Voxel Custom firmware build for R7800 v. 1.0.2.80SF & v. 1.0.2.80.4SF & v. 1.0.2.80.5SF

[R. Gerrits]

hmmm, indeed seems to be a bug..

Advanced -> ReadySHARE -> Advanced Settings
there enable FTP(via internet) and enable Admin Password Protection

But then I still can logon to FTP anonymously:

PS C:\Users\richard> ftp
ftp> open xxxxx.mine.nu
Connected to xxxxx.mine.nu.
220 ProFTPD Server (NETGEAR-R7800) [::ffff:94.213.xxx.xxx]
200 UTF8 set to on
User (xxxxx.mine.nu:(none)): anonymous
331 Anonymous login ok, send your complete email address as your password
Password:
230 Anonymous access granted, restrictions apply
ftp> ls
200 PORT command successful
150 Opening ASCII mode data connection for file list
shares
226 Transfer complete
ftp: 11 bytes received in 0.00Seconds 3.67Kbytes/sec.
ftp> cd shares
250 CWD command successful
ftp> ls
200 PORT command successful
150 Opening ASCII mode data connection for file list
USB_Storage
226 Transfer complete
ftp: 16 bytes received in 0.00Seconds 16.00Kbytes/sec.
ftp>

I'll have a look if I can see where it goes wrong.

 

[rick131]

That was exactly what it was about. The question is how to fix it.

 

[R. Gerrits]

A quick solution:
Edit /sbin/cmdftp and change

# ftp_passwd_enable=`${config} get usb_passwdFvia`
ftp_passwd_enable=0

into

ftp_passwd_enable=`${config} get usb_passwdFvia`
# ftp_passwd_enable=0

 

[rick131]

o thanks, when i try to edit this file via telnet or ssh, after trying to write it throws out no permissions. I'm logging as admin.

 

[R. Gerrits]

so you do

nano /sbin/cmdftp

then you make the change
and do CTRL-X, Y, enter

And then you get the error?
Can you check the output of

ls -al /sbin |grep cmdftp

 

[danlat1415]

GPL source codes of 1.0.2.74 are not published yet, check this link:

https://kb.netgear.com/2649/NETGEAR-Open-Source-Code-for-Programmers-GPL

latest codes: 1.0.2.68

So changes in 1.0.2.74 could not be processed by me and are not included into my build.

But I am a bit skeptical regarding:



First, as I guess security all the fixes are related to some issues (IMO not so serious) re: access to router GUI via https. At least partially fixed in 80SF (https://github.com/SVoxel/R7800/commit/6079a3e0a7fac3c928a001b984314eb81273fdb7).
There are a lot of much more serious CVEs left in the stock. Briefly I've checked the binary firmware, and e.g. 1.0.2.74 is still using OpenSSL 1.0.2h (May 2016), and not accelerated. There following CVE since this version detected (I exclude platform-specific CVE):

CVE-2016-7052
CVE-2016-7055
CVE-2017-3731
CVE-2017-3735
CVE-2017-3737
CVE-2018-0739
CVE-2018-0737
CVE-2018-0732
CVE-2018-0734
CVE-2018-5407
CVE-2019-1559
CVE-2019-1547
CVE-2019-1563

Version of libjson-c in 1.0.2.74 is 0.12.1. So at least CVE-2020-12762.

Well, there are other old packages.


Fix of corrupted files in AP mode according to NG official announcement above is related to wireless clients. I've faced such corrupted files using wired client. So I am not sure that everything is really fixed.

I do not remember such bug reports re: disappeared devices from attached device list with my latest releases.

Minor bugs... Well. As a rule there could be new bugs after such fixes by NG/DNI.


So, let us wait for GPL.

Voxel.

The GPL has now been released for .74 firmware update;

https://www.downloads.netgear.com/files/GPL/R7800-V1.0.2.74_gpl_src.tar.bz2.zip

 

[R. Gerrits]

And @Voxel, a proper solution for the issue that rick131 discovered:
Edit /sbin/cmdftp and change

# ftp_passwd_enable=`${config} get usb_passwdFvia`
ftp_passwd_enable=0

into

if [ "$($config get usb_passwdFvia)" = "1" -o "$($config get usb_passwdFtp)" = "1" ]; then
  ftp_passwd_enable=1
else
  ftp_passwd_enable=0
fi

Perhaps you can include this in the next release of your firmware.

 

[rick131]

It's ok now, thanks a lot

 

[rick131]

Hello. And another problem with the FTP server. Logging on to the server (USB drive), each client shows incorrect file sizes above 4 GB. There is a way around it, apparently it's a common problem with proFTPd. I checked both ntfs and ext4 drives - same problem.

 

[R. Gerrits]

Hypothetical:
proftpd in firmware is compiled without option --enable-largefile
(see https://svn.dd-wrt.com//changeset/23546 where they fixed a similar issue for dd-wrt).

If this indeed is the case, then it can only be fixed by @Voxel in a new release.

(or if you have entware installed, you could attempt recompiling proftpd from source with that option enabled, and replace the firmware binary with the compiled binary)

 

[rick131]

Thank you for your response. Hope Voxel can fix it. I wonder if the problem was also on the stock firmware ...

 

[Voxel]

Hope Voxel can fix it. I wonder if the problem was also on the stock firmware ...

