a rebate and can't return the router now. Does anybody have advice on using TFTP? Should uploading factory-to-ddwrt.bin renamed to archerc9v1_tp_recovery.bin reload DD-WRT?
The absolute latest (beta 01-02-2016-r28647) appears to brick and not be recoverable from the recovery images in http://ww2.dd-wrt.de/phpBB2/viewtopic.php?t=283784 - the router downloads the image off of tftp, but it stays in the crashloop (otoh, I didn't pay enough attention to what exact version of stock I had on originally, though it the upgrade obviously did attempt to work). I haven't tried serial or anything (and am unlikely to bother).
I just ran into this same problem, I had to revert to stock firmware using the serial unbrick method. Once back into stock, I used the web panel to flash 12-24-2015-r28598 and now it's working great!
The simpler recovery+TFTP method doesn't work, the crash loop persists after flashing. I was lucky that I happened to have the required hardware (USB UART and jumper wires) from my PS4 mouse/keyboard setup.
If anyone tries the serial unbrick method, make sure that you plug in the board and boot it into the crash loop before connecting the USB to your PC, otherwise it won't boot. You'll then see the boot output in putty (including the disk read error that results in the crash) and you can hold down ctrl-c to get into the CFE console on next boot.
lmpqwe wrote:
I want to get back to a working DD-WRT. I'm able to get into recovery mode and serve archerc9v1_tp_recovery.bin from a TFTP server. After the file transfer is done, nothing happens. The single light with the down and up arrows remains on and the router doesn't restart or anything. I read in other posts that it should reboot itself immediately after getting the recovery bin.
Now what?
This sounds similar to my issue, to fix it you'll need to use the serial unbrick method.
Has anyone tried a console flash of this build yet?
I have searched and am not able to find the changes made between revisions. Can someone point me to the release notes or git logs between revision r28882 and previous builds?
Has anyone tried a console flash of this build yet?
I have searched and am not able to find the changes made between revisions. Can someone point me to the release notes or git logs between revision r28882 and previous builds?
Has anyone tried a console flash of this build yet?
I have searched and am not able to find the changes made between revisions. Can someone point me to the release notes or git logs between revision r28882 and previous builds?
Appreciate all who work on dd-wrt! Thanks.
Being the first time ever flashing a router I was pretty darn nervous. However, after reading many many articles on this site I finally gave it a try. It appears the firmware flash was successful using 01-19-2016-r28882/tplink_archer-c9v1/factory_to_ddwrt.bin. With that said, I do have some current issues using openvpn and wireless AC.
When openvpn is configured the logs show the its connected but none of the LAN connections are able to get out on the WAN. I followed the setup instructions for two different providers (expressvpn and airvpn). Neither worked. I talked with ExpressVPN and they said the current firmware is not able to do Openvpn because the firmware has to be either huge or mega.
The other issue is with wireless port on 5G. My roku4 device is AC compatible and its not able to see the 5g wireless lan; only the 2.4g lan.
Again, this could simply be because I'm a first time user of dd-wrt. Bottom line, archer c9 router is more stable now with dd-wrt. Only time will tell if I can get the two issues resolved.
EDIT:
Figured out both issues. The openvpn was fixed by going to airvpn and using their setup instructions.
Got Roku4 to see the 5g network after manually setting the wireless channel to 36 and the channel width to 40. Setting the wireless channel to auto or any other setting > 36 the Roku4 does not see the ac network.
Last edited by twisted_wrister on Thu Feb 04, 2016 17:11; edited 1 time in total
Posted: Sat Jan 23, 2016 20:40 Post subject: Issue with r28598 on Archer C9 V1
I been testing the r28598 build since I read a few others have had success. My luck hasn't been so good, unfortunately. The two most prominent issues:
1) The web-UI only start every 3rd or 4th restart of the router. I can always get into the telnet/ssh admin though, but don't know how to diagnose past that (and help would be welcome - is there an in-memory log?).
Here is what a ps showed during one boot where there was no web-UI:
2) The per-host connection-counts ("Conn. Count") under Status->LAN don't seem right. On my own PC's LAN connection for example, the count shows MUCH higher than what netstat tells me (active and time_wait). A side effect, and what caused me to notice this, is that if you set "IP Filter Settings (adjust these for P2P)" under Admin-Mgmt to something low like 1024 - then just f5, you can see the total connection count get to that point then the webUI no longer responds - even though back on my own netstat I only show a couple hundred connections.
(NB: should/does the web UI use HTTP-keep-alives?)
Does any one else see this? If you go to Status->LAN and f5 over and over, does your host-count creep into the hundreds/thousands?
EDIT - yes I know the ps shows what some will consider sensitive info. I don't mind that so PLEASE, thank you for but don't mention it
