1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Tomato RAF Releases

Discussion in 'Tomato Firmware' started by Victek, Dec 28, 2012.

  1. Victek

    Victek Network Guru Member

    I personally think a web gui is not needed for udpxy because the usage is very simple. Some prior considerations to the usage of udpxy:

    udpxy converts multicast protocols RTP or UDP to unicast protocol HTTP.

    Why?
    • Some devices in our network don't support multicast protocol (some switches)
    • You are not able to see video stream in one device in your network not supporting multicast protocol (i.e: some smartphone or android tablet)
    • You want to access to streaming from a different network (i.e: streaming from one location to other location over Internet or WAN/VPN).
    How to run and settings for udpxy?
    • Access to your router IP via telnet command. (telnet 192.168.1.1).
    • Simply type udpxy from cli prompt and you have access to the user guide.
    Examples:
    • You have some examples in the user guide (bottom) when you typed udpxy in cli mode.
     
    Elfew likes this.
  2. Elfew

    Elfew Addicted to LI Member

    Thank you! I will check this :)
     
  3. Victek

    Victek Network Guru Member

    Perfect, I think it's useful to see TV program in your tablet or android not supporting VLC .. ;)
     
    Elfew likes this.
  4. emporium

    emporium Reformed Router Member

    Thanks Victek for your work with this firmware.

    I wonder if it is possible to add iodine to the firmware. It's source code is available but I don't know if it can be built for tomato.

    I think it can be useful having it's funcionality in tomato.

    What do you think about.

    Thanks.
     
  5. Victek

    Victek Network Guru Member

    Just a quick hands up note provided by RMerlin (Thank you),

    Asus will release RT-N66U new batches, the Memory Flash chip is NOT compatible with the existing kernel code in Tomato builds available till now.

    The new units can be identified (not confirmed) by label B2 version in the unit plate.

    Don't try to upgrade the unit with existing Tomato builds, you may find a kinda bricked unit. RMerlin suspects that it should still be recoverable through FW recovery mode, not confirmed yet.

    In order to avoid surprises a Tomato RAF version for RT-N66 with a patch for the new batches is available in victek's site now, once you select to download RT-N66 firmware you will get beta release V.1.1wx.

    This version is compatible with up to now sold RT-N66 and also new RT-N66 batches to come labeled B2 (not confirmed yet).

    Patch it's available also in the Tomato-RAF-RT-N git repository to be added in other Tomato builds.

    Thanks!

    I must declare it as one great step ahead that demostrate a teamwork for Tomato firmware. Thanks to RMerlin for the information. You have the available firmware before the unit it's in your hands (I hope !)....
     
    zapoqx, Marcel Tunks and Elfew like this.
  6. Elfew

    Elfew Addicted to LI Member

    Last hw revision before new model? Maybe. Anyway, thank you Merlin and Victek
     
  7. RMerlin

    RMerlin Network Guru Member


    There's already a new model out (RT-AC56U), with its big brother RT-AC68U following sometime late summer/early autumn.
     
  8. Elfew

    Elfew Addicted to LI Member

    But it is not tomato capable, am I right?
     
  9. Victek

    Victek Network Guru Member

    Too early to say no .. it's another challenge for Tomato with the new Broadcom dual core ARM controller. If we work as a team and we have resources (test unit), We can. ;)
    Look at: http://goo.gl/0dPgH

    It promises 802.4 Mbps WAN-LAN for RT-N56U ... and 901.4Mbps for Linksys EA6300-EA6400...
     
    Elfew likes this.
  10. RMerlin

    RMerlin Network Guru Member


    Technically, it's possible. It runs Asuswrt like the other Asus routers, which is actually a Tomato fork. It will only require developers with the willingness and the time to port Tomato to it. It's based around Broadcom 6.x SDK, so Shibby's RT-AC66U port could be used as a starting point once it's fully working. You'll then need the 2.6.36 kernel that Asus uses (with Broadcom's SDK applied to it), and the ARM toolchain. After that, new binary drivers, and some tweaks here and there to handle the differences related to toolchain/SMP/ARM.

    I've had an RT-AC56 for about two months now, got my own firmware running on it. It's a really nice router, and the dual-core 800 MHz CPU would be a boon to Tomato. I did a few routing tests with CTF disabled, and I was able to reach 240 Mbits between WAN and LAN.
     
    Elfew likes this.
  11. Elfew

    Elfew Addicted to LI Member

    Nice, thank you for info! I think it will be next major step for Tomato project in near future...
     
  12. JugsteR

    JugsteR Serious Server Member

    I think you are correct in one of two accounts. Depending on what near means of course. :)

    Signature? Signature!
     
  13. Jeffers

    Jeffers Reformed Router Member

    Victek - thanks for the excellent firmware. I just came over from dd-wrt and have been very impressed.

    One question - I'm trying to connect it to my existing mrtg setup but mrtg is throwing errors because the interface isnt reporting a max speed when querying the WAN port on my E3000:

    ### Interface 30 >> Descr: 'No-Description' | Name: 'No Name' | Ip: '192.168.0.1' | Eth: 'No Ethernet Id' ###
    ### The following interface is commented out because:
    ### * has no ifSpeed property
    ### * got 'Received SNMP response with error code
    ### error status: noSuchName
    ### index 1 (OID: 1.3.6.1.2.1.2.2.1.10.30)
    ### SNMPv1_Session (remote host: "router" [192.168.0.1].161)
    ### community: "public"
    ### request ID: -413816998
    ### PDU bufsize: 8000 bytes
    ### timeout: 2s
    ### retries: 5
    ### backoff: 1)' from interface when trying to query

    Is this a known issue? Im using 1.28.9013

    Many thanks
    Jeffers
     
  14. koitsu

    koitsu Network Guru Member

    It's going to vary depending on version of the SNMP daemon and underlying code used by the SNMP daemon. Reliable interface speed detection on Tomato-based routers has not been something commonly available; please do not say "DD-WRT has it so why can't Tomato". The problem relates to the immense amount of variance between PHYs and chips used across the various models of routers, and in some cases even sub-revisions of the same router model (ex. hardware revision 1 vs. 2). The userland programs to query the PHY information are equivalently unreliable (for example there are all sorts of forked versions of the robocfg script which work on model X but not Y, or throw out random/insane values (such as -84238923493) on other models); what may work for person X doesn't work at all for person Y, or in some cases may induce a kernel panic.

    Your mrtg scripts can cease querying of that OID and instead just hard-code the value to either 100000000 (100mbit) or 1000000000 (1000mbit), depending on the kind of device you have attached to your WAN port. There should (hopefully) be some visual indicator, e.g. LED, on one of the two devices that tells you what its speed is. Other cases (such as on cable modems) there are usually administrative interfaces available at http://192.168.100.1/ which can sometimes help determine that. For example on my SB6121 the negotiated interface speed is indicated via an LED on the modem itself.