OK. wait for a quickfix version. I think I'll release it today. Plus fixing SAMBA issue on Android/iOS:

When I login to \\readyshare on a PC and Stream media through VLC from the HDD, video plays perfectly fine. However, when I use VLC on Android and iOS, I select my Network Location R7800 [SMB] and I cannot get video to play back on any of my devices. I tried different programs, different video players, still no joy.

Well,

Hypothetical:
proftpd in firmware is compiled without option --enable-largefile
(see https://svn.dd-wrt.com//changeset/23546 where they fixed a similar issue for dd-wrt).

Thank you.
Most important in dd-wrt fix is this "sed" command. As far as I understand --enable-largefile is used by default. But I added this configure option explicitly to Makefile as well.

Voxel.

 

[Voxel]

This release is definitely the fastest yet.

I think that upcoming release I mentioned above will be a bit faster. Compare the screenshot:

https://www.snbforums.com/threads/custom-firmware-build-for-r7800-v-1-0-2-79sf.65364/post-604899

and current screenshot:

Thanks to @vladlenas for his testing and confirmation of slight Wi-Fi acceleration.

I hope that rest will be OK as well...

Voxel.

 

[Voxel]

Quickfix/Snapshot version of my custom firmware build: 1.0.2.80.4SF.

Changes (vs 1.0.2.80SF):

1. Toolchain: Go is upgraded 1.15.1->1.15.2.
2. Toolchain: binutils version is upgraded 2.35->2.35.1.
3. Add ARM acceleration to kernel crypto SHA1.
4. OpenSSL v. 1.1.1 package is upgraded 1.1.1g->1.1.1h.
5. jansson package is upgraded 2.12->2.13.1.
6. libjson-c package is upgraded 0.14->0.15.
7. expat package is upgraded 2.2.9->2.2.10.
8. nano package is upgraded 5.2->5.3.
9. transmission-web-control package is upgraded 2019-07-24->1.6.1+ (2020-09-26).
10. Change SAMBA config generation (for Android/iOS gadgets, issue reported by Rustypouch).
11. Make an order in samba36 Makefile.
12. Fix proftpd issue: change cmdftp (thanks to R. Gerrits).
13. Fix proftpd issue: display size of large file (thanks to R. Gerrits).

Voxel.

 

[danlat1415]

Thanks.

I've installed .80.4SF without issue and running fine so far.

Before updating to 80.4SF, I tried out Netgear's .74 firmware.

I've noticed that they've now added VPN Client to their stock firmware.

 

[SteevieWeevie]

Yes, so I updated to the 80.4 this morning and my speeds are:

I updated my previous post with the WiFi sweet spot data. These are from the same position and on wifi.

TBH I don't really understand how updates to elements of the firmware can produce such a difference in WiFi throughput, it's like some dark magic. But nevertheless I am extremely happy and grateful to Voxel for working his magic on this.

My WiFi speed has never been an issue for me as neighbours here seem to stick to good etiquette using single channel 2.4g on 1,6 and 11 and my 5g WiFi is 80Mhz on channel not overlapping with any other received 5ghz networks. Still, seeing an increase to my ISP download speeds for the price of a free and vastly superior firmware use is very nice.

 

[Voxel]

Well, thank you for your benchmark. I am interested in such tests.

TBH I don't really understand how updates to elements of the firmware can produce such a difference in WiFi throughput, it's like some dark magic.

A bit overcomplicated technical details. I was able to extract the the part of kernel source codes from more new kernel (namely 3.10.x, R8900/R9000) and insert them into kernel of R7800 (3.4.103). P. "3." in my changes.log. These changes mean acceleration of some operations exactly on ARM platform. So as I expected: slight acceleration. Anyway: any acceleration is good, right?

According to @vladlenas he had acceleration from 1 to 13 per cents (depending on client, testing by iperf3). Your results are about 5.6%. Good: AVG. So, well... Thank you.

Still working on this: re get maximum what we can from R7800 hardware. I am owner of R7800 and I am interested in this. Hardware should work.

Voxel.

 

[SteevieWeevie]

Anyway: any acceleration is good, right?

Yes! Absolutely.

According to @vladlenas he had acceleration from 1 to 13 per cents (depending on client, testing by iperf3). Your results are about 5.6%. Good: AVG. So, well... Thank you.

Thanks are due to you, not me. I am here and freely benefit, as many do, from work you have done. Your attitude toward these routers I agree with completely. If only the manufacturers shared this view, but then I suppose it would not make business sense to have a product perform so well as to make upgrading to the new thing unnecessary because 'capitalism' and such.

Unlike many users on the forum I am not advanced and do not go into nitty gritty bits. Even so, I understand enough to know that your firmware cleans up a lot of mess from Netgear's work and that having latest libraries, code bases etc is better for security and - as so clearly demonstrated - overall performance. So far as using the router day to day the connection almost feels a little snappier, which is probably placebo effect but the numbers on paper do not lie and, as you said, any acceleration is good. This is doubly the case if it comes at no cost real cost except of course your time and skill.

So thank you again Voxel. A happy fan here

 

[kokishin]

Due to corona virus, a @Voxel update brings some needed excitement to my life (in a good way).

Just updated to 80.4SF.

Thanks you Voxel!

 

[Voxel]

but then I suppose it would not make business sense to have a product perform so well as to make upgrading to the new thing unnecessary because 'capitalism' and such.

OOT. Funnily. I think that I was able to change (a bit) NG's business plans re: R7800 life and EoL for R7800. E.g. vs their pushed by them XR500. In spite of their silent refusal to cooperate with me. But anyway I think my work brings the profits to NG. So... Just interesting.

Voxel.