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

Toastman RT-N Tomato Firmware on Asus RT-N66U 'Dark Knight' Dual Band Wireless-N900 Gigabit Router

Discussion in 'Tomato Firmware' started by though, Feb 24, 2012.

  1. though

    though Network Guru Member

    Figured it would be a good idea to start an official thread discussing Toastman's tomato on this router. Please keep all messages related to this router & firmware in this thread.
     
  2. ArCan

    ArCan Connected Client Member

    Hi!
    I don't need any options BT etc. and I'm using right now the
    tomato-K26USB-1.28.0495.3MIPSR2-Toastman-VLAN-RT-N-VPN.trx
    I like it very much!
    Thank you Toastman!

    P.S. I'm watting very much for firmware from Toastman special for RT-N66U's chipset.
    I'm not programmer but somebody told that only firmware under new SDK
    will open all futures new RT-N66U. I don't know how it right and true.
     
  3. Planiwa

    Planiwa LI Guru Member

    It may not be too late for the originator of this thread to edit the thread origin so as to state very clearly whether there actually exists a Toastman firmware for this router, and if so, exactly where it can be found.
     
  4. though

    though Network Guru Member

  5. Planiwa

    Planiwa LI Guru Member

  6. Planiwa

    Planiwa LI Guru Member

    I am considering getting an N66U in the hope that 5GHz will overcome the severe wireless interference from numerous strong N routers in the neighbourhood.

    Is it safe to assume that a device less than 3 years old will be able to handle 5GHz N?
     
  7. though

    though Network Guru Member

    no. the N16 uses the RT release, not the RT-N release.
     
  8. though

    though Network Guru Member

    absolutely not safe to assume that.
     
  9. kthaddock

    kthaddock LI Guru Member

    RT-N build also cover RT-N16 with newer wifi drivers, so both can bee used by RT-N16.
     
  10. Planiwa

    Planiwa LI Guru Member

  11. Planiwa

    Planiwa LI Guru Member

    Does that mean that RT-N might be a better choice for RT-N16, wifi-wise?
     
  12. Planiwa

    Planiwa LI Guru Member

    So, is it safe to assume that if a client is "N"-capable, that it will be able to handle 5GHz?
     
  13. eahm

    eahm Addicted to LI Member

    Not necessarily a better choice but most of the times new is better so just test and let us know which one is better for you.

    5Ghz is 802.11a btw.

    What do you mean with safe? Just check the specs of the device, it will tell you if it handles 5Ghz bands (or 802.11a).
     
  14. ArCan

    ArCan Connected Client Member

    Hi!
    I'm using tomato-K26USB-1.28.0495.3MIPSR2-Toastman-VLAN-RT-N-VPN.trx with RT-N66U.
    All works fine except Scheduler. It can do Reboot one time but not All Days.
    For example, if I check All Days it will reboot router only one time at the first.
    But after that router works without rebooting.
    I tryed check Custom setting with Reboot command and checked All Days with same results.
    By the way (!) same situation was with Shibby's firmware. I really don't know why.
    May be something wron with my configuration or even with my router...
     
  15. Victek

    Victek Network Guru Member

    No, all of us use the same source, I can't reproduce in RT-N66, In my RT-N16 I tested and it works fine.. we need more info about RT-N66, shibby have one unit I think.
     
  16. Mirko Baila

    Mirko Baila Reformed Router Member

    Try to disable USB support,reboot router,rehabilitation USB support once started
     
  17. though

    though Network Guru Member

  18. shibby20

    shibby20 LI Guru Member

    leave Auto mode.
     
  19. rhdcheme

    rhdcheme Networkin' Nut Member

    Did you try multiSSID on your RT-N66U? I did not try 0495.3, but it does not seem to work in 0496.2 (does not save guest network in "Overview" tab of Virtual Wireless).
     
  20. ladysman

    ladysman LI Guru Member

    I also can't get it to work. I haven't had time to mess with it since last weekend. See the "Multiple SSID" thread
     
  21. rhdcheme

    rhdcheme Networkin' Nut Member

    I got it to work. According to teaman (in a thread regarding a similar problem for an E3000), you have to save the Virtual Wireless before changing the Wireless Auto Mode in Basic Network. I did just that and it took the "Save" in Advanced Virtual wireless Overview. Then I changed the Auto Mode to N only.

    However, I am finding the wireless is much slower than my E4200. I experienced slow fluctuating wireless speeds in the stock firmware as well, so I am wondering if I got a lemon. I do have a lot of neighborhood interference, though.
     
  22. ladysman

    ladysman LI Guru Member

    I'll have to look but i'm pretty sure I never changed the Wireless Auto Mode in basic network. I left it at auto.
     
