Topic: davidc502 1900ac 3200acm builds

The content of this topic has been archived between 26 Feb 2018 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.

Page 148 of 229

Post #3676

WWTK

23 Oct 2017, 14:49

Relaxation wrote:

went back to DDWRT in the meantime. I just want a stable build with the Krack fix.

While I agree the fix is important, I personally feel its being blown out of proportion a bit since it requires the attacker to be physically close enough to be able to send and receive your network signal. Then they have to be smart enough to actually code software to take advantage of it.

While this is an important issue to be sure since it is a glaring hole in the process, the likelihood that anyone who could actually would bother with an individual is pretty darn slim IMHO. Things like ATMS or credit card readers or businesses well, that's another story all together

Post #3677

nitroshift

23 Oct 2017, 15:34

@davidc502

Yeah, I'm working with Rango's only, my Mamba's ended up at my mom's and mother-in-law's places.

nitroshift

Post #3678

davidc502

23 Oct 2017, 16:29

The following article doesn't specifically name anything in the 1900, 1200, or 3200 line. However, I'm a bit concerned about the article none the less.

http://www.securityweek.com/no-patches- … ss-routers

Post #3679

WWTK

23 Oct 2017, 16:46

davidc502 wrote:

The following article doesn't specifically name anything in the 1900, 1200, or 3200 line. However, I'm a bit concerned about the article none the less.
http://www.securityweek.com/no-patches- … ss-routers

Yeah, we all know linksys firmware sucks big blue donkey balls...... Hooray for LEDE/OpenWrt!!!!

Post #3680

bill1228

23 Oct 2017, 22:56

Relaxation wrote:

went back to DDWRT in the meantime. I just want a stable build with the Krack fix.

I did the same thing. Running DDWRT r33555 for last 3 days without issue. When LEDE gets things figured out I will be back though.

Post #3681

starrbuck

24 Oct 2017, 00:32

My 3200acm is still rebooting every few hours. :-(

Post #3682

davidc502

24 Oct 2017, 00:36

starrbuck wrote:

My 3200acm is still rebooting every few hours. :-(

You changed from 9.3.0.7 to 9.3.0.8?

I would change back to 9.3.0.7.

Post #3683

starrbuck

24 Oct 2017, 03:04

davidc502 wrote:

starrbuck wrote:
My 3200acm is still rebooting every few hours. :-(
You changed from 9.3.0.7 to 9.3.0.8?
I would change back to 9.3.0.7.

Yessir, but I was having reboots on 9.3.0.7 as well.

Post #3684

davidc502

24 Oct 2017, 03:17

starrbuck wrote:

davidc502 wrote:
starrbuck wrote:
My 3200acm is still rebooting every few hours. :-(
You changed from 9.3.0.7 to 9.3.0.8?
I would change back to 9.3.0.7.
Yessir, but I was having reboots on 9.3.0.7 as well.

good to know... Thank you.

For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.

Post #3685

degroe

24 Oct 2017, 04:02

davidc502 wrote:

starrbuck wrote:
davidc502 wrote:
You changed from 9.3.0.7 to 9.3.0.8?
I would change back to 9.3.0.7.
Yessir, but I was having reboots on 9.3.0.7 as well.
good to know... Thank you.
For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.

I am using David's original 5113 release and 9.3.0.8 firmware on 3200ACM. The router has been up for over two days with no issue except some warning in system log. The warning message entries in system log are from "daemon.warn miniupnpd" and "user.warn igmpproxy".
I also want to report that the legacy 802.11G device compatibility issue is also gone. I have Logictech squeezebox radio and it connected with the 2.4G radio with no issue with r5113 firmware. The same device won't hold connection with 3200 with DDWRT or older LEDE firmware.
I installed LEDE from Linksys firmware; then restored configuration from previous LEDE archive.

Post #3686

eduperez

24 Oct 2017, 08:04

davidc502 wrote:

For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.

I'm currently running LEDE 17.01.4 on a WRT3200ACM + updated mwlwifi drivers (firmware 9.3.0.8); this device has been up since I did the last system upgrade, 3 days ago, and I am not seeing anything suspicious on the logs.

Post #3687

tapper

24 Oct 2017, 09:23

eduperez wrote:

davidc502 wrote:
For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.
I'm currently running LEDE 17.01.4 on a WRT3200ACM + updated mwlwifi drivers (firmware 9.3.0.8); this device has been up since I did the last system upgrade, 3 days ago, and I am not seeing anything suspicious on the logs.

Same here mate! I am not sure that it's the fermwair making the router reboot.

Post #3688

mrremo

24 Oct 2017, 13:23

davidc502 wrote:

For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.