    And besides, if the ifSpeed was to change one day (say from 1000mbit to 100mbit), your graphs would look all screwed up and any mathematics used to determine rates/speeds relying on that number would then become skewed. Think about that. :) Just hard-code the value in your mrtg.cfg and be done with it.
     
  15. Victek

    Victek Network Guru Member


    Thanks for your test, as koitsu said it depends on phy version and custom modifications (heads up about phy module in tomato raf, it's not standard broadcom code, you are able to see additional elements like the ethernet port monitor), so, some 'ID' have been changed.
    In this case, socket specs are equal, so max speed for wan it's available but don't know if your settings now are same as you had with dd-wrt firmware, interface IP for instance it's not the tomato default, could you double check settings like a simple ifconfig and compare?

    I never user rtgm since networks activity and statistics in Tomato are one of the strong points with svg graphics compared with average firmware but I'll give a try to get familiar with this utility.

    Thanks
     
  16. Planiwa

    Planiwa LI Guru Member

    Did I mention that I use standard shell tools?
    So, then I did throughput measurements with ethernet:

    91 Mb/s
    100% router CPU utilization.

    Hm.

    I'd be interested in what others may have found, using standard shell tools to test throughput.

    Thanks!
     
  17. Elfew

    Elfew Addicted to LI Member



    I just know that next my new router will be only router with Tomato support!
     
    Victek likes this.
  18. eahm

    eahm LI Guru Member

    Ok, tired of waiting. Flashing N66U wx.

    1) How the hell can a new firmware optimizes so much from a few months old one? I was using the latest .8 Toastman and was (and still) downloading ~25 torrents to test line and router.
    With Toastman I couldn't surf much, I thought it was normal since the number of torrents downloading but with RAF wx I can surf and download apps like there are no torrents.
    This is insanely fast, fastest firmware I've ever tried, and this is only with a 10 mins test.

    I will keep using Toastman for business but I think I am pretty much converted when a "stable" 1.2 arrives.

    2) I don't use eMule but I use it to test UPnP capabilities since it's the only app that sometimes gives me problems opening ports by itself.
    With Shibby and Toastman it never works, sometimes I really have to open them manually. With RAF wx UPnP worked right away during the port test before opening eMule.
    I think P2P test are great for a router. I will keep testing.

    3) I am testing WiFi for the first time with USA country setup and 0 power. I always use Singapore and 0, let's see how it goes.
     
    Victek likes this.
  19. Victek

    Victek Network Guru Member

    Waiting your wireless test, thanks for testing. About your comments.. each Beta is one step ahead, there is a lot of work inside Tomato RAF and hundred of hours since I went back to my hobby earlier this year.
     
    eahm and Elfew like this.
  20. Victek

    Victek Network Guru Member


    Yes, normal stats, could you test CPU usage via cli instead of gui? top command.. Thanks!
     
  21. Victek

    Victek Network Guru Member

    Did you tried dnscrypto-proxy? Basic/Network.. middle of the page .. test and feedback please. I didn't saw your post before, sorry for delay answering.
     
  22. Planiwa

    Planiwa LI Guru Member

    By shell I mean what some people call "cli". By standard tools I mean such programs as top and iostat, etc. I'd use iostat, but it's not included in any "standard" distro. No out-of-the-box Tomato distro includes a standard Unix shell environment, even for the RT-N66U. (Yes, I know about optware and entware and Rodney's fabulous work.)

    In any case, I would like to compare notes with peers (ideally sysadmins) who are interested in using standard Unix shell tools, rather than "installing" all sorts of fancy "applications" on their "Windows" "PC" or whatever. In this particular instance, I would like to compare notes with peers who have also measured WiFi throughput, using standard Unix shell tools, on their routers and their LAN hosts.
     
  23. Victek

    Victek Network Guru Member

    Ok, I understood wrong then, cli (custom little Interface) it's the 'router shell', we don't have all 'shell' commands. ;)
     
  24. Planiwa

    Planiwa LI Guru Member

    I am taking a close look at the first part of the first web page only:

    Code:
    status-overview.asp
     
    Name                            N66U
    Model                          Asus RT-N66U
    Chipset                        Broadcom BCM5300 chip rev 1 pkg 0
    CPU Clock                      600 MHz
    CPU Utilization                3.29%
    Load Avg. (1 / 5 / 15 mins)    0.00 / 0.00 / 0.00
     
    Date & Time                    Sun, 14 Jul 2013 01:24:26 -0400
    Uptime                          2 days, 01:30:51
    RAM Flash Size                  32 MB
    RAM Used / Total                249.96 MB / 236.22 MB (94.50%)
    NVRAM Used / Total              64.00 KB / 38.74 KB (60.53%)
    Code:
    [/FONT]
    [FONT=courier new]Tomato v1.28.9013 MIPSR2-RAF-V1.1w K26 USB VLAN-NGINX-64K[/FONT]
    [FONT=courier new]nvram show |tail -1[/FONT]
    [FONT=courier new]1180 entries, 25855 bytes used, 39681 bytes free.


    Should these two sets of figures not agree?


    As for RAM memory -- it appears that the 256 MB is used by 2 file systems as well as main memory.
    I have placed a file of 100MB on the /dev file system:

    Code:
    [/FONT]
    [FONT=courier new]  df -h[/FONT]
    [FONT=courier new]Filesystem                Size      Used Available Use% Mounted on[/FONT]
    [FONT=courier new]/dev/root                9.0M      9.0M        0 100% /[/FONT]
    [FONT=courier new]tmpfs                  125.0M      3.2M    121.7M  3% /tmp[/FONT]
    [FONT=courier new]devfs                  125.0M    95.5M    29.4M  76% /dev[/FONT]
     
    [FONT=courier new]  free[/FONT]
    [FONT=courier new]            total        used        free      shared      buffers[/FONT]
    [FONT=courier new]Mem:        255956      138768      117188            0        11204[/FONT]
    [FONT=courier new]-/+ buffers:            127564      128392[/FONT]
    [FONT=courier new]Swap:            0            0            0[/FONT]
    [FONT=courier new]


    It seems to me that I have used more than half the RAM, but the GUI page says 95% "free".

    (I'm tired of battling the stupid formatting!)
     
  25. Victek

    Victek Network Guru Member

    curious.. here my stats..