  23. hk2k2

    hk2k2 Network Newbie Member

    Hello! I´ve also bought the RT-N66U Router now, and flashed tomato - toastman mod.

    i´ve also tried flashing 60k nvram version, and seems to work fine.

    anyone can confirm this too ?

    i´ve added lot of stuff, to fill up the nvram up to 56k...

    no problem at all, but now i only use 27k again.

    what´s the safest way to check if this works 100% without brick the router?

    fill up and nvram ? then doing a reboot?

    what´s the worst case? any jtag option available for recovering ? (i have jtag cables..)

    thanks ,br hk2k2
     
  24. kthaddock

    kthaddock LI Guru Member

    Next time you filled NVRAM up run this comand and see space: "nvram show |grep entries"
    kthaddock
     
  25. rhdcheme

    rhdcheme Networkin' Nut Member

    Yes, I tried both versions and they seem to work. However, not sure which one should be used or is the recommended version for this router. Anyone know?
     
  26. Toastman

    Toastman Super Moderator Staff Member Member

    Please use only the version that is designed for your router, whatever the model. You can't use the 60K version on routers that do not have 60K of NVRAM. That should be rather obvious.

    Now, please don't try to be smart and keep asking whether this only applies to model XXX router.

    If your router - ANY MAKE, ANY MODEL - does not have 60K of NVRAM then you can't use the 60K version of the software, OK? It's not negotiable.

    You can't stick the wheel of a huge truck on your toyota either, can you? What's the problem here?

    Please read through the following threads before you dice with the death of your router.

    http://tomatousb.org/forum/t-284878/increasing-nvram-size
    http://tomatousb.org/forum/t-423607
    http://www.linksysinfo.org/index.php?threads/asus-rt-n16-bricked-using-60k-nvram-version.37025/



    Arcan - this is the reason why Gigabit ethernet setting is not, and should not, be part of that menu:

    http://www.homelabnetwork.com/learning/do-not-force-gigabit-ethernet-auto-negotiation-to-1000full/
     
  27. rhdcheme

    rhdcheme Networkin' Nut Member

    Thanks for the reply. Under your download database, I could not find any specific build for RT-N66U, even though the main folder seems to indicate that RT-N66u is supported. So just to clarify, is the version below the version that should be flashed on the RT-N66U? Or is the RT-N66u not supported yet? Thanks.

    tomato-K26USB-1.28.0496.2MIPSR2-Toastman-VLAN-RT-N-Ext
     
  28. hk2k2

    hk2k2 Network Newbie Member

    thans for reply,

    @ toastman, your linked thread is about RT-N16-not 66! till now i haven´t found a thread about a bricked RT-N66U caused by 60k nvram.

    and on your download page, there is no specific RT-N66U version.

    @kthaddock,

    last time i filled up i´ve already done this, to be sure :)

    but i will try again NOW:

    here nvram show | grep entires BEFORE i filled up stuff:
    Code:
    root@master:/tmp/home/root# nvram show | grep entries
    1194 entries, 27451 bytes used, 33989 bytes free.
    
    and now filled up with stuff:
    Code:
    root@master:/tmp/home/root# nvram show | grep entries
    1195 entries, 52889 bytes used, 8551 bytes free.
    
    then - NO RISK NO FUN- shutdown my router.

    remove Powersupply, wait some seconds, and reboot it.

    voila!
    still ~53kB used :)

    i´ve dumped the mtdblock4, removed all FF bytes, and about 48k are used.

    Code:
    Tomato v1.28.0496 MIPSR2-Toastman-VLAN-RT-N K26 USB VPN
    root@master:/tmp/home/root# nvram show | grep entries
    1194 entries, 52855 bytes used, 8585 bytes free.
    root@master:/tmp/home/root#
    
    router survives a reboot with filled nvram? so is this a safe confirmation that 60k NVRAM images works fine on RT-N66U ?

    ;) hope this helps someone....

    thanks, br hk2k2
     
  29. kthaddock

    kthaddock LI Guru Member

    @hk2k2
    Thanks for the test. People think that was risky, but seems to work. I'm a chicken ! :eek:
     
  30. MacJohnMcC

    MacJohnMcC Network Newbie Member

    I would love to try out Tomato on my RT-N66U but I get an error:

    "Firmware upgrade fail. It may result from incorrect image or error transmission. Please check the version of firmware and try again."

    My routers current firmware version is the ASUS 3.0.0.3.108.

    Is there a trick I'm missing? I've tried downloading the file with both Chrome and Firefox on Mac OS X 10.7.3.
     
  31. fubdap

    fubdap Networkin' Nut Member

    Did you try this procedure:
    http://www.shadowandy.net/2012/03/asus-rt-n66u-tomatousb-firmware-flashing-guide.htm
     
  32. Toastman

    Toastman Super Moderator Staff Member Member

    FLASHING 60K builds on 32K routers

    Just a warning to everyone.

    As forum moderator, I have considered whether to delete the posts recommending that others do something that I know will brick their router, but I decided against it.

