[Fork] Asuswrt-Merlin 374.43 LTS releases (Archive)

[john9527]

It's also possible to redirect syslog to /tmp by

$ rm /jffs/syslog.log && ln -s /tmp/syslog.log /jffs/syslog.log

The link will survive across reboots.

Not a good thing to do....

First...the way this works is that a copy is made of /tmp/syslog.log to /jffs. So you've essentially created a recursive symlink

Second, if you look you'll see that the symlink is removed on a reboot...it may take a minute or more after the reboot for it to happen.

 

[Builder71]

This is highly interesting, anyone try this yet?

That exact setting caught my interest.
If I want to try -1dB or -2dB, what nvram value do I have to use?

 

[knighthawk]

I took a quick look at the webpage.....if you can try something. If you disable prefix delegation (DHCP-PD) and assign a LAN address I think it may work. Not familiar enough with IPv6 to know if this is something you can do.

I could technically (and just tried it expecting what would happen) but then Comcast will not issue me my ipv6 IP prefix or even a wanip, which sort of negates the whole point - unless as I suspect you just want to know if dhcp6s's conf is then generated correctly when using say a static address (or native set with no -pd set) and stateles-dhcpv6 is selected.

In both 'for debugging purpose' cases (static\native with no pd) & removing any personal workarounds the result is ipv6_dhcp_start and ipv6_dhcp_end up having the full address(ie ipv6_prefix::1000) , and by extension a resulting perfectly correct and working stateful dhcps6 config (minus of course no inet-access in this test) after a reboot.

I think that would tend to indicate to me Asus in general is assuming those start\end nvram var's DO contain the full start\end addresses when they go to use them elsewhere (services.c previously mentioned) but in the case of DHCP-PD being enabled they are not being set, or not being reset at the right time via webpage or other events, ie after dhcp6c obtaining an address (at least in dhcp-pd mode anyway).

So perhaps some progress in at least narrowing that in cases other than dhcp-pd those start\end vars are expected to be full addresses and not the ::1000 and ::2000 that end up there in the case of dhcp-pd being enabled.

 

[john9527]

In that both cases (static\native with no pd) & [removing any personal workarounds] the result is ipv6_dhcp_start and ipv6_dhcp_end up having the full address(ie ipv6_prefix::1000) (and a resulting perfectly working stateful config (minus of course inet-access in this test) after a reboot.

Ok....this confirms I'm at least in the right neighborhood.....looks like maybe a bad assumption on their part. I'll dig in a bit more now and see where it makes sense to try a fix.

 

[knighthawk]

Ok....this confirms I'm at least in the right neighborhood.....looks like maybe a bad assumption on their part. I'll dig in a bit more now and see where it makes sense to try a fix.

Oh you're definitely in the right neighborhood, if I can do anything else to help or test let me know , I try to help myself so to speak as much as I can, but then trying to compile the firmware of a system I really don't know the ins and outs of without screwing up... well frankly kind of scares the sh1t out of me at the moment. That said I'll try and put a second set of eyes on the webpage and related stuff again and see if I can come up with anything that might be of help, if nothing else I might learn something in the process.

Thanks.

 

[netware5]

Hi guys, I'm following a conservative policy concerning FW updates of my N66. That mean I try to not change FW too frequently. So now it is time to move from Merlin's .42 to newer one. Now the big question is which FW to choose - Merlin's .48 or this beautiful fork with all security fixes ported from last Merlin's. As this fork incorporates almost all security fixes from last Merlin FW my understanding is that the *MAIN* difference between .43 fork and later Merlin's FWs is the ability to change region related Wi-Fi settings. Am I right? As my flat is small and the surroundings are not very crowded by other Wi-Fi networks I am fine with my CFE regional settings and the ability to change Wi-Fi channels and power output is not very important for me. The other significant difference is the SNMP abilities, but they are also not so important for me. From the other point of view this fork originates from very stable .43 which is based on .374 Asus code. My feeling is that the next .376 generation and also all Merlin's .43+ FWs are more oriented to the newer Asus routers and do not give too much improvement to my old N66.

So I will appreciate any advices what to do - to stay with Merlin or to switch to this fork.

 

[Deleted member 22229]

I would use the fork for the N66 the 376 builds really take away from signal and range on the N66 and offer nothing other then some eye candy in the UI. I have been using Merlin/John fork since it came out on my N66 and they have all worked great. Just my opinion as always YMMV.

 

[netware5]

I would use the fork for the N66 the 376 builds really take away from signal and range on the N66 and offer nothing other then some eye candy in the UI. I have been using Merlin/John fork since it came out on my N66 and they have all worked great. Just my opinion as always YMMV.