    RAM Flash Size 32 MB
    Total / Free RAM 249.96 MB / 238.41 MB (95.38%)
    Total / Free NVRAM 64.00 KB / 38.68 KB (60.44%)


    Tomato v1.28.9013 MIPSR2-RAF-V1.1x K26 USB VLAN-NGINX-64K
    root@RT-N16:/tmp/home/root# df -h
    Filesystem Size Used Available Use% Mounted on
    /dev/root 9.0M 9.0M 0 100% /
    tmpfs 125.0M 160.0K 124.8M 0% /tmp
    devfs 125.0M 0 125.0M 0% /dev
    root@RT-N16:/tmp/home/root# free
    total used free shared buffers
    Mem: 255956 24572 231384 0 3276
    -/+ buffers: 21296 234660
    Swap: 0 0 0
    root@RT-N16:/tmp/home/root#top
    Mem: 24684K used, 231272K free, 0K shrd, 3276K buff, 9696K cached
    CPU: 0% usr 0% sys 0% nic 100% idle 0% io 0% irq 0% sirq


    No sirq, key improvement :)
     
  26. Planiwa

    Planiwa LI Guru Member

    I was making 6 points, the first 4 about Presentation, the last 2 about content:

    1. "CPU Utilization" would be more standard, meaningful, and clear than "CPU Load".
    2. "Load Avg." would be more standard, meaningful, and clear than "CPU Load".
    3. "RAM Used / Total" would be very much more standard, meaningful, and clear than "Total / Free RAM".
    4. "NVRAM Used / Total" would be very much more standard, meaningful, and clear than "Total / Free NVRAM".

    5. The values given for "Free RAM" are completely false.
    6. The values given for "Free NVRAM" are completely false.

    = = = = =

    #1 and #2 are simply a matter of standard usage.
    #3 and #4 have been utterly confused and uncorrected for 10 years.

    The correct source for #5 is "free".
    The correct source for #6 is "nvram -show|tail -1"

    = = = = =

    If you want to monitor changing memory use every 10 seconds, for 5 minutes, you can run this:

    Code:
    N=0;while [ $N -lt 30 ]; do N=$((N+1));set -- $(free); echo "$N Mem Used / Total: $8 / $7 $((100*$8/$7))%"; sleep 10; done &
    
    Then you can run this to slowly (taking more than 2 minutes) create a 100MB file:

    Code:
    dd < /dev/urandom > /tmp/100MB bs=1000000 count=100
    
    And you should be able to watch the "Status: Overview" page update, if it works properly.

    (Which it does not!)

    = = = = = [Edit] = = = = =

    Perhaps I should add that these problems were not introduced by this particular sub-version of Victek's FW. They have been around for a long time, in every Tomato FW.

    Perhaps some people feel that the time has not yet come to fix these problems.

    Perhaps some people feel that a Beta tester should ignore old bugs and only focus on the new "features".

    Obviously I am not one of those people. ;-)
     
  27. eahm

    eahm LI Guru Member

    I will keep testing but these are the first impressions:

    USA drivers = starts around 16MB/s and it stabilizes around 13.5MB/s

    Singapore drivers = starts around 16MB/s and it stabilizes at 15.5MB/s

    Still very good and fast since I've never reached over 12MB/s with stabilization around 9MB/s before.

    My wife's laptop is a Lenovo T510 with an Intel Centrino Ultimate-N 6300 AGN 450mb/s WiFi card. The laptop is around 12 meters (40 feet) away with a wall in the middle. The router (N66U) is at 1.5 meters (5 feet) from the ground.
     
  28. zapoqx

    zapoqx Networkin' Nut Member

    So, this morning, I woke up to find no one had internet connection. Seemed that my scheduled router reboot somewhat failed. Unfortunately, I don't know what caused it. I set the router to reboot at 5 AM on Sunday Morning. That said, the last time a random error like that happened, it was... hmm... before the 1.0 release and only once every few weeks. I just want to see if anyone else had trouble with the schedule reboot function. I unfortunately don't have the logs set to somewhere off-site to review anything on what stopped it (Like if it was before reboot or during).
     
  29. JugsteR

    JugsteR Serious Server Member

    What does rebooting give you? I am always online 24/7 and have never experienced any problems. Surely if something is leaking memory or there is another problem that requires you to reboot, it would be known by now?

    Can this be carrier related, I don't know...
    Maybe I'm just lucky.

    Signature? Signature!
     
  30. zavar

    zavar Networkin' Nut Member

    I haven't seen the reboot issue. Mine is scheduled for every Monday morning. Though I know it's likely not necessary, I would rather have it reboot when no one is using it then have the family call me when I'm at work and say there's no internet.
     
  31. zapoqx

    zapoqx Networkin' Nut Member

    I made it a thing to do each week regardless of flavor of tomato in part that first started it all was when one time, one of the tomato flavors didn't want to release ports sometimes and caused issues as well as some issues from IPv6. So its just at this point standard for me. I know, not really necessary these days, but sometimes, its good to just be safe about certain things. That way, I don't have to hear "Hey! [Something] isn't working!" with me saying "A reboot shouldn't be required" just to find out a reboot was needed. And their method of reboot really screws up my log info as they just turn the router off and on.
     
  32. Planiwa

    Planiwa LI Guru Member

    Code:
    13.07.14_13:30:01  eth1: -79 dBm Noise  47°C  3 WiFis  eth2: -84 dBm Noise  50°C  1 WiFis
     
    Connections:79 TCP:27 UDP:51 DNS:46
     
    Boot 35d16h ago    INIT 8 Jul 01:00    WAN up 10 Jul 20:03    FW up 10 Jul 20:04  QOS: 0
    
    (35 missed opportunities to have a failed boot. :)

    Interesting question. I wonder if there are common (observed) reasons for failures of scheduled re-boots, perhaps observed at sites that do have external syslogs.

    I wonder if there is a state where the system is "up" enough to be able to examine its own state and restart if it finds it is stuck. Some kind of pre-Init minimalist consciousness.
     
  33. Victek

    Victek Network Guru Member


    It's part of Broadcom code adapted for each kernel, CPU or platform, sincerely... I think I can live with it and this data doesn't affect router operations.
     
  34. Victek

    Victek Network Guru Member


    About calendar reboot, if you like to help me please don't do it, we are in beta phase and you lost opportunity to report bugs.
     
    Elfew and koitsu like this.
  35. Elfew

    Elfew Addicted to LI Member

    Planiva please think about yourself and your words in your post...
     
  36. eahm

    eahm LI Guru Member

    When the drivers is set to USA the router is much more responsive, pages open quickly and everything is faster. Don't really know why, maybe the DNS lookup time changes?
     
  37. Victek

    Victek Network Guru Member


    eahm .... If your wife is happy don't touch anything else and don't ask questions or doubts .. You make her day, let it be. ;)
    Now seriously... these figures makes me ask you three questions:
    a) For wireless ... Did you fixed a channel or it's in auto mode?
    b) Sometimes a signal excess provokes retries in wireless stream, the rates are a good information but also the information about signal error (retries) helps to find the right setting. Retries or transmission error using SG or USA country?. It may explain the responsiveness when you select USA country... (one reason), power without control is = null.
    c) All these test are 2.4 or 5 GHz band? Channel width 20 or 40MHz?