    You do this at your own risk. I do not recommend it, and I would hope that people here do not recommend that others try it. If you decide to do so, please read through the following thread on tomatousb.org regarding the attempt to enable more than 32K of nvram, and try to understand why it is fraught with danger. I would consider RAY123 to be the best authority that we have on this subject and it might be wise to listen to him. Please note that the attempt was not ultimately successful and was removed from TomatoUSB, so there is no auto-detection of NVRAM size, it is hard-coded in the cfe.

    http://tomatousb.org/forum/t-284878/increasing-nvram-size

    While you may think you are getting something for nothing by flashing the 60K versions on routers with only 32K, one day when the cfe tries to read something that isn't in the right place, one of two things may happen. If you're very, very lucky, it may reset using defaults, or, more likely, it will brick. And once it bricks, much of the time the serial port doesn't work any more and the only way to recover is with JTAG.

    And many of the newer routers aren't supported yet by JTAG ..... so it's only good for a doorstop.

    The RT-N66U cfe is designed to work with 32K of NVRAM, just the same as the RT-N16U and most other routers apart from the Linksys series. When your router ends up in the skip don't say you weren't warned.


    Please don't keep asking if I "think" this or that - you can't "negotiate" this, it doesn't matter a damn what anybody "thinks".

    Also, stop saying that ASUS or LINKSYS or anyone should help - they don't and they won't. It's pointless posting stuff like this.

    And now it's time to lay this to rest, as the complaints have already started pouring in from people with bricked routers who have been listening to the bad advice, and I'm beginning to get very weary of this topic and the resulting mail asking me to help fix it. Sorry, I won't. You did what you were told very clearly was a bad idea, and that's your own fault.

    NB - There is no specific (named) version for this router because it is supported by the normal build.
     
  33. hk2k2

    hk2k2 Network Newbie Member

    hm ok, so you think even with my tested reboots with 56k used nvram, it could happen after a while that cfe will try to read something from wrong adress ?

    too bad :) or is there a small chance, that it will work normally :)

    when CFE will read something in nvram? on normal reboot there was no problem at all..

    why they just dont offer full 128k flash partition size for nvram :)

    maybe asus will help *G*

    br
     
  34. mstombs

    mstombs Network Guru Member

    Can someone explain hk2k2's results on the 60K firmware on RT-N66U? Would be good to see serial console dump of what CFE reports during boot.

    Asus did leak source-code for a version of the RT-N16 CFE, which makes it clear that they used Broadcom's historical default NVRAM_SPACE occupying the last 32kB of the flash chip, the GPL router sourcecode for the N66 uses the same. I do not know why they don't just use the whole of the last flash block - since it all gets written everytime nvram is committed. I have experimented with 128kB in CFE (on WRT54G-TM) and sourcecode before and it did work, for a while. There must be a reason why Linksys use 60 rather than 64 which has been the minimum erase block size on most common flash chips (128 or 256 also used).

    The router source code (shared by CFE) does search for the start of the nvram in a number of places, one of which matches the 60k location - it looks for the token:-

    Code:
    #define NVRAM_MAGIC 0x48534C46 /* 'FLSH' */
    BUT I have not seen anything that suggests that the size of the nvram is also auto-detected by the CFE, NVRAM_SPACE is a compile #define, used in many places - which is set differently by Linksys (calculated at compile time in tomatousb to support the different sizes).
     
  35. Planiwa

    Planiwa LI Guru Member

    Yes.
    Code:
        </dev/mtd4 wc
          63      672    131072
     
        </dev/mtd4 tr '\377' '\n' |wc
      106239      672    131072
     
        </dev/mtd4 tr '\377' '\n' |grep -n -v '^$' |head
    98305:FLSH@a
    98307:åZ*&rrule0=0|1320|300|31|||word text
    98308:^begins-with.domain.
    98309:.ends-with.net$
    98310:^www.exact-domain.net$|0|example
    98311:wl_mac_deny=
    98312:wl_radius_port=1812
    98313:sb/1/ofdm2gpo=0x66666666
    98314:pptp_client_mru=1450
    98315:https_crt=
     
        nvram show --nosort |head -5  ### (NB: this sequence is not fixed!)
    rrule0=0|1320|300|31|||word text ^begins-with.domain. .ends-with.net$ ^www.exact-domain.net$|0|example
    wl_mac_deny=
    wl_radius_port=1812
    sb/1/ofdm2gpo=0x66666666
    pptp_client_mru=1450
    
    I've been meaning to start a conversation about NVRAM towards understanding the relationship between the actual NVRAM, the working NVRAM image, and the various processes that interact with (mostly) the latter.

    (I'll probably get to it after the wireless -- wl thread :))

    FWIW, to get the size of the NVRAM image:
    Code:
    nvram show|tail -1
    [above commands were on an N16 -- others may wish to run on N66U, compare and report.]
     
  36. ArCan

    ArCan Connected Client Member

    Do somebody know something about DNSCRYPT in Toastman firmware?
    Will it able in future and if YES that when?
    Thank you.
     
  37. eahm

    eahm Addicted to LI Member

    Toastman's?
     
