[Preview] Asuswrt-Merlin 380.63 alpha builds are available

[RMerlin]

The initial plan was for the next release to be based on 3.0.0.4.382, however it seems that 382 is taking longer than expected, so there will be one more build in the 380 series.

Early 380.63 test builds are available on Mediafire. These are pre-beta work-in-progress, and still a lot can change before the beta/final stage.

Those builds are released as-is, with no support for them. I might occasionally refresh them as development progress.

https://www.mediafire.com/folder/bj94sbhrh7e49/Test_Builds

24-Oct-2016: Refreshed builds.

7b46daf kernel: Fixed bad merge for RT-AC3200 kernel
c90170a webui: Added contextual help for NAT passthrough settings
fa53022 Updated documentation
f9aa712 kernel-mips: Patch the 2.6.22.19 kernel against CVE-2016-5195
7e18948 tor: update to 0.2.8.9 (contains security fixes; closes #1067)
3c81765 Updated documentation
bef8fab webui: Created contextual help popups for various Merlin-specific settings
5171f34 webui: remove extra space between "Manufacturer" and ":" to match the other fields on the MAC popups
fafbbb4 webui: Implement our own help array for openHint() so we won't collide with Asus's, and implement a first entry for the IPv6 firewall
b8e8d67 firewall: support fixed interface ID (EUI64) ipv6 destination addresses (based on john9527's patch)
cdefe67 helper: added pc-delete()
821c324 vpn: Made policy rules start at 10000 instead of 1000
16fcf51 webui: Traffic Manager pages do not reside under QoS, so don't label them as such on bwdpi-enabled models
4636162 webui: Use localized strings for Download/Upload labels on Traffic Monitor charts, and remove extra leading space to scale label
7a95789 kernel-arm: Patched all three 2.6.36 kernels against CVE-2016-5195 (Closes #1066)
343d6f9 ipset_arm: some build systems try to link against the wrong library, ensure we do link against our built libmnl.so (closes #1063)
f022993 dropbear: disable MD5 support, for security reasons
74eeda2 rc: re-create passwd/shadow/group after the JFFS2 partition is mounted, in case we have custom configs stored there (fixes #1061)

 

[conquer22]

Thanks for all your support.

 

[RMerlin]

Shots of the new charts:

 

[Adamm]

It seems after the IPSet update, IPTables is having trouble using "--match-set". Good to finally see an updated version though.

admin@RT-AC68U-EE20:/jffs/scripts# ipset list
Name: Blacklist
Type: hash:ip
Revision: 0
Header: family inet hashsize 1024 maxelem 500000
Size in memory: 8248
References: 0
Members:

admin@RT-AC68U-EE20:/jffs/scripts# iptables -I INPUT -m set --match-set Blacklist src -j DROP
iptables: No chain/target/match by that name.

 

[john9527]

It seems after the IPSet update, IPTables is having trouble using "--match-set".

Module names have changed......you need to load xt_set...

http://www.snbforums.com/threads/se...t-build-for-rt-ac88u.31609/page-5#post-285912

 

[Adamm]

Module names have changed......you need to load xt_set...

http://www.snbforums.com/threads/se...t-build-for-rt-ac88u.31609/page-5#post-285912

Perfect, that was it, thanks.

 

[ChriscomIT]

AWESOME !!! Try it after work!!! <3 RMerlin

 

[liamhere]

trying it now on my AC3200

 

[Bitgod]

Ooooo, purty charts!

 

[swetoast]

Wonder if asus are gonna patch for Dirty Cow exploit

 

[tomsk]

Wonder if asus are gonna patch for Dirty Cow exploit

I had to find out what a dirty cow was...... and its got a logo!

 

[RMerlin]

Note that the issue is not as serious for a router that it is for a server/workstation, since the router is pretty much a single-user device, where most stuff runs as root already.

 

[Fitz Mutch]

Wonder if asus are gonna patch for Dirty Cow exploit

Isn't it Broadcom's responsibility to patch the kernel? It's their SDK, "Linux version 2.6.36.4brcmarm".

 

[swetoast]

True and my guess that they arent gonna do it beside as RMerlin said most of the stuff runs on root anyways but it would be nice with a newer kernel.

 

[RMerlin]

Isn't it Broadcom's responsibility to patch the kernel? It's their SDK, "Linux version 2.6.36.4brcmarm".

Patching can be done by the OEM/third party.

In any case, this is all off topic here.

 

[RussellInCincinnati]

Shots of the new charts:

View attachment 7515 View attachment 7512 View attachment 7514

This is such a great idea. Am a happy user of QOS quality of service on an AC-68u, just the most basic bandwidth limiting to get rid of "buffer bloat" pauses, but soon you will be giving us the tools and insight to figure out what traffic is the most important to prioritize. Thanks much, Merlin.

 

[jeff288]

How much trouble would it be to port the QOS chart to visualize the firewall log? It'd be great to get an idea of what countries keep knocking at the doors so much and what ports are being attacked without having to do some manual searching.

 

[Shark6969]

Thanks... checking version.

 

[muffintastic]

@RMerlin - Any chance you can try and fix the DMZ and UPNP issues in the next build? I've had to revert back to a previous build where it worked correctly.

If you don't know what I'm talking about I posted it in the 380.62_1 thread (here)

 

[dsring]

@RMerlin - Any chance you can try and fix the DMZ and UPNP issues in the next build? I've had to revert back to a previous build where it worked correctly.

If you don't know what I'm talking about I posted it in the 380.62_1 thread (here)

What build are you running?