    Thanks for helping!
     
  38. eahm

    eahm LI Guru Member

    Sorry I forgot to say. The WiFi test, with the T510, was done in 5GHz at 40MHz and the different response I noticed was in my fixed workstation, with a cable.
     
  39. Planiwa

    Planiwa LI Guru Member

    We each contribute according to our own ability. My car's fuel gauge and odometer do not affect the engines's operation. Yet, I do prefer them to give accurate readings.
     
  40. Victek

    Victek Network Guru Member

    Good information, thanks!
    Well, Tomato is a multi router model firmware, if we made just for one model as your car it can be precise and accurate as your fuel gauge or odometer. Since we don't have access to the source code for drivers and vital part of firmware it's difficult to be accurate.

    Edit: I saw this post, http://www.linksysinfo.org/index.php?threads/save-button-not-working.68777/#post-230863 . Really surprises me, Why a free third part firmware deserves this comment? ... if you are interested to get accurate results for your particular router you can order and pay a custom version for it same as you paid for your car. I can try to adjust kernel values for your model. ;) .. Enjoy your day :D
     
    jerrm, Elfew, 56kb and 3 others like this.
  41. Elfew

    Elfew Addicted to LI Member

    @Planiwa - it is only cosmetic stuff, it wont bring better stability or speed...

    I think Victek knows better than you, and it is his firmware... if you wanna these changes just build your own firmware, sources are public available.
     
  42. RMerlin

    RMerlin Network Guru Member

    One man with only limited time to work on the car has to either fix the flat tire, or replace the odometer.

    Priorities.
     
    Toastman, Elfew, Victek and 1 other person like this.
  43. asammar

    asammar Network Guru Member

    Victek,

    I noticed that there are no beta build for the E2000 or the Belkin F7D4301, which are the 2 routers I have right now. Is it easy enough for me to build a load from source for these 2? Are they going to be supported in the final release (pretty please)? :)

    As always, thank you to all the Tomato developers. Without you guys, none of this would be possible!
     
  44. Victek

    Victek Network Guru Member

    asammar, I plan to release firmware for other models at the end of beta test. Also you can build your own version using the source code provided, might be possible that some modules will not fit due to flash ram capacity.

    Thanks!
     
  45. lefty

    lefty Networkin' Nut Member

    Thanks for the great tomato build Victek, it is working fine for my E2000 and E1550, atleast basic features, haven't tried anything else for now, but wanted to let you know is working fine for me on those units. :)
     
  46. Victek

    Victek Network Guru Member

    lefty, thanks for the info, much appreciated.
     
  47. gffmac

    gffmac Serious Server Member

    Just a quick one.. is the 'Use dnscrypt-proxy' option only applicable when using OpenDNS?

    Thanks
     
  48. Victek

    Victek Network Guru Member

    No, you can type additional dns, then the request will be routed by the strict order in resolv.conf file .. look at log file in tomato, you can see the available dns (127.0.0.1:40 is dnscrypto-proxy route to OpenDNS).
    In your PC if windows type ipconfig /all and you can see the used dns..
     
  49. macgyver

    macgyver Reformed Router Member

    RT-N66U results with build "WX":
    mac addresses for both wireless interfaces were wrong...once I clicked "default" they went to the proper values...also noticed I had a different WAN IP address using MerlinWRT so Im not sure what exactly is happening...I only noticed after checking the mac address field when I saw my WAN IP was different...TX power is not working for 5GHz band at all..with stock FW I get -23dbm for 2.4GHz -31dbm on 5GHz, using tomato its -23dbm for 2.4 and -42dbm for 5GHz...another thing I noticed is the output power on 5GHz changes for a brief second then goes back down..its a big difference -11dbm I had better 5GHz coverage on E3000:eek:
     
  50. Victek

    Victek Network Guru Member

    Glad to read you again...
    Changes in MAC address might happen if you was using another build, just upgrade erasing nvram.
    Different WAN IP might change also, then your ISP can assign another IP...
    I limited power for 5GHz in this release after some tests (You can see that value for power it's between 0-400), if you like noise power then you can unlock by changing value in nvram table. More power doesn't mean higher wireless speed, you might encounter a lot of retries. In any case here is the key value to let you experiment:

    default. pa0maxpwr=400
    change. pa0maxpwr=500

    Other values to experiment...
    root@RT-N66:/tmp/home/root# nvram set wl0_country_code="SG"
    root@RT-N66:/tmp/home/root# nvram set wl1_country_code="SG"
    root@RT-N66:/tmp/home/root# nvram commit

    Thanks!
     
  51. macgyver

    macgyver Reformed Router Member

    I just picked up the N66 yesterday for testing...out of the box I updated to merlins RT-N66U_3.0.0.4_372.30_3 build then went to recovery mode, erased nvram then loaded wx build...I will flash it again now and reconfigure it to see but it seems like the power is not holding whatever variable it is set to..2.4GHz is giving same dbm on both firmware but 5GHz seems to be not holding..will do some experimenting now and see with new toy :D
     
  52. Victek

    Victek Network Guru Member

    Enjoy the toy ;) .. serious toy by the way !
     
  53. macgyver

    macgyver Reformed Router Member

    clearly tx power working on 2.4 band..havent re-flashed yet but here are the results so far from a different test location:
    0mw -28dbm
    5mw -44dbm
    40mw -35dbm
    80mw -32dbm
    120mw -30dbm
    160mw -28dbm

    5GHz band makes no difference...hopefully re-flashing does something...I have both countries set for US right now and when I changed to singapore it went from -51 to -55dbm but nothing changed when I try changing tx power

    Edit:
    just re-flashed with "wx" build and MAC addresses end like this:
    Router LAN MAC............. F:68
    WAN Port.......................F:69
    WLeth1..........................F:68
    WLeth2..........................F:6C

    When I click "default" they change to:
    Router LAN MAC............. F:68
    WAN Port.......................F:69
    WLeth1..........................F:6A
    WLeth2..........................F:6B
     
  54. Victek

    Victek Network Guru Member

    Good, as I said, try to get the best transmission speed independently of the power. I'm sure you will get worst statistics (collisions and retries) when the power increases....

    For 5GHz probably the driver don't responds to power change. The digit you type in window power is not equivalent to mW, it's just a numeric value without any correspondence to mW since EIRP can't be measured by the RSSI and the sensitivity of your wireless card. Your values are a good reference for your structure but can't be compared with any other user structure.

    MAC address are ok for me, I got same range as you. Have to check this part of the code and why it assigns same MAC for LAN and WLAN1, I'll check also in E4200 since it shares similar structure. It's a bug but I don't know if it's shared with other models using Tomato.
     