  38. ArCan

    ArCan Connected Client Member

    Yes, of course "Toastman's".
    The last Shibby's f/w has this option.
    What about DNSCRYPT in Toastman's f/w?
    Thank you.
     
  39. eahm

    eahm Addicted to LI Member

    Got it sorry, I thought you said Toastman's already had it.
     
  40. hk2k2

    hk2k2 Network Newbie Member

    here is my bootlog:

    http://pastebin.com/0f2u9rYn

    i can see a kernel oops and segfault at the end, what did this mean :) what causes this ?!



    all used functions works as normal here (except some problems with crond and scheduler....)

    thanks br hk2k2
     
  41. shibby20

    shibby20 LI Guru Member

    You`ve got Segmentation fault, module fault and not corrected detection of flash memory (Found an serial flash with 0 0KB blocks; total size 0MB)

    but you still tell us all is good, all is fine.... !!!

    I tested NVRAM60 build on wnr3500Lv1, v2 and rt-n16.

    All bricked!
    .
     
  42. hk2k2

    hk2k2 Network Newbie Member

    ;) have you tried on your RTN66U too ?

    the problem with the serial flash exist also!!! in normal 32kB nvram version.

    and i found it also in bootlog from ORIGINAL ASUS FIRMWARE:
    Code:
    
    Creating 5 MTD partitions on "Physically mapped flash":
    0x00000000-0x00040000 : "pmon"
    0x00040000-0x01fe0000 : "linux"
    0x0013b400-0x012a0000 : "rootfs"
    0x012a0000-0x01fe0000 : "jffs2"
    0x01fe0000-0x02000000 : "nvram"
    Found an  serial flash with 0 0KB blocks; total size 0MB
    sflash: found no supported devices
    
    ...

    and the kernel oops, seems to accur only if i activate usb support on my router... i will check this out.

    thanks, br hk2k2
     
  43. ArCan

    ArCan Connected Client Member

    Same with RT-N66U and Toastman's (32k) f/w :
    ...
    Dec 31 19:00:07 user.notice kernel: number of CFI chips: 1
    Dec 31 19:00:07 user.notice kernel: cfi_cmdset_0002: Disabling erase-suspend-program due to code brokenness.
    Dec 31 19:00:07 user.notice kernel: Flash device: 0x2000000 at 0x1c000000
    Dec 31 19:00:07 user.notice kernel: Creating 5 MTD partitions on "Physically mapped flash":
    Dec 31 19:00:07 user.notice kernel: 0x00000000-0x00040000 : "pmon"
    Dec 31 19:00:07 user.notice kernel: 0x00040000-0x01fe0000 : "linux"
    Dec 31 19:00:07 user.notice kernel: 0x00127000-0x00780000 : "rootfs"
    Dec 31 19:00:07 user.notice kernel: 0x00780000-0x01fe0000 : "jffs2"
    Dec 31 19:00:07 user.notice kernel: 0x01fe0000-0x02000000 : "nvram"
    Dec 31 19:00:07 user.warn kernel: Found an serial flash with 0 0KB blocks; total size 0MB
    Dec 31 19:00:07 user.err kernel: sflash: found no supported devices
    Dec 31 19:00:07 user.warn kernel: u32 classifier
    ...
     
  44. Washu-Chan

    Washu-Chan Connected Client Member

    I'm considering getting the RT-N66U in the future. I currently have the RT-N16.

    Is loading the Tomato firmware on the RT-N66U nearly identical to the RT-N16's (i.e., put router in recovery mode and use the Asus Firmware Restoration Utility)?
     
  45. Mirko Baila

    Mirko Baila Reformed Router Member

    Yes!Same procedure

    Other question:Have a bug in Time?In Italy change the hour,but in the router not change?
    Why?
     
  46. fubdap

    fubdap Networkin' Nut Member

    Follow the link on post #31.
     
  47. Washu-Chan

    Washu-Chan Connected Client Member

    I live in the US, so I can't say.

    In my case, I set the time using the NIST servers (http://tf.nist.gov/tf-cgi/servers.cgi), and set the time to Eastern Time (UTC -0500)
     
  48. mstombs

    mstombs Network Guru Member

    The tomatousb kernel probes for different types of flash chip, not a problem if a particular type not found - obviously a big problem if none found. The kernel oops is a problem, but probably not related to nvram size, I'm surprised no relevant messages in hk2k2's logs, looks like it will work until nvram gets shuffled and important bits not found by CFE in first 32k?

    The DST time change fail is a classic bug in uclibc see
     
  49. hk2k2

    hk2k2 Network Newbie Member

    from time to time i got this OOPSES again, also with 32k and 60k version.

    what does is mean? what causes this ? anyone can read out something?

