Asuswrt-Merlin 374.42_2 is out

[RMerlin]

Hi everyone,

374.42 is now available for download for all supported models. This release is mostly a maintenance version.

EDIT: 374.42_2 is now available for the RT-AC56 and RT-AC68, addressing the Time Machine issues.

Highlights of this release:

• Merged with Asus's 374_5656 GPL, which included a few additional security fixes, and a reworked web interface for FTP/SMB management.
• Various user-contributed tweaks and fixes related to IPv6 were merged in.
• OpenVPN and miniupnpd were updated to their latest versions. The miniupnpd update includes some improvements to PCP support.

Additional details:
The 3G/4G USB support should work better in this build than the 374.41 build in the case of MIPS-based routers (N16/N66/AC66). That's because I didn't have the latest kernel code for these routers for 374.41, which means I was missing the new cdc_ncm driver which is now used by various devices. Let me know if you get better luck with this release with 3G/4G devices if they stopped working in 374.41.

Mac users, also see if PCP now works properly for you (an Asuswrt-Merlin user reported an issue to the miniupnpd author, who has since fixed it - this build includes his fixed version).

Changelog is here.
Downloads are here.
More details can be found in the complete changelog.

MD5 checksums:

4bb39a92912c50efeb22eb84a210385b  RT-AC56U_3.0.0.4_374.42_0.trx
3e559f06d9a72dceb0a704b61ec40438  RT-AC66U_3.0.0.4_374.42_0.trx
fc40bb66ec5d541021da8b9777eebaa4  RT-AC68U_3.0.0.4_374.42_0.trx
f0171e4dbb7b55bce68e7ce124495cff  RT-N16_3.0.0.4_374.42_0.trx
8d30f1ee3e14dabcbd75bc5d3b22693b  RT-N66U_3.0.0.4_374.42_0.trx

5db83e282b973e9332e4d63d088d9e67  RT-AC56U_3.0.0.4_374.42_2.trx
079f310c747d4beeb59d7040eb166437  RT-AC68U_3.0.0.4_374.42_2.trx

Enjoy!

 

[Daybreak]

Howdy,

Loaded up 374.42.0 on a RT-N66U and a RT-AC68 without issue. I had .41 build on both units without issue before for 22 days. upgrading... because I can

 

[lbradio]

Hi everyone,

374.42 is now available for download for all supported models. This release is mostly a maintenance version.

Highlights of this release:

• Merged with Asus's 374_5656 GPL, which included a few additional security fixes, and a reworked web interface for FTP/SMB management.
• Various user-contributed tweaks and fixes related to IPv6 were merged in.
• OpenVPN and miniupnpd were updated to their latest versions. The miniupnpd update includes some improvements to PCP support.

More details can be found in the complete changelog.

MD5 checksums:

4bb39a92912c50efeb22eb84a210385b  RT-AC56U_3.0.0.4_374.42_0.trx
3e559f06d9a72dceb0a704b61ec40438  RT-AC66U_3.0.0.4_374.42_0.trx
fc40bb66ec5d541021da8b9777eebaa4  RT-AC68U_3.0.0.4_374.42_0.trx
f0171e4dbb7b55bce68e7ce124495cff  RT-N16_3.0.0.4_374.42_0.trx
8d30f1ee3e14dabcbd75bc5d3b22693b  RT-N66U_3.0.0.4_374.42_0.trx

Enjoy!

No direct link to the download? #because-I'm-lazy.
I will certainly give it a try, thanks for the great contribition.

 

[Fester1952]

No direct link to the download?

https://www.mediafire.com/folder/bkfq2a6aebq68//Asuswrt-Merlin

 

[sinshiva]

https://community.openvpn.net/openvpn/wiki/ChangesInOpenvpn23

OpenVPN 2.3.4

Arne Schwabe (1):
Fix man page and OSCP script: tls_serial_{n} is decimal

Dmitrij Tejblum (1):
Fix is_ipv6 in case of tap interface.

Gert Doering (8):
IPv6 address/route delete fix for Win8
Add SSL library version reporting.
Minor t_client.sh cleanups
Repair --multihome on FreeBSD for IPv4 sockets.
Rewrite manpage section about --multihome
More IPv6-related updates to the openvpn man page.
Conditionalize calls to print_default_gateway on !ENABLE_SMALL
Preparing for release v2.3.4 (ChangeLog, version.m4)

James Yonan (2):
Use native strtoull() with MSVC 2013.
When tls-version-min is unspecified, revert to original versioning approach.

Steffan Karger (4):
Change signedness of hash in x509_get_sha1_hash(), fixes compiler warning.
Fix OCSP_check.sh to also use decimal for stdout verification.
Fix build system to accept non-system crypto library locations for plugins.
Make serial env exporting consistent amongst OpenSSL and PolarSSL builds.

Yawning Angel (1):
Fix SOCKSv5 method selection

kangsterizer (1):
Fix typo in sample build script to use LDFLAGS

OpenVPN 2.3.3

Alon Bar-Lev (1):
pkcs11: use generic evp key instead of rsa

Arne Schwabe (8):
Add support of utun devices under Mac OS X
Add support to ignore specific options.
Add a note what setenv opt does for OpenVPN < 2.3.3
Add reporting of UI version to basic push-peer-info set.
Fix compile error in ssl_openssl introduced by polar external-management patch
Fix assertion when SIGUSR1 is received while getaddrinfo is successful
Add warning for using connection block variables after connection blocks
Introduce safety check for http proxy options

David Sommerseth (5):
man page: Update man page about the tls_digest_{n} environment variable
Remove the --disable-eurephia configure option
plugin: Extend the plug-in v3 API to identify the SSL implementation used
autoconf: Fix typo
Fix file checks when --chroot is being used