  55. macgyver

    macgyver Reformed Router Member

    I just flashed to merlin build to see and with that firmware its showing same MAC addresses on BR0, ETH0, ETH1, and VLAN1 all ending in F:68 and the only one different is for 5GHz which is F:6C:confused:
    This also explains the IP change on WAN port....using tomato my WAN MAC ends in :69, with merlin build ends in .68
     
  56. Elfew

    Elfew Addicted to LI Member

    I had this.problem.with different mac adresses too, but it was with first beta versions... (I had this issue with Shibby mod.too). I fixed it easily - full reset, NVram reset followed by factory reset. Problem gone and since that no problem
     
  57. Victek

    Victek Network Guru Member


    Do as Elfew suggest, enter cli and type nvram erase, then type reboot and see what's going on.
     
  58. macgyver

    macgyver Reformed Router Member

    I don't know if all the MAC are supposed to be the same or not with the exception of 5GHz, but erase nvram reverts them back to this, which is almost the same as merlinWRT with the exception of WAN MAC...very confusing:confused:
    Router LAN MAC............. F:68
    WAN Port.......................F:69
    WLeth1..........................F:68
    WLeth2..........................F:6C
     
  59. Jeffers

    Jeffers Reformed Router Member

    Sorry for the late reply. I did initially try to adapt my original ddwrt settings with no success.

    Anyway - I managed to fix the issue, not sure if its a bug or not. On the E3000 the ip assigned to the WAN port shows in ifconfig as vlan2 but that ip is shown as attached to a different interface when queried via snmp - this interface doesnt have a speed exposed. I have no idea what the interface 32 relates to within ifconfig.

    For completeness I have uploaded the ifconfig and cfgmaker output to pastebin:
    http://pastebin.com/anZrDrHG

    I have replaced my actual public ip with 'my.public.ip.here'. 192.168.0.0/24 is my normal lan; 10.0.0.0/24 is the network my adsl modem uses (on the wan port).

    For what its worth, the ethernet speeds are being reported correctly in the webUI and via SNMP so seemingly no issues with the PHY driver.

    Let me know if you need to know anything else, and, again, thanks for the excellent firmware!

    Jeffers
     
  60. Victek

    Victek Network Guru Member

    Jeffers,

    In Tomato VLAN1 is LAN, VLAN2 is WAN, the IP you talk might be attached to ppp interface?... In any case I like to read that reported data to snmp are correct.

    Thanks for your test, I had no time to try it yet.

    Woud you confirm this point ' which is almost the same as merlinWRT with the exception of WAN MAC...very confusing' ... can you paste MAC using RMerlin build and Tomato RAF after nvram erase?

    I just try to see what are the differences in MAC assignment... Thanks for your help and enjoy the toy!
     
  61. macgyver

    macgyver Reformed Router Member

    Ifconfig from merlinwrt 3.0.0.4.372.30_2:
    br0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:6C
    vlan1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    vlan2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68

    If I enter MAC address during setup wizard (eth0 MAC changes):confused:
    br0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:69
    eth1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:6C
    vlan1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    vlan2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68

    been flashing all day the joys of new toys;) ...with merlinWRT if you don't specify a MAC address for WAN during the initial setup it clones the LAN MAC for WAN MAC (seems to like it a lot) :D
    vlan2 MAC stays :68 with merlinWRT and :69 with TomatoRAF

    ifconfig from TomatoRAF below:
    br0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:6C
    vlan1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    vlan2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:69

    If I click "default" beside MAC address clone they become as below but NVRAM erase reverts them to as above:
    br0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth0 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    eth1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:6A
    eth2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:6B
    vlan1 Link encap:Ethernet HWaddr xxxxxxxxxxxF:68
    vlan2 Link encap:Ethernet HWaddr xxxxxxxxxxxF:69
     
  62. asammar

    asammar Network Guru Member

    @lefty, which file did you use on the e2000 please?
     
  63. macgyver

    macgyver Reformed Router Member

    What are your MAC address assignment now? could u pls post
     
  64. lefty

    lefty Networkin' Nut Member

    The one that i built myself.
     