    Code:
    CPU 0 Unable to handle kernel paging request at virtual address d661f000, epc == 80087044, ra == 80087034
    Oops[#2]:
    Cpu 0
    $ 0   : 00000000 00000000 00020001 d661f000
    $ 4   : d661f000 00020000 00000000 00000000
    $ 8   : 80306fb0 73756464 702e7265 6974736f
    $12   : 00000001 00000011 00000011 622f7470
    $16   : 00000000 ffffffff 87538000 87acd3e0
    $20   : 00000080 00000000 802e0000 80270000
    $24   : 00000000 8001d8a8
    $28   : 87886000 87887dd8 004c541c 80087034
    Hi    : 00000000
    Lo    : 00000000
    epc   : 80087044     Tainted: P
    ra    : 80087034 Status: 1100fc03    KERNEL EXL IE
    Cause : 00000008
    BadVA : d661f000
    PrId  : 00019749
    Modules linked in: usblp ohci_hcd hfsplus hfs leds_usb led_class ledtrig_usbdev ip6table_mangle ip6table_filter xt_webmon xt_recent xt_IMQ im
    q ehci_hcd vfat fat ext2 ext3 jbd mbcache usb_storage sd_mod scsi_wait_scan scsi_mod usbcore wl(P) et(P) igs(P) emf(P)
    Process sh (pid: 15363, threadinfo=87886000, task=8136a000)
    Stack : 87538030 87538000 00000000 800bd1b4 87538000 87887f30 87887e18 87887f30
            802aa700 800bcb0c 000200d2 00000000 fe141000 fe141000 87538002 00000000
            6e69622f 0068732f 000200d2 00000000 004c5420 800677a4 87887e2c 00000010
            000200d2 00000000 87538130 87503000 8753807c 00000000 0001ffd7 00000000
            004c5420 004c5404 8753807c 00000001 0001fe90 00000000 00000002 812de460
            ...
    Call Trace:[<802b0000>][<80003204>][<80001548>]
    
    Code: 00402021  00501821  24a20001 <80710000> 00402821  ae420104  26100001  24021000  1202ffe5
    note: sh[15363] exited with preempt_count 1
    BUG: scheduling while atomic: sh/0x10000001/15363
    Call Trace:[<80006024>][<80006a50>][<80001500>][<802b0000>][<80003204>][<80001548>]
    thanks ,br
     
  50. quietsy

    quietsy Addicted to LI Member

    Can anyone confirm this strange behavior?

    Create a test script /temp/test.sh:
    #!/bin/sh
    echo $1

    And execute it a bunch of times with a parameter:
    root@Router:/tmp# /tmp/test.sh 2452454
    tmp/test.sh
    root@Router:/tmp# /tmp/test.sh 2452454
    2452454
    root@Router:/tmp# /tmp/test.sh 2452454
    2452454
    root@Router:/tmp# /tmp/test.sh 2452454
    2452454
    root@Router:/tmp# /tmp/test.sh 2452454
    tmp/test.sh
     
  51. Planiwa

    Planiwa LI Guru Member

    What is your context and purpose? Are you suggesting that the shell sometimes acts as if ther had been $0 instead of $1?

    Ah, wait, $0 would be /tmp/test.sh , not tmp/test.sh

    Can you replicate this at the command line?
    Code:
    set -- foo bar # do this once
    while sleep 1;do echo $1;done
    
    break out with Control+c


    Does it still happen when your script only contains the second line?

    Can you remove all the visual clutter?

    Here is something from my .profile:
    export PS1=" "

    This gives me a clean prompt of two spaces.

    (instead of the ugly uninformative one)

    but some folks might confuse two blanks with nothing. :)
     
  52. quietsy

    quietsy Addicted to LI Member

    I removed the visual clutter, edited the test.sh script to look like this:
    #!/bin/sh
    set -- foo bar
    while sleep 1;do echo $1;done

    And the result is:
    ./test.sh
    foo
    foo
    foo
    foo
    foo
    foo
    foo
    foo
    foo
    foo
    foo

    Removed the 2nd line (set -- foo bar) and executed a bunch of times:
    ./test.sh 123
    123
    123
    123

    ./test.sh 123
    /test.sh
    /test.sh
    /test.sh
    /test.sh

    ./test.sh 123
    /test.sh
    /test.sh

    ./test.sh 123
    /test.sh
    /test.sh

    ./test.sh 123
    123
    123

    I'm still puzzled.
     
  53. Planiwa

    Planiwa LI Guru Member

    I meant interactive . . .
     
  54. quietsy

    quietsy Addicted to LI Member

    It seems consistent:
    set -- foo bar # do this once
    while sleep 1;do echo $1;done
    foo
    foo
    foo
    foo
    foo
    foo
    foo
    foo

    while sleep 1;do echo $1;done
    foo
    foo
    foo

    while sleep 1;do echo $1;done
    foo
    foo
    foo
    foo

    while sleep 1;do echo $1;done
    foo
    foo
    foo
    foo
    foo
     
  55. Planiwa

    Planiwa LI Guru Member

    Peculiar. what do you get with:
    Code:
      echo 'echo $0 $1' >foo ; cat foo
    
    and then run
    Code:
      sh -vx foo bar
    
    will it still act strange?
     