I'm running the r5113 build. After 1 1/2 day uptime the 5GHz is completly gone. I had to reboot the WRT3200ACM. I've seen it twice in a row now. My router has never restarted itself.
The performance of the 5GHz is as good as never before.

root@LEDE:~# cat /sys/kernel/debug/ieee80211/phy0/mwlwifi/info
driver name: mwlwifi
chip type: 88W8964
hw version: 7
driver version: 10.3.4.0-20171011
firmware version: 0x09030008
power table loaded from dts: no
firmware region code: 0x30
mac address: ******************
2g: disable
5g: enable
antenna: 4 4
irq number: 45
ap macid support: 0000ffff
sta macid support: 00010000
macid used: 00000001
radio: enable
iobase0: e1080000
iobase1: e1300000
tx limit: 1024
rx limit: 16384

(Last edited by mrremo on 27 Oct 2017, 06:25)

Post #3689

WWTK

24 Oct 2017, 14:31

davidc502 wrote:

starrbuck wrote:
I'm on 5117 and it came with 9.3.0.7. I used this guide when building:
https://lede-project.org/docs/guide-dev … ild-images
The latest mwlwifi commit was not pushed by LEDE trunk, so to get the latest firmware in the build it required some adjustments prior to compiling.

So i built a new firmware from github now running r5126, but it indeed has 9.3.0.7 driver.

I'm not sure how to tell if the KRACK fixes are there or how to upgrade the mwlwifi or if that would even be beneficial to the group/discussion.

I see the newer firmware is on the router (along with 2 others) but I don't know where to tell it which one to use.

root@LEDE-3200ACM:/lib/firmware/mwlwifi# ls -l
-rwxr-xr-x    1 root     root        118776 Oct 23 10:23 88W8864.bin
-rwxr-xr-x    1 root     root        489932 Oct 23 10:23 88W8897.bin
-rwxr-xr-x    1 root     root        447620 Oct 23 10:23 88W8964.bin
-rw-r--r--    1 root     root          2140 Oct 23 10:23 Marvell_license.txt

If you think it would be useful for me to select the newer driver please let me know and provide a bit of direction on how to pick that one.

Also, the second 5ghz radio is showing up in this build, not sure but I don't think it was there before, but honestly its been a while since I looked at the interfaces lol. The first interface is set to 160mhz which i thought used both radios to accomplish that.

(Last edited by WWTK on 24 Oct 2017, 14:39)

Post #3690

starrbuck

24 Oct 2017, 16:19

davidc502 wrote:

starrbuck wrote:
davidc502 wrote:
You changed from 9.3.0.7 to 9.3.0.8?
I would change back to 9.3.0.7.
Yessir, but I was having reboots on 9.3.0.7 as well.
good to know... Thank you.
For me, the router has a uptime of two days on 9.3.0.7 and nothing in the logs and no reboots. However, would like to hear from other 3200acm users who also reverted firmware.

You're welcome. The only difference with mine is I've tried 9.3.0.7 and .8 and I am using r5117 from when I cloned. Also I turned IPv6 support in packages off but that is the only change I made before I compiled.

The reboots were annoying as they were happening during my WoW raids, so I am trying out DD-WRT. I prefer LEDE and will return when some progress is made on this. So far, DD-WRT has been up for over 15 hours.

Post #3691

davidc502

24 Oct 2017, 16:38

From what everyone is reporting, this is completely a mixed bag (opinions welcome). The only thing I'm thinking at this point is that it could be a bad build. However, we had one person who built his own and was having issues, but was on 9.3.0.7...... So, I'm confused.

I do know my router is going up on 3 days of uptime with no issues on 9.3.0.7... The most was 2 days uptime on 9.3.0.8.

(Last edited by davidc502 on 24 Oct 2017, 16:38)

Post #3692

WWTK

24 Oct 2017, 16:45

davidc502 wrote:

From what everyone is reporting, this is completely a mixed bag (opinions welcome). The only thing I'm thinking at this point is that it could be a bad build. However, we had one person who built his own and was having issues, but was on 9.3.0.7...... So, I'm confused.
I do know my router is going up on 3 days of uptime with no issues on 9.3.0.7... The most was 2 days uptime on 9.3.0.8.

I agree with the bad build possibility, using the current tree, I am building without any errors just make -j4 and I remember reading somewhere back that you and others had errors trying to compile.

Maybe its some other package that not everyone is using that has an issue with the driver?

I would like to build with the new driver, but I don't see it as a choice in menuconfig. I see it in the Make file for the drivers as to be copied, but i'm at a loss how to select it at this point. sorry for the noob issues but gotta learn somehow LOL

Post #3693

davidc502

24 Oct 2017, 17:32

WWTK wrote:

davidc502 wrote:
From what everyone is reporting, this is completely a mixed bag (opinions welcome). The only thing I'm thinking at this point is that it could be a bad build. However, we had one person who built his own and was having issues, but was on 9.3.0.7...... So, I'm confused.
I do know my router is going up on 3 days of uptime with no issues on 9.3.0.7... The most was 2 days uptime on 9.3.0.8.
I agree with the bad build possibility, using the current tree, I am building without any errors just make -j4 and I remember reading somewhere back that you and others had errors trying to compile.
Maybe its some other package that not everyone is using that has an issue with the driver?
I would like to build with the new driver, but I don't see it as a choice in menuconfig. I see it in the Make file for the drivers as to be copied, but i'm at a loss how to select it at this point. sorry for the noob issues but gotta learn somehow LOL

Download the new driver (new firmware will be bundled in) from Kaloz's site. Unzip and then re-zip into a .tar.xz compression format. Get a sha256 and sha1 sum of the package, and edit the mwlwifi makefile and enter the sha's and save. From there just start compiling, and drop the .tar.xz driver package into the dl directory (It doesn't exist until the make process starts). From there the build should use the package that was placed into the dl directory.