  65. macgyver

    macgyver Reformed Router Member

    Power comparison between MerlinWRT and TomatoRAF...all the limits are higher in MerlinWRT and 5GHz power works:
    MerlinWRT:
    Regulatory Limits:
    CCK : 25.50
    CCK CDD 1x2 : 25.50
    CCK CDD 1x3 : 25.50
    Legacy OFDM 20MHz SISO : 25.50
    Legacy OFDM 20MHz CDD : 25.50
    MCS 0-7 20MHz 1 Tx : 25.50
    MCS 0-7 20MHz 2 Tx : 25.50
    MCS 0-7 20MHz 3 Tx : 25.50
    MCS 8-15 20MHz 2 Tx : 25.50
    MCS 8-15 20MHz 3 Tx : 25.50
    MCS 16-23 20MHz 3 Tx : 25.50
    Legacy OFDM 40MHz SISO : 24.50
    Legacy OFDM 40MHz CDD : 24.50
    MCS 0-7 40MHz 1 Tx : 24.50
    MCS 0-7 40MHz 2 Tx : 24.50
    MCS 0-7 40MHz 3 Tx : 24.50
    MCS 8-15 40MHz 2 Tx : 24.50
    MCS 8-15 40MHz 3 Tx : 24.50
    MCS 16-23 40MHz 3 Tx : 24.50
    MCS 32 : 24.50
    20UL CCK : 25.50
    20UL CCK CDD 1x2 : 25.50
    20UL CCK CDD 1x3 : 25.50
    20UL Legacy OFDM : 25.50
    20UL Legacy OFDM CDD : 25.50
    MCS 0-7 20UL 1 Tx : 25.50
    MCS 0-7 20UL 2 Tx : 25.50
    MCS 0-7 20UL 3 Tx : 25.50
    MCS 8-15 20UL 2 Tx : 25.50
    MCS 8-15 20UL 3 Tx : 25.50
    MCS 16-23 20UL 3 Tx : 25.50
    Board Limits:
    CCK : 23.50
    CCK CDD 1x2 : 23.50
    CCK CDD 1x3 : 23.50
    Legacy OFDM 20MHz SISO : 22.50
    Legacy OFDM 20MHz CDD : 22.50
    MCS 0-7 20MHz 1 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20MHz 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20MHz 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 16-23 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    Legacy OFDM 40MHz SISO : 20.50
    Legacy OFDM 40MHz CDD : 20.50
    MCS 0-7 40MHz 1 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 0-7 40MHz 2 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 0-7 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 8-15 40MHz 2 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 8-15 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 16-23 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 32 : 20.50
    20UL CCK : 23.50
    20UL CCK CDD 1x2 : 23.50
    20UL CCK CDD 1x3 : 23.50
    20UL Legacy OFDM : 22.50
    20UL Legacy OFDM CDD : 22.50
    MCS 0-7 20UL 1 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20UL 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20UL 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 16-23 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    Power Target:
    CCK : 22.0
    CCK CDD 1x2 : 22.0
    CCK CDD 1x3 : 22.0
    Legacy OFDM 20MHz SISO : 21.0
    Legacy OFDM 20MHz CDD : 21.0
    MCS 0-7 20MHz 1 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 0-7 20MHz 2 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 0-7 20MHz 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 8-15 20MHz 2 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 8-15 20MHz 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 16-23 20MHz 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    Legacy OFDM 40MHz SISO : 19.0
    Legacy OFDM 40MHz CDD : 19.0
    MCS 0-7 40MHz 1 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 0-7 40MHz 2 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 0-7 40MHz 3 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 8-15 40MHz 2 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 8-15 40MHz 3 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 16-23 40MHz 3 Tx : 19.0 19.0 19.0 19.0 16.0 16.0 16.0 16.0
    MCS 32 : 19.0
    20UL CCK : 22.0
    20UL CCK CDD 1x2 : 22.0
    20UL CCK CDD 1x3 : 22.0
    20UL Legacy OFDM : 21.0
    20UL Legacy OFDM CDD : 21.0
    MCS 0-7 20UL 1 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 0-7 20UL 2 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 0-7 20UL 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 8-15 20UL 2 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 8-15 20UL 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    MCS 16-23 20UL 3 Tx : 21.0 21.0 21.0 21.0 21.0 19.0 17.50 16.0
    Maximum Power Target among all rates: 22.0 22.0 22.0
    Rate index with Maximum Power Target: 0 0 0
    Last adjusted est. power : 18.50 0.0 18.25
    ------------------------------------------------------------------------------
    TomatoRAF "WX"
    Regulatory Limits:
    CCK : 30.0
    CCK CDD 1x2 : 30.0
    CCK CDD 1x3 : 30.0
    Legacy OFDM 20MHz SISO : 30.0
    Legacy OFDM 20MHz CDD : 30.0
    MCS 0-7 20MHz 1 Tx : 19.0
    MCS 0-7 20MHz 2 Tx : 19.0
    MCS 0-7 20MHz 3 Tx : 19.0
    MCS 8-15 20MHz 2 Tx : 19.0
    MCS 8-15 20MHz 3 Tx : 19.0
    MCS 16-23 20MHz 3 Tx : 19.0
    Legacy OFDM 40MHz SISO : 16.50
    Legacy OFDM 40MHz CDD : 14.0 14.0 14.0 14.0 14.0 15.0 15.0 15.0
    MCS 0-7 40MHz 1 Tx : 16.50
    MCS 0-7 40MHz 2 Tx : 14.0 14.0 14.0 14.0 14.0 15.0 15.0 15.0
    MCS 0-7 40MHz 3 Tx : 14.0 14.0 14.0 14.0 14.0 15.0 15.0 15.0
    MCS 8-15 40MHz 2 Tx : 15.0 15.0 15.0 15.0 15.50 15.50 15.50 15.50
    MCS 8-15 40MHz 3 Tx : 15.0 15.0 15.0 15.0 15.50 15.50 15.50 15.50
    MCS 16-23 40MHz 3 Tx : 15.0 15.0 15.0 15.0 15.50 15.50 15.50 15.50
    MCS 32 : 16.50
    20UL CCK : 30.0
    20UL CCK CDD 1x2 : 30.0
    20UL CCK CDD 1x3 : 30.0
    20UL Legacy OFDM : 30.0
    20UL Legacy OFDM CDD : 30.0
    MCS 0-7 20UL 1 Tx : 19.0
    MCS 0-7 20UL 2 Tx : 19.0
    MCS 0-7 20UL 3 Tx : 19.0
    MCS 8-15 20UL 2 Tx : 19.0
    MCS 8-15 20UL 3 Tx : 19.0
    MCS 16-23 20UL 3 Tx : 19.0
    Board Limits:
    CCK : 23.50
    CCK CDD 1x2 : 23.50
    CCK CDD 1x3 : 23.50
    Legacy OFDM 20MHz SISO : 22.50
    Legacy OFDM 20MHz CDD : 22.50
    MCS 0-7 20MHz 1 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20MHz 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20MHz 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 16-23 20MHz 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    Legacy OFDM 40MHz SISO : 20.50
    Legacy OFDM 40MHz CDD : 20.50
    MCS 0-7 40MHz 1 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 0-7 40MHz 2 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 0-7 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 8-15 40MHz 2 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 8-15 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 16-23 40MHz 3 Tx : 20.50 20.50 20.50 20.50 17.50 17.50 17.50 17.50
    MCS 32 : 20.50
    20UL CCK : 23.50
    20UL CCK CDD 1x2 : 23.50
    20UL CCK CDD 1x3 : 23.50
    20UL Legacy OFDM : 22.50
    20UL Legacy OFDM CDD : 22.50
    MCS 0-7 20UL 1 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20UL 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 0-7 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20UL 2 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 8-15 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    MCS 16-23 20UL 3 Tx : 22.50 22.50 22.50 22.50 22.50 20.50 19.0 17.50
    Power Target:
    CCK : 22.0
    CCK CDD 1x2 : 22.0
    CCK CDD 1x3 : 22.0
    Legacy OFDM 20MHz SISO : 21.0
    Legacy OFDM 20MHz CDD : 21.0
    MCS 0-7 20MHz 1 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 0-7 20MHz 2 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 0-7 20MHz 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 8-15 20MHz 2 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 8-15 20MHz 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 16-23 20MHz 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    Legacy OFDM 40MHz SISO : 15.0
    Legacy OFDM 40MHz CDD : 12.50 12.50 12.50 12.50 12.50 12.50 13.50 13.50
    MCS 0-7 40MHz 1 Tx : 15.0
    MCS 0-7 40MHz 2 Tx : 12.50 12.50 12.50 12.50 12.50 13.50 13.50 13.50
    MCS 0-7 40MHz 3 Tx : 12.50 12.50 12.50 12.50 12.50 13.50 13.50 13.50
    MCS 8-15 40MHz 2 Tx : 13.50 13.50 13.50 13.50 14.0 14.0 14.0 14.0
    MCS 8-15 40MHz 3 Tx : 13.50 13.50 13.50 13.50 14.0 14.0 14.0 14.0
    MCS 16-23 40MHz 3 Tx : 13.50 13.50 13.50 13.50 14.0 14.0 14.0 14.0
    MCS 32 : 15.0
    20UL CCK : 22.0
    20UL CCK CDD 1x2 : 22.0
    20UL CCK CDD 1x3 : 22.0
    20UL Legacy OFDM : 21.0
    20UL Legacy OFDM CDD : 21.0
    MCS 0-7 20UL 1 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 0-7 20UL 2 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 0-7 20UL 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 8-15 20UL 2 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 8-15 20UL 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    MCS 16-23 20UL 3 Tx : 17.50 17.50 17.50 17.50 17.50 17.50 17.50 16.0
    Maximum Power Target among all rates: 22.0 22.0 22.0
    Rate index with Maximum Power Target: 0 0 0
    Last adjusted est. power : 22.0 22.0 22.0
     
  66. Victek

    Victek Network Guru Member

    Thanks for the info macgyver, about MAC:
    The merge of various builds and flashing can corrupt nvram but it seems weird the MAC assignment in asuswrtmerlin firmware, I think the best solution is as Elfew pointed, nvram erase, reset, and then configure. My MAC details...