  56. quietsy

    quietsy Addicted to LI Member

    This looks good as well.
    The inconsistency shows only when calling a script with #!/bin/sh and some parameters.
     
  57. Planiwa

    Planiwa LI Guru Member

    So, what happens when you invoke the original script with sh -vx /tmp/test.sh foo ? Does it still act up?
     
  58. quietsy

    quietsy Addicted to LI Member

    It works fine with sh -vx /tmp/test.sh foo and with sh /tmp/test.sh foo, but not with /tmp/test.sh foo.
     
  59. Planiwa

    Planiwa LI Guru Member

    I wonder if #!/bin/ash is the same.
    what version is your busybox? sh --help
    also
    cat /etc/motd
     
  60. cohomology

    cohomology Serious Server Member

    I just flashed Tomato Firmware v1.28.0497 MIPSR2-Toastman-VLAN-RT-N K26 USB VPN

    Absolutely love it. However, the QoS graph seems to be broken? Every entry in the bandwidth distribution(In/Out bound both) is 0. Can anyone tell me what is going on?
     
  61. ArCan

    ArCan Connected Client Member

    You are not alone with QoS graphs. I have same issue.
    And somwhere before somebody told about it several times.
    It's wellknown issue.

    And I have still two more.
    Scheduler doesn't work after reboot.
    User MAC for 2.4 band gone away after reboot too...

    P.S. I did reset, 30/30/30 reset, restore stock firmware and back to Tomato
    and always with same issues
    QoS graphs
    Scheduler
    user MAC for 2.4 band
    :-(
     
  62. cohomology

    cohomology Serious Server Member

    OK I C. How can I know QoS is working though?
     
  63. ArCan

    ArCan Connected Client Member

    I'm almost sure that QoS works fine.
    Because I try to change the "Max Bandwidth Limit"
    and mesure the trafic for different class of trafic
    P2P, WEB, VOiP etc.
    and always I had results proporcional my settings.
    Or simple answer even I have max P2P trfic
    my VOiP and other trafics with higher priority works
    fine without timeouts...

    I mean even you don't see QoS grahps QoS really do what it should!


    Thanks to Tomato developers!
     
  64. quietsy

    quietsy Addicted to LI Member

    You are experiencing side effects of the /bin/sh randomness, I also have this inconsistency with QOS, I believe that the QOS scripts under /tmp are causing it.
     
  65. cohomology

    cohomology Serious Server Member

    I have not tried Shiby's build yet. Does it have the same issue with QoS?
     
  66. Planiwa

    Planiwa LI Guru Member

  67. ArCan

    ArCan Connected Client Member

    This is repost from http://tomatousb.org/forum/t-461749

    "Here is a minimalist demonstration of the Hash-Bang Shell Parameter Bug:
    Paste this:

    cd /tmp
    echo "#!/bin/sh" >bug
    echo 'echo -n $1.' >>bug
    chmod +x bug
    for i in 0 1 2 3 4 5 6 7 8 9; do bug $i; done

    You should get a line of 10 punctuated digits.
    But on the N66U, you might see a few "ug"s instead.
    You can do this from Tools>System — no need to ssh or telnet."

    Oh my God...
    RT-N66U is here and
    every time when I execute it I don't have any result without "ug" !!!
    0.1.2.3.ug.5.6.7.8.9.
    ug.1.ug.ug.ug.5.ug.7.8.ug.
    ug.1.2.3.ug.ug.ug.ug.8.ug.
    etc.
    ...

    I did same test on E2000 and it works fine without "ug" all the time.
    "0.1.2.3.4.5.6.7.8.9." Good result !

    P.S. On both routers I have last version of Toastman f/w.
    I would like to know results of such test with DD-WRT on RT-N66U...
     
  68. Planiwa

    Planiwa LI Guru Member

    Does factory N66U firmware have a command line interface?
    If so, does it have the bug?
     
  69. ArCan

    ArCan Connected Client Member

    The RT-N66U with factory firmware pass "ug" test without any problems.

    To ToastMAN.
    Thank you for new version of firmware.
    Mr. Toastman, do you have any opinion about "ug"-issue (RT-N66U only!) with your and Shibby's firmware ?
     
  70. Mirko Baila

    Mirko Baila Reformed Router Member

  71. Toastman

    Toastman Super Moderator Staff Member Member

    No. I can't replicate it on any router I have, it seems to be an issue with this router, but I can't imagine why.
     
  72. Planiwa

    Planiwa LI Guru Member

    Just to disambiguate, please:

    Are you saying that you tried (unsuccessfully) to replicate it on several RT-N66U's?

    (It is not clear that that was your intended meaning.)

    This problem has only been observed on the RT-N66U, running Tomato.
    This problem has not been observed with Asus / DD-WRT FW.
    There have been multiple reports of this problem with Toastman, Shibby, Victek builds, by multiple users with or without memory cards, and/or optware.
    It is intermittent, but frequent when present.
    There have been reports of Tomato on RT-N66U's where the problem has not been replicated.