Thank you for the prompt response. That is the clue - *RANGE*. Does the range issues related only to the region in CFE or there is a general decrease? Until now I have no range problems as my flat is small. If the only other difference is the "eye candy in the UI" I will switch to the Jonh's fork, but I am interested if there are any other differences in stability and functionality.

 

[Deleted member 22229]

Thank you for the prompt response. That is the clue - *RANGE*. Does the range issues related only to the region in CFE or there is a general decrease? Until now I have no range problems as my flat is small. If the only other difference is the "eye candy in the UI" I will switch to the Jonh's fork, but I am interested if there are any other differences in stability and functionality.

Here is the thread to the latest Merlin 376.48_1 release. All you can do is try them both and see what you like.

http://forums.smallnetbuilder.com/showthread.php?t=20605

 

[netware5]

Here is the thread to the latest Merlin 376.48_1 release. All you can do is try them both and see what you like.

http://forums.smallnetbuilder.com/showthread.php?t=20605

Thanks, I am reading the Merlin's threads constantly. Unfortunately since SDK6 the postings about range and signal issues are prevailing, so it is really difficult to focus on other functionality. Or may be I am too lazy in reading?

OK, may be I'll give a try of .48 and if unsatisfied will revert to John's .43 fork. Thanks again.

 

[Deleted member 22229]

Thanks, I am reading the Merlin's threads constantly. Unfortunately since SDK6 the postings about range and signal issues are prevailing, so it is really difficult to focus on other functionality. Or may be I am too lazy in reading?

OK, may be I'll give a try of .48 and if unsatisfied will revert to John's .43 fork. Thanks again.

Yep thats what i would do if i was undecided. As you stated your in a small apt so the reduction in range may not really matter in your case. Flashing back and forth is easy and quick. Remember to do a factory reset when you upgrade to any new firmware.

 

[sidamos]

What are the main advantages of this firmware vs. the last Merlin sdk5 version for the N66U?

Thanks!

 

[Deleted member 22229]

What are the main advantages of this firmware vs. the last Merlin sdk5 version for the N66U?

Thanks!

For one and foremost many many security fixes. SDK5 has not been used in quite some time. Bottom line SDK5 did offer excellent signal output but is just not safe to run any longer.

 

[ml70]

That exact setting caught my interest.
If I want to try -1dB or -2dB, what nvram value do I have to use?

Lower 6 bits for full db's is interpreted as a signed number, so your range for positive gain is 0-31, for negative gain -1 - -32. Maximum positive value with quarter db bits (7 and 8) on is 1101 1111 = 223 decimal.

 

[Deleted member 27741]

Hey, john9527- I have an idea for the firmware or scripting.

Do you think it would be cool if login/logout events for the webUI and SSH would show up in the log? I think VPN events do show up in the log, but it would be cool if we could have login/logout events for other stuff end up there too.

I'm sure there is a script way to do this or maybe it is already possible and I'm missing something. I tried maxing out the verbosity of syslog- no dice.

 

[john9527]

Hey, john9527- I have an idea for the firmware or scripting.

Do you think it would be cool if login/logout events for the webUI and SSH would show up in the log? I think VPN events do show up in the log, but it would be cool if we could have login/logout events for other stuff end up there too.

I'm sure there is a script way to do this or maybe it is already possible and I'm missing something. I tried maxing out the verbosity of syslog- no dice.

Telnet logins are logged, but not logouts.

Nov 9 07:58:47 login[8801]: root login on 'pts/1'

SSH logins/logouts are logged by dropbear which provides the SSH service.

Nov 9 08:00:47 dropbear[8818]: Child connection from 192.168.1.100:14867
Nov 9 08:00:47 dropbear[8818]: Password auth succeeded for 'admin' from 192.168.1.100:14867
Nov 9 08:01:36 dropbear[8818]: Exit (admin): Exited normally

You are correct that login/logout for the webui is not done....that's something I can take a look at.

 

[Deleted member 27741]

______SWEET!______

 

[ualdrivr]

Any guys or gals here with iPhone 6 or 6 plus, experiencing random reboots, while the phone is on the Asus 68U Network, either 2.4 or 5 GHz with John's latest fork? Is there something going on with iOS 8.1 and Asus hardware/Firmware. I've been out on a trip and my 6 is on several different hotel Networks without a reboot. When I'm home sometimes I get two or three in a day, sometimes more, it's nuts!!!

 

[john9527]

More than a few are having problems with IOS 8.1 and disconnects. See this thread http://forums.smallnetbuilder.com/showthread.php?t=20497

One thing that has seemed to help is to set Regulatory mode to 802.11d+h on the Wireless/Professional tab (the default is Off).

 

[RocketJSquirrel]

One thing that has seemed to help is to set Regulatory mode to 802.11d+h on the Wireless/Professional tab (the default is Off).

I have read that after January 01, 2015, the FCC will not allow the use of 802.11d.