    LAN Router MAC Address xx:xx:xx:01:07:88
    WAN MAC Address xx:xx:xx:01:07:89
    WLAN 1 (eth1) MAC Address xx:xx:xx:01:07:8A
    WLAN2 (eth2) MAC Address xx:xx:xx:01:07:8B

    About 5GHz settings, well, it's build depending and if Rmerlin version shows it then is ok, again, more power means more coverage but no relation with reliable data communication. Try both versions (and others) then select the best for your need. ;)
     
  67. emporium

    emporium Reformed Router Member

    Nowadays I can't feedback because I'm using an old stable tomato RAF version, and it doesn't support dnscrypto-proxy. I'm waiting for the 1.2 stable version.

    I don't think dnscrypto-proxy and iodine has the same funcionality.
    AFAIK dnscrypto-proxy works encrypting all DNS traffic. But it only encrypts DNS traffic.
    Iodine lets you tunnel all ipv4 data encapsulated as DNS traffic, so you can use internet or access to your router or servers through a connection that only permits DNS traffic.

    I don't know much about dnscrypto-proxy, but I think it has a different goal.

    Let me know if I am wrong, because I'm not an expert.

    I'd like to have iodine as command in cli, it's not necessary to be in the gui.

    Thanks.
     
  68. Victek

    Victek Network Guru Member

    ;) try the last beta, it's near release candidate and then you can feedback your findings. I'll take a look after your feedback. ;)
     
  69. Elfew

    Elfew Addicted to LI Member

    This is confusing... and I am not alone. WAN should be WAN, LAN should be LAN1-4 etc. At least add note below the device list window.

    But it is offtopic now, what about MAC addreses? Is it working for you now? I solved this isssue by complete reset 30/30/30, NVram reset and fresh flash.
     
  70. RMerlin

    RMerlin Network Guru Member

    Asus has done a lot of tweaking to output power a few months ago. Keep in mind that their settings might be more fine-tuned to the RT-N66U, while Tomato has to share the same settings across a wide range of routers.
     
  71. Victek

    Victek Network Guru Member


    Will do, easy request ;) .. but unfortunately it's not the same for all models... and tomato versions, VLAN build differs. As RMerlin said, we have to make it compatible with many models now...
     
  72. Elfew

    Elfew Addicted to LI Member

    I understand... it is only your decision. It is only a minor change and I think there is a problem with my solution which I posted you, am I right? Because it is different for every model :(

    @RMerlin - interesting, and what are the results of this improvement? Better stability or higher throughput?
     
  73. RMerlin

    RMerlin Network Guru Member

    No idea. Too many factors changing at the same time, so only their engineers would know.
     
  74. macgyver

    macgyver Reformed Router Member


    I always clear nvram before and after flashing...I did it through the recovery webpage @192.168.1.1 prior to loading new firmware; I didnt use any special software only what is on the router when in rescue mode...right now I can test whatever needs to be tested as I took the router offline for troubleshooting purposes...I will start again from scratch with pleasure...all in the name of advancement ;)
    In this case ...more power means signal or no signal but that is not really an issue in the big picture once the bugs are sorted
    Im willing to compare the code and help in any way possible...first I will sort the MAC issue then move on to further testing...I have only had this router for a day so still learning tricks...
    Handy trick to know if the recovery mode doesnt work from holding reset button then power on is to hold down WPS button and power on, then release WPS button after 10 seconds...then when all lights flash on router press and hold reset button until Power LED starts flashing...then u will have recovery mode!....I thought I bricked the router after flashing DDWRT and that was the only way to get back in to recovery mode...on to more testing...Teamwork always pays off....more than happy to assist all I can:D
     
  75. macgyver

    macgyver Reformed Router Member

    Hi Merlin and thanks for your contributions to the router world....I noticed with the latest 370.2 firmware if I hold down reset then power on, recovery mode lights flash to indicate rescue mode but there is no access to minicfe server unless hard reset of the router first, but using tomatoRAF I have access to it by pressing button
     
  76. macgyver

    macgyver Reformed Router Member

    I have done nvram erase, reset, flash, reset, reset and the values are still off for WLAN1 and WLAN2 unless I click "default" then they change to how you have them....otherwise as in my posts above...I had a feeling I was clearing nvram properly but through telnet, minicfe server, WPS button or reset button or all of those 1 after another still results in same thing...Im at a loss but to me something seems wrong with the MAC assignments in the tomato code as DDWRT assigns the proper MACs according to CFE...in my CFE it shows:
    et0macaddr=xxxxxxxxxx3F:68
    pci/1/1/macaddr=xxxxxx3F:68
    pci/2/1/macaddr=xxxxxx3F:6C
    pmon_ver=CFE 5.100.127 odmpid=ASUS bl_version=1.0.1.3

    Ideally (and logically) it should be the way you listed your MAC addresses..I want to edit mine knowing this now as it just annoys me...also wondering if your CFE matches the same values? I was expecting to see this:
    et0macaddr=xxxxxxxxxx3F:68
    pci/1/1/macaddr=xxxxxx3F:6A
    pci/2/1/macaddr=xxxxxx3F:6B
     
  77. Victek

    Victek Network Guru Member

    Well, do it as I did, press default in MAC gui, save in config file and done. I agree it's a 'mild' bug to be investigated not only for RT-N66U, also for other models. ;) but It doesn't affect router operations, priorities and bugs first.
     
  78. RMerlin

    RMerlin Network Guru Member


    The firmware being run shouldn't have any effect, since the CFE is totally separate from the FW itself, and is run before booting the FW.
     
  79. JAC70

    JAC70 Networkin' Nut Member

    Hi, and thanks for the reply. I did as you suggested, but unfortunately WAN speed remained the same.

    I verified the command took, with nvram show | grep ctf_ and it confirmed ctf_disable=0.

    Even overclocking the router CPU had no effect, so it doesn't appear to be the CPU limiting the newest firmware.

    Thanks.
     
  80. RMerlin

    RMerlin Network Guru Member


    Check that CTF actually gets loaded:

    Code:
    lsmod | grep ctf
    
    If nothing comes up, then it means hardware acceleration isn't loaded and working.
     