    So far, no programmer has said that he replicated it and took diagnostic measures.
    Shibby has said that he plans to . . .
     
  73. Toastman

    Toastman Super Moderator Staff Member Member

    OK

    No. I can't replicate it on any routers (of different models and makes) that I have in use, it seems to be an issue with the RT-N66U router, but I can't imagine why.

    The firmware is compiled exactly the same.

    I have no plans to look at this issue. Currently I working away from home and I am too busy.
     
  74. Planiwa

    Planiwa LI Guru Member

    I take that to mean that Toastman has no access to an RT-N66U.
    Then, obviously, he cannot diagnose the problem. He can't even see it.

    I too, have no access to an RT-N66U. [I find it bizarre that there are people who have this device, which has a catastrophic problem, and who don't have system programming skills, but who would rather live with such a ticking time-bomb than give remote access to someone who might be willing and able to help. -- If I had such a router with this problem, I would certainly not "live with it".]

    So far, Shibby is the only programmer who has said that he has access to an RT-N66U.
    But he has not even said that he has checked if this problem exists on his RT-N66U.

    I understand quite well that compiling program takes different skills from recognizing, tracking, diagnosing, and fixing software problems, especially ones that are both intermittent and limited to particular hardware. These are the worst kind.

    Edited . . .

    It is available for $215, and in stock, in my neighbourhood.

    Maybe the people who own the problem don't really want it diagnosed?

    Otherwise, maybe if 8 people could pay $20 each, I'll pay the rest.
    If I can't find the problem myself, I can make it accessible online to programmers with diagnostic tools, etc.

    Meanwhile, I challenge anyone who owns the problem to consider this:

    1. It's your problem. Not my problem.
    2. Even though, I'm willing to help.
    3. As long as you refuse to do the following, I doubt that you really want this problem fixed:

    4. Produce a procedure of the form:

    ... A. Starting with a factory fresh unit, flash XXX-version Tomato.
    ... B. Log in, try the shell parameter test, and it fails.

    I would like to see reports of people who have done that.

    I don't care about people who have no RT-N66U -- this thread is about the RT-N66U.
    I don't care about people who have stuck any hardware into their RT-N66U.
    I don't care about people who have stuck any software into their RT-N66U.
    I don't care about people who have made any configuration changes to their RT-N66U.

    So, if a few people care enough, maybe I'll buy one.
    And (if I can't track down the problem myself) I can make it accessible to other programmers.

    Or, maybe someone here has a better idea? :)
     
  75. ArCan

    ArCan Connected Client Member

    Hi!
    Could sombody tell me about the command to see the temperature of CPU in RT-N66U?
    I think if stock firmware has this options I could run special command in System in Tomato
    to see the result for count the temperature of processor...
    Thank you.
     
  76. Mirko Baila

    Mirko Baila Reformed Router Member

    interested :)
     
  77. Planiwa

    Planiwa LI Guru Member

    Why is it that numerous people say:

    I have an RT-N66U device. I want technical experts to help me.

    But no one says:

    I will provide access to my RT-N66U router to bona fide technical experts so they can help me.

    As far as I know, no technical expert has access to an RT-N66U.

    Shibby seems to own one, but in an entire week has not been able to access it for a few seconds, in order to determine whether it has the Shell Parameter Bug.
     
  78. ArCan

    ArCan Connected Client Member

    Thanks to Toastman for new 1.28.0498.2 version.
    (- IP Traffic: Fix JS error causing initial load delay when default nvram 'tomato' skin
    - ash: fix execution of shell scripts without shebang
    - PPPoE Status Progress - overview page
    - add PPTPD to ext builds
    - Dropbear: 2012.55 update
    - Dnsmasq: 2.60 update)

    I tryed on it already.
    I'm not using IP Traffic, PPPoE, PPTPD and don't execute "shell scripts without shebang"...
    May be it's the reason why I don't see any differences.

    P.S. I'm not sure but seems to me that GUI in some places of menu slower than before.
    For example, I have to wait around 5 seconds to open Device List or IP Traffic...
    ("ug"-test still show negative results)

    Any way thanks a lot for you business Toastman.
     
  79. Toastman

    Toastman Super Moderator Staff Member Member

    OK, don't worry about the slow page, there is a bug with that, I pulled that build. But it made no difference to the ash problem as everyone already suspected.
     
  80. shibby20

    shibby20 LI Guru Member

    [info] i succesful upgraded broadcom driver using Asus`s sources for RT-N66u :) Result:

    Both radios works :) Now i do more tests.
     
  81. Mirko Baila

    Mirko Baila Reformed Router Member

    Wow!!!Great work Shibby!!!
    Thank'you :)
     
  82. linkmaniac

    linkmaniac Reformed Router Member

    Is there a trick to install a tomato build on this router? I tried both toastman and shibby builds, upload is successful but I still cannot access the gui. On the other hand stock firmware installs without a problem. Has anyone had this problem before or am I the only one?