If not clear enough let me know and try to be more specific.

*EDIT*

About the failed build packages.... You can leave out the V=s from the make command. From there as packages begin to compile, you will see the ones that fail. If you are compiling all possible packages, you should see a number of them that do fail.

(Last edited by davidc502 on 24 Oct 2017, 17:36)

Post #3694

Villeneuve

24 Oct 2017, 18:13

The rango reboot issue looks like it is client dependant, with a time component(leak maybe), but does not appear to be happening with LEDE stable.

A way to deal with the BLOB if you want to switch between:
get the BLOB from github
ssh to router and scp BLOB to /tmp on you router

wifi down
cd /lib/firmware/mwlwifi
mv 88W8964.bin 88W8964.bin.9307
mv /tmp/88W8964.bin 88W8964.bin.9308
ln -s 88W8964.bin.9308 88W8964.bin
wifi up

which will allow you to change the BLOB in play depending on which you want to test.

(Last edited by Villeneuve on 22 Nov 2017, 20:54)

Post #3695

WWTK

24 Oct 2017, 18:37

davidc502 wrote:

WWTK wrote:
davidc502 wrote:
From what everyone is reporting, this is completely a mixed bag (opinions welcome). The only thing I'm thinking at this point is that it could be a bad build. However, we had one person who built his own and was having issues, but was on 9.3.0.7...... So, I'm confused.
I do know my router is going up on 3 days of uptime with no issues on 9.3.0.7... The most was 2 days uptime on 9.3.0.8.
I agree with the bad build possibility, using the current tree, I am building without any errors just make -j4 and I remember reading somewhere back that you and others had errors trying to compile.
Maybe its some other package that not everyone is using that has an issue with the driver?
I would like to build with the new driver, but I don't see it as a choice in menuconfig. I see it in the Make file for the drivers as to be copied, but i'm at a loss how to select it at this point. sorry for the noob issues but gotta learn somehow LOL
Download the new driver (new firmware will be bundled in) from Kaloz's site. Unzip and then re-zip into a .tar.xz compression format. Get a sha256 and sha1 sum of the package, and edit the mwlwifi makefile and enter the sha's and save. From there just start compiling, and drop the .tar.xz driver package into the dl directory (It doesn't exist until the make process starts). From there the build should use the package that was placed into the dl directory.
If not clear enough let me know and try to be more specific.
*EDIT*
About the failed build packages.... You can leave out the V=s from the make command. From there as packages begin to compile, you will see the ones that fail. If you are compiling all possible packages, you should see a number of them that do fail.

OK running the 9.3.0.8 now with r5126 I'll post back results later

Post #3696

WWTK

24 Oct 2017, 18:38

Villeneuve wrote:

The rango reboot issue looks like it is client dependant, with a time component(leak maybe), but does not appear to be happening with LEDE stable.
A way to deal with the BLOB if you want to switch between:
get the BLOB from github
ssh to router and scp BLOB to /tmp on you router
iw down
cd /lib/firmware/mwlwifi
mv 88W8964.bin 88W8964.bin.9307
mv /tmp/88W8964.bin 88W8964.bin.9308
ln -s 88W8964.bin.9308 88W8964.bin
iw up
which will allow you to change the BLOB in play depending on which you want to test.

Thanks that was way easier!!

Post #3697

WWTK

24 Oct 2017, 20:06

@davidc502

Clearly im not building everything you are since my output is only about 9mb but no, i don't get any errors.

Post #3698

davidc502

24 Oct 2017, 21:28

WWTK wrote:

@davidc502
Clearly im not building everything you are since my output is only about 9mb but no, i don't get any errors.

That is probably where the difference is... Total space on each build is close to 500MB.

Post #3699

WWTK

24 Oct 2017, 21:29

davidc502 wrote:

WWTK wrote:
@davidc502
Clearly im not building everything you are since my output is only about 9mb but no, i don't get any errors.
That is probably where the difference is... Total space on each build is close to 500MB.

i'm talking about the generated firmware, yours is a bit over 10mb

Post #3700

davidc502

24 Oct 2017, 22:26

WWTK wrote:

davidc502 wrote:
WWTK wrote:
@davidc502
Clearly im not building everything you are since my output is only about 9mb but no, i don't get any errors.
That is probably where the difference is... Total space on each build is close to 500MB.

i'm talking about the generated firmware, yours is a bit over 10mb

The difference there would probably just be the included packages.

Are building all the possible modular packages? Defined as "M" in the make menuconfig?

Page 148 of 229