  81. Elfew

    Elfew Addicted to LI Member

    CTF is not working in Tomato... There was a GUI for this but it didnt work... I think this GUI is still in Toastman builds under Misc. tab. I think it has never been working in Tomato -> It has been removed from GUI by Teddy Bear from his builds.

    check this:
    http://tomatousb.org/forum/t-343748
     
  82. JAC70

    JAC70 Networkin' Nut Member

  83. Victek

    Victek Network Guru Member

    Many parts of the code (QoS, BW limiter) disables ctf module even you try to enable. It's not build in RAF version (even as a module), but it's compiled and available as extra module in the build for people interested to include it in their build (ctf.ko). I'll do some test with wideband network to check if we can enable when other modules are also disabled...
     
    eviltone and Elfew like this.
  84. macgyver

    macgyver Reformed Router Member

    I pressed default but at the same time had to get to the bottom of it...turns out it reads the MACs from the CFE and the asus/merlinWRT assumes that eth1 is same MAC as LAN (it clearly doesnt read the CFE for that because I changed the value and eth1 MAC is still the same as LAN MAC after FW loads, but for eth2 it read the MAC as I had in CFE)..I edited the CFE and changed the MACs to end in A and B, now tomato is reading the MACs as the CFE has them, and I don't have to press default in MAC GUI. Also DDWRT assigns the MACs correctly...works for me on to bigger bugs:D
     
  85. Victek

    Victek Network Guru Member

    You modified the CFE? :eek:
     
  86. macgyver

    macgyver Reformed Router Member

    I modified the offset 65D for eth1 MAC and changed from "8" to "A" and offset B4C for eth2 MAC and changed from "C" to "B"...I put the original CFE back on afterwards all is well
    I did CFE edit on my 610v2 years ago to convert to E3000 and also modified default clocks at same time...all the routers still running strong at 532/266/133 but can't compare E3000 to RTN66 except for 5GHz wireless signal strength which is sad in its own way but thats a bug with 5GHz power level adjustment:(
     
  87. Victek

    Victek Network Guru Member

    ;) I did the same on my WRT610N..... before dying.. How it runs now?.

    By the way, yesterday I lost 5GHz band in RT-N66U after trying one dd-wrt version. Led was off and MAC for eth2 (wl1) was empty, erase nvram didn't help, WPS and power up nothing, fill new MAC for 5GHz .. nothing.

    I solved after restoring a previous configuration I had for the router when both bands was working.
     
  88. macgyver

    macgyver Reformed Router Member

    I have around 40 E3000 in the field all of them overclocked with CFE mod for years now and none are dead...I must be lucky :eek:
    I had a similar issue flashing DDWRT and the SSID would come up for a few seconds but unable to connect to it...I had to get into recovery mode as I outlined a few posts earlier...the usual way didnt work and I thought the router was bricked...
    One thing I noticed with MAC address on TomatoRAF when you click "default" for eth2, it changes the 2nd value for some reason...for example 00:00:00:00:00:0C becomes 02:00:00:00:00:0B and the actual mac has neither the 2 or the B to begin with in the CFE ...this happens on RT-N66U and E3000, but on E3000 it only changes the 2nd value, on RT-N66U it changes the 2nd and last values:confused:
    curious if you had clicked "default" which would have changed your MAC to be x:2 instead of :x0 for 5GHz band, and if that could have had anything to do with it...Tomato seems to think all eth2 mac address the 2nd value is a 2 if you click "default"....could that have anything to do with things?
     
  89. Victek

    Victek Network Guru Member

    I killed WRT610N when I was soldering a new RAM for 128MB ...

    About the issue, it happened after a battle firmware comparison, kong firmware, merlin, toastman, shibby, dd-wrt, and then the 5GHz led was killed in last dd-wrt test (some nvram values are missed). The base values for mac are in the CFE .. but Tomato expand these value when it reserves MAC for virtual WLAN (also for pci/2/1 5GHz radio) but in the reboot it shows eth%d (no device found) as pci/2/1 instead of eth2...
    Only restoring a previous backup realive 5GHz in my case... I have to check what values was restored or created again.
     
  90. Mysteron

    Mysteron Networkin' Nut Member

    @Victek

    I had one of your old builds on my Asus WL-500GPv2. Are any of the new (2013) builds compatible with this router?

    Thanks.
     
  91. macgyver

    macgyver Reformed Router Member

    bug in port forwarding...UPnP/NAT-PMP tab
    If I untick either "Enable UPnP" or "Enable NAT-PMP" it says "NAT-PMP or UPnP must be enabled in at least 1 LAN bridge" and by default there is no tick in any LAN box but its enabled:confused: , and to disable UPnP I have to tick the "Listen on LAN" box, and it also allows both "Enable UPnP" and "Enable NAT-PMP" boxes unticked which is contradictory to what the message says
     
  92. Victek

    Victek Network Guru Member


    No Mysteron, new builds are not compatible with your router, sorry for that.


    What version do you use?, I checked and I have LAN enable .... did you restored previous settings in this new release? I'm using now wx release.

    Edit: Ok, found the bug... creating new nvram values when VLAN switch is ON. New commit in the git.
     
    Elfew likes this.
  93. Elfew

    Elfew Addicted to LI Member

    I noticed this bug too, thx for fix :)
     
  94. Victek

    Victek Network Guru Member

  95. M0g13r

    M0g13r LI Guru Member

    webserver / nginx gui
    its buggy in IE8 too :\
    and dont say to upgrade my browser :) i dont use IE .... if i found some IE related things ..... i found it at work :)

    nginx_gui.JPG
     
  96. Victek

    Victek Network Guru Member

    Ok, as you said IE is buggy.. but nevertheless I'll try to solve this incompatibility issue too. Thanks! ;) , To all of you, please use Firefox, Opera, Chrome ...
     
  97. Elfew

    Elfew Addicted to LI Member

  98. Victek

    Victek Network Guru Member

    Elfew likes this.
  99. BlaSTiWi

    BlaSTiWi Network Guru Member

    Victek,

    I'm using E3000 + v1.1w beta ... while configuring I noticed JFFS section section showed 64K and got error everytime I enable it:

    Jul 27 15:46:00 router user.info init[1]: notice[jffs]: Error mounting JFFS. Check the logs to see if they contain more details about this error.
    Jul 27 15:46:00 router user.info kernel: JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
    Jul 27 15:46:00 router user.err kernel: jffs2: Too few erase blocks (1)
    Jul 27 15:48:19 router user.info init[1]: notice[jffs]: Stopped

    Any pointer would be great ... tkx much in adv.!
     
  100. lefty

    lefty Networkin' Nut Member

    As noted in the log, too few erase blocks - meaning not enough flash space left to use any kind of JFFS. Only other pointer from here would be to either use the E3000's USB port for external storage or use some kind of other tomato mod that is a smaller image and would leave you enough space left over that you could use JFFS. This mod image for the E3000 is already trimmed down as much as it could be.
     

Share This Page