    So far tried these builds:

    tomato-K26USB-1.28.7498.1MIPSR2-Toastman-VLAN-RT-Lite.trx
    tomato-K26USB-1.28.RT-N5x-MIPSR2-085V-VPN.trx
    tomato-K26USB-1.28.RT-N5x-MIPSR2-092-AIO.trx

    none of them worked...
     
  83. linkmaniac

    linkmaniac Reformed Router Member

    Success!!! I `ve flashed first an older stock version (3.0.0.3.72) and then tomato. Now seems everything OK. The problem was flashing from the latest stock version (3.0.0.3.112) to tomato.
     
  84. Handycracker2k2

    Handycracker2k2 Addicted to LI Member

    hmm, for me both radios worked fine also in older versions? or what is the point ? :)

    and an other question, i use toastman mods, and wanna ask, if there is a chance to get multple ssid to work in future ? :)

    if i activate it, gui is comletely damaged (no 5ghz settings visible...) and the 2nd ssid has no tx power..(logged with a notebook, also next to the router i got a VERY SMALL signal...)

    thanks ,br hk2k2
     
  85. Gitsum

    Gitsum LI Guru Member

    New Broadcom drivers. Thats nice to hear, Shibby. Thank you.
    I also read some e-mails from ASUS tech support over at the DD-wrt forums that ASUS will be having a firmware update next month that will increase the NVRAM to 64kb.
     
  86. ArCan

    ArCan Connected Client Member

    Dear Toastman,
    I read message from PeterV on http://wl500g.info wher he wrote interesting information about the commad to see the themperature sensors inside of RT-N66U.

    wl -i eth1 phy_tempsense
    wl -i eth2 phy_tempsense

    to see the temperature like on graph should to use the formula
    x*0.5+20

    The themperature of the RT-N66U the subject of conversation on many fourums.

    Could you include the usfull options to monitor the temperature in one of your next firmware for RT-N66U ?

    Thank you.
     
  87. Toastman

    Toastman Super Moderator Staff Member Member

    I won't add it because I have no RT-N66U to test it on. This command is not supported by other routers.

    While it is a subject of conversation, I see the usual trend - a new router - any problems are being blamed on overheating already.

    I hear rumours that the proposed 64K nvram upgrade does not properly change the cfe and is unstable, but I hope that's wrong.
     
  88. ArCan

    ArCan Connected Client Member

    OK. I see. Thanks.
     
  89. shibby20

    shibby20 LI Guru Member

    Radios have better performance in OFW. This is why i`m trying to move this driver from OFW to Tomato :)
     
  90. ArCan

    ArCan Connected Client Member

    Shibby,
    I think you do absolutely right job!
    Thanks.
     
  91. Toastman

    Toastman Super Moderator Staff Member Member

    If Shibby is successful and the wireless signal gets better, then that might also be useful for other routers. So far, all the "ND" wireless drivers I've tried have been the same. The last drivers used in Tomato that had excellent strength were the older non-ND versions used on the WRT series etc. Remember those? (The ones that had the problems with Intel wireless cards causing the routers to reboot).
     
  92. Gitsum

    Gitsum LI Guru Member

    Shibby, why not trying Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 5.100.138.20 ?
     
  93. Planiwa

    Planiwa LI Guru Member

    Who knows which CPU Frequencies are allowable?
    Apparently, half speed (300MHz) is not allowed.
    Thanks.
     
  94. shibby20

    shibby20 LI Guru Member

    I will try port drivers from WNDR4500 :)
    eth0: Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 5.100.138
    PCI: Enabling device 0000:01:01.0 (0000 -> 0002)
    eth1: Broadcom BCM4331 802.11 Wireless Controller 5.100.138.903
    PCI: Enabling device 0000:02:01.0 (0000 -> 0002)
    eth2: Broadcom BCM4331 802.11 Wireless Controller 5.100.138.903
     
  95. Planiwa

    Planiwa LI Guru Member

    What are they saying?
    Is any of it in English?

    Is there a temperature sensor on anything other than the two radios?

    What did you mean by:

    "to see the temperature like on graph should to use the formula x*0.5+20"?
     
  96. hawkmat

    hawkmat Serious Server Member

  97. SteveE

    SteveE Network Newbie Member

    Shibby... how are you getting on with improved wireless/port drivers?
     
  98. shibby20

    shibby20 LI Guru Member

    at the moment i have (tested in RT-N10U C1):

    about
    dmesg
    IMO stability i better. I didnt test performance yet but wireless is working very nice.
     
  99. Toastman

    Toastman Super Moderator Staff Member Member

    That's good news .. is signal any stronger?
     
  100. SteveE

    SteveE Network Newbie Member

    I'm interested in what speeds you are able to get on wireless...
    I have been running tomato-K26USB-1.28.RT-N5x-MIPSR2-092-AIO.trx for a week or two, with about 60 users on wired and wireless and it's been very stable. No problems at all. But I notice the wireless speed was not as high as I would expect.
     

Share This Page