Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Howlingwolf]

I don't know if there is a way of doing it via telnet but even so, Asbokid's guide is probably the easiest method.

[D-an-W]

All done, I used "bcm96368MVWG_fs_kernel_HG612V100R001C01B030SP06_unlockedgui-nobtagent" was that the best choice?

[les-70]

   B030SP06_unlockedgui-nobtagent is running fine for me on adsl2 annex m.  The GUI allowed all the usual changes that I make re ADSL and IP addresses and it really helpful to be able to save the config file.  Reloading it also worked fine. Your edit to allow saving the config file in B028SP10_no-btagent is also a real help if that firmware proves better in other ways.  I did look at the serial output but I did not learn much - just seemed to be a few extra complaints during the initial startup but not when making the GUI changes. 

[roseway]

B030SP06_unlockedgui-nobtagent is working fine here too, and everything seems to be according to specification. The GUI has all the expected features, and there are no BTAgent processes running.

Code: [Select]

BusyBox v1.9.1 (2013-06-01 18:30:08 CST) built-in shell (ash)
Enter 'help' for a list of built-in commands.

# ps
  PID  Uid        VSZ Stat Command
    1 0           336 S   init               
    2 0               SWN [ksoftirqd/0]
    3 0               SW< [events/0]
    4 0               SW< [khelper]
    5 0               SW< [kthread]
   19 0               SW< [kblockd/0]
   32 0               SW  [pdflush]
   33 0               SW  [pdflush]
   34 0               SWN [kswapd0]
   35 0               SW< [aio/0]
   49 0               SW  [mtdblockd]
   59 0           400 S   -/bin/sh
  101 0               SW  [dsl0]
  116 0               SW  [eth0]
  127 0           504 S   mc
  128 0           284 S   console
  129 0           504 S   mc
  130 0           504 S   mc
  131 0           380 S   /bin/msg msg
  132 0           664 S   /bin/log
  133 0           464 S   /bin/klog
  134 0           664 S   /bin/log
  135 0           664 S   /bin/log
  136 0          1120 S   /bin/dbase
  137 0          1352 S   /bin/web
  138 0          1352 S   /bin/web
  139 0          1352 S   /bin/web
  140 0           992 S   /bin/cli
  141 0           992 S   /bin/cli
  142 0           992 S   /bin/cli
  143 0           992 S   /bin/cli
  144 0           992 S   /bin/cli
  145 0           992 S   /bin/cli
  146 0          1640 S   /bin/cms
  147 0          1148 S   /bin/cwmp
  148 0          1640 S   /bin/cms
  193 0           336 S   zebra -f /var/zebra/zebra.conf
  194 0           284 S   console
  195 0           284 S   console
  196 0           284 S   console
  198 0           336 S   ripd -f /var/zebra/ripd.conf
  277 0          1640 S   /bin/cms
  278 0          1640 S   /bin/cms
  310 0           616 S   dhcps
  312 0           240 S   ipcheck
  522 0           488 S   pppc -I ppp256
  528 0           504 S   monitor
  531 0           504 S   monitor
  532 0           504 S   monitor
  546 0           380 S   dnsmasq --conf-file=/var/dnsmasq.conf
  626 0           440 S   sntp
  636 0           292 S   sshd -E
  655 0               SWN [jffs2_gcd_mtd1]
  657 0          1020 S   MidServer
 1136 0           992 S   /bin/cli
 1137 0           400 S   /bin/sh -si
 1138 0           992 S   /bin/cli
 1139 0           404 S   /bin/sh -si
 1144 0           324 R   ps
#


[Chrysalis]

can you guys using B030SP06_unlockedgui-nobtagent do this test? thanks

enable QoS in gui (if disabled) do a upload speedtest.
then disable QoS in gui, do another speedtest and confirm you get roughly extra 10% speed.

thanks

toggling QoS doesnt restart connection.

[Chrysalis]

will the fixed firmware be labeled differently on the site to download?

Sorry. Perhaps I could have made that clearer.

It's already done. The 'new' version has the same filename with a later upload time (2013-10-24 17:40).

I didn't want to have multiple versions of the same firmware hanging around, (xxx_unlocked, xxx_unlocked-fixed, xxx_unlocked-003). Particularly as there may be a new... new version soon. It could get confusing 

I noticed the timestamp on the BO28SP10 is still earlier, that one is safe?

[Ixel]

can you guys using B030SP06_unlockedgui-nobtagent do this test? thanks

enable QoS in gui (if disabled) do a upload speedtest.
then disable QoS in gui, do another speedtest and confirm you get roughly extra 10% speed.

thanks

toggling QoS doesnt restart connection.

I would but unfortunately a while back I managed to accidentally firewall block myself out of the web interface. I can still use telnet, but I don't know if or how to regain access to the web interface. I thought flashing the new firmware (unlockedgui-nobtagent or w/e it's called) would've done so, but I think it kept the original settings I use.

So far I've noticed that the number of CRC errors and consequently ES have been significantly lower (approximately 40% less). Bitswapping count seems higher, with over 3000 in the last hour on the downstream. I've also noticed some parameters on xdslcmd that I haven't noticed before, e.g. --maxdatarate, --ginp, unless I overlooked them.