Davide Brini (1):
Document authfile for socks server

Gert Doering (9):
Fix IPv6 examples in t_client.rc-sample
Fix slow memory drain on each client renegotiation.
t_client.sh: ignore fields from "ip -6 route show" output that distort results.
Make code and documentation for --remote-random-hostname consistent.
Reduce IV_OPENVPN_GUI_VERSION= to IV_GUI_VER=
Document issue with --chroot, /dev/urandom and PolarSSL.
Rename 'struct route' to 'struct route_ipv4'
Replace copied structure elements with including <net/route.h>
Workaround missing SSL_OP_NO_TICKET in earlier OpenSSL versions

Heikki Hannikainen (1):
Always load intermediate certificates from a PKCS#12 file

Heiko Hund (2):
Support non-ASCII TAP adapter names on Windows
Support non-ASCII characters in Windows tmp path

James Yonan (3):

TLS version negotiation
Added "setenv opt" directive prefix.
Set SSL_OP_NO_TICKET flag in SSL context for OpenSSL builds, to disable TLS stateless session resumption.

Jens Wagner (1):
Fix spurious ignoring of pushed config options (trac#349).

Joachim Schipper (3):
Refactor tls_ctx_use_external_private_key()
--management-external-key for PolarSSL
external_pkcs1_sign: Support non-RSA_SIG_RAW hash_ids

Josh Cepek (2):
Correct error text when no Windows TAP device is present
Require a 1.2.x PolarSSL version

Klee Dienes (1):
tls_ctx_load_ca: Improve certificate error messages

Max Muster (1):
Remove duplicate cipher entries from TLS translation table.

Peter Sagerson (1):
Fix configure interaction with static OpenSSL libraries

Steffan Karger (7):
Do not pass struct tls_session* as void* in key_state_ssl_init().
Require polarssl >= 1.2.10 for polarssl-builds, which fixes CVE-2013-5915.
Use RSA_generate_key_ex() instead of deprecated, RSA_generate_key()
Also update TLSv1_method() calls in support code to SSLv23_method() calls.
Update TLSv1 error messages to SSLv23 to reflect changes from commit 4b67f98
If --tls-cipher is supplied, make --show-tls parse the list.
Add openssl-specific common cipher list names to ssl.c.

Tamas TEVESZ (1):
Add support for client-cert-not-required for PolarSSL.

Thomas Veerman (1):
Fix "." in description of utun.

for those interested

 

[cofetym]

Asuswrt-Merlin 374.42

Wireless seems quicker than 374.41

 

[Raiu]

I like how he drops these at night so when we get a hold of it we can upload it during a simi down time for the house and we can upgrade our equipment.

Thanks man! Seems a lot faster for some reason maybe it's just placebo

 

[chercm]

will monitor for any issues before installing on mine . hope it is more stable

 

[nostromo]

I have lost Comcast IPV6 with .42. I came from .41b1 worked fine except for the neighborhood overflow. Tried .41b2 and lost IPV6 too. I did a factory reset but to no avail. N66U

 

[jim769]

OK here is the deal here.. Merlin 374.42 So far works great AC68U comcast IPv6 works good. Speeds are right on par. 5Ghz led is fixed. Thanks Merlin !!

 

[jerry6]

so far working well on rt n-66

 

[sm00thpapa]

Using my WRT1900AC but uploaded this to my RT-N66U and RT-N66W and both work great with no issues. Thanks once again Merlin.

 

[Daybreak]

I have lost Comcast IPV6 with .42. I came from .41b1 worked fine except for the neighborhood overflow. Tried .41b2 and lost IPV6 too. I did a factory reset but to no avail. N66U

Howdy,
Standard operating procedures.... power down your cable modem, power down the router, wait a few minutes, power up your cable modem, power up your router. Have you gotten your ipv6 back?

(firmware changes sometimes makes it look like different hardware, so the cable modem doesn't understand what happened)

 

[Poseidon]

Is a Factory Restore/Clear NVRAM required when upgrading from Merlin's .41 to .42?

 

[sinshiva]

Is a Factory Restore/Clear NVRAM required when upgrading from Merlin's .41 to .42?

No.

4567890

 

[InkyRag]

Channel 161 is back when set to 40 MHz channel width.

 

[jim769]

Channel 161 is back when set to 40 MHz channel width.

Yes that's because this version is based on Asus 5656 code..

 

[RMerlin]

I like how he drops these at night so when we get a hold of it we can upload it during a simi down time for the house and we can upgrade our equipment.

Thanks man! Seems a lot faster for some reason maybe it's just placebo

Well I can't release during the day because I work

I do try to release on Friday or Saturday tho, to give the weekend for people to test and report back on it. If there's any last minute issue, then that leaves me some time to push out an emergency update before the week starts.

 

[RMerlin]

I have lost Comcast IPV6 with .42. I came from .41b1 worked fine except for the neighborhood overflow. Tried .41b2 and lost IPV6 too. I did a factory reset but to no avail. N66U

Comcast's IPv6 is like lottery. People reporting random issues with every single firmware release while at the same time other people reports no issue, so I've assumed by now that either Comcast's IPv6 support is very shaky on some specific network nodes, or there is some specific voodoo involved in getting it to work after any router reboot (which is most likely what triggers new issues, not the firmware update in itself).

I recommend doing a 5-10 mins power down of the modem for people having issues with native IPv6 support.

 

[L&LD]

Wireless seems quicker than 374.41

No doubt about it; browsing in general is quicker with magic 42.

RT-N66U

Especially with my banking sites the difference is night and day (and 41 was the quickest previously too).


Throughput is also ~5% quicker at a nominally worse location than where I normally test from in my home.

Thank you RMerlin, this made my weekend.