[roseway]

can you guys using B030SP06_unlockedgui-nobtagent do this test? thanks

enable QoS in gui (if disabled) do a upload speedtest.
then disable QoS in gui, do another speedtest and confirm you get roughly extra 10% speed.

thanks

toggling QoS doesnt restart connection.

It didn't make any difference here. I assume that you meant us to simply tick or untick the QoS option, without setting up any actual QoS parameters?

[Howlingwolf]

All done, I used "bcm96368MVWG_fs_kernel_HG612V100R001C01B030SP06_unlockedgui-nobtagent" was that the best choice?

My reasoning behind providing both was to give people the choice of allowing Openreach to update their modem or not, without forcing people who don't to fiddle about manually killing the BTAgent process every time the modem is restarted.

just seemed to be a few extra complaints during the initial startup

What are the new errors you are seeing?  The only one I saw reported during my tests was 'Bad rule' from iptables. Most of the others seemed to be common across all the variations I tested.

[les-70]

Likewise - nothing significant

QOS
on      down 14.52 up 1.48
                 14.54    1.50

off             14.59    1.48
                 14.53    1.50

on             14.53    1.50
                14.56    1.50     
       
off            14.59    1.49   
                14.54    1.50 

  What were you expecting to happen?   (I notice that my HG622 completely loses internet access with QOS turned on and default settings.  The ethernet connection looks fine but no internet. )

[Howlingwolf]

I noticed the timestamp on the BO28SP10 is still earlier, that one is safe?

Yes. It was only the new B030SP06 which needed replacing and even that was just a precaution.

I would but unfortunately a while back I managed to accidentally firewall block myself out of the web interface. I can still use telnet, but I don't know if or how to regain access to the web interface. I thought flashing the new firmware (unlockedgui-nobtagent or w/e it's called) would've done so, but I think it kept the original settings I use.

I'm afraid so. There's an earlier post in this thread if you want the technical details but the long and the short of it is you need to restore the default settings to regain access.

So far I've noticed that the number of CRC errors and consequently ES have been significantly lower (approximately 40% less). Bitswapping count seems higher, with over 3000 in the last hour on the downstream. I've also noticed some parameters on xdslcmd that I haven't noticed before, e.g. --maxdatarate, --ginp, unless I overlooked them.

No, those are new. No idea what they do yet, although maxdatarate seems pretty obvious.

[Ixel]

So far I've noticed that the number of CRC errors and consequently ES have been significantly lower (approximately 40% less). Bitswapping count seems higher, with over 3000 in the last hour on the downstream. I've also noticed some parameters on xdslcmd that I haven't noticed before, e.g. --maxdatarate, --ginp, unless I overlooked them.

No, those are new. No idea what they do yet, although maxdatarate seems pretty obvious.

I'd also speculate that ginp is related to the impulse noise protection value, however I'm not sure how that is set, and if it even works. I tried messing around with maxdatarate thinking I could band my connection speed as a test (worked on the Fritz!Box 7390), but I was unsuccessful. For example, --maxdatarate 50000 15000 0 didn't make any difference and the values were still 0 when checking --cfg.

[les-70]

What are the new errors you are seeing?  The only one I saw reported during my tests was 'Bad rule' from iptables. Most of the others seemed to be common across all the variations I tested.

   I should have looked more carefully.  Having done a careful check of the startup with asbo's new blob and B030SP06_unlockedgui-nobtagent the only extra errors follow the eth0... section. the "Error, Control port.'s   

"eth0.5 entered promiscuous mode
 Error, ControlPort.
 Error, ControlPort.
 Error, ControlPort.
 Error, ControlPort."

  You seem to have got rid of  a usbmount not found message.

   but the "iptables: Bad rule" message is common between these versions along with a few other errors

[les-70]

t I was unsuccessful. For example, --maxdatarate 50000 15000 0 didn't make any difference and the values were still 0 when checking --cfg.

    If you try smaller values e.g. on adsl2  "--maxDataRate 1500 150 160  or 150 15 160 the values DO show when checking with --cfg but don't seem to make any difference!!   I am also not sure what maxaggrdatarate might be.  Maybe something else also needs enabling at the same time?

   Edit ----- In fact it does not like maxaggrdatarate being zero, that seems to be the issue with it not showing with --cfg, - I changed two things at once!! I have tried a wide rage of values now but with no impact. Be aware that this gives a resync every time you try it.  Not a problem on TT but would be for some experimenters.

[Howlingwolf]

What are the new errors you are seeing?  The only one I saw reported during my tests was 'Bad rule' from iptables. Most of the others seemed to be common across all the variations I tested.

   I should have looked more carefully.  Having done a careful check of the startup with asbo's new blob and B030SP06_unlockedgui-nobtagent the only extra errors follow the eth0... section. the "Error, Control port.'s   

"eth0.5 entered promiscuous mode
 Error, ControlPort.
 Error, ControlPort.
 Error, ControlPort.
 Error, ControlPort."

  You seem to have got rid of  a usbmount not found message.

   but the "iptables: Bad rule" message is common between version along with a few other errors

Those control port errors exist in the original openreach update as well. I would hazard a guess it's something Openreach stripped to save space and didn't clean up properly. The usbmount msg in the B028SP10 firmware is probably more of the same which was fixed in B030SP10.

Still, I will go through them again in case I missed something. It's going to take a while as there are six variations to flash and check - B028SP10: locked/unlocked/nobtagent and B030SP06: locked/unlocked/gui_nobtagent.