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

[john9527]

2.4ghz will truncate the SSID

What is your 2.4GHz SSID? Also your 2.4GHz client names? Any special characters (characters other than alphanumerics, dash and underscore)?

 

[Badlandz]

What is your 2.4GHz SSID? Any special characters (characters other than alphanumerics, dash and underscore)?

Yes, both SSIDs have special characters. Other than alphanumerics are these * < / - @ among the SSIDs.
The SSIDs for 2.4ghz and 5ghz are actually the same but in reverse order. They are also either max length(32 charcters?) or close to it. If that is the case it does not affect the 5G portion of the display.

Same SSIDs in V10 and to my knowledge prior versions of fork displays correctly on System Log>Wireless Log.

As far as 2.4ghz client names they are mostly unnamed and therefore just show the MAC ID. The ones that are named are simple 2-3 alphanumeric characters. Example: RBR, RLR, WP, XB1

 

[Techno Pride]

No, it's not that simple.

The RT-AC56S is a single core CPU. The firmware needs to use different versions of numerous precompiled binary blobs, all the RT-AC56U files are compiled for dual core CPUs. Quite a few of these components will crash on a single core CPU.

ok then. How much effort are we talking about here?

a) We've to recompile the necessary files
b) Modify the build file to pick up the correct sdingle core binaries?

 

[RMerlin]

ok then. How much effort are we talking about here?

a) We've to recompile the necessary files
b) Modify the build file to pick up the correct sdingle core binaries?

You have to modify the actual build process, and dig through Asus's GPL releases to retrieve those single-processor versions. They are closed-source, so you can't compile them yourselves.

It's technically doable, but a good amount of work, one which I don't intend to put into it, since that model is not available here, and only sold in very limited markets.

I only focus on high-end models, as I don't have the time to support too many models at the same time.

 

[sbailey4]

Well did a factory reset this weekend working on a small issue I was having ( turns out not the router) and noticed that the Max TCP connections were reset back to 16387 again not the 300,000 as default that was addressed in this build (or 13j). Something apparently still off at lease for the AC66.

 

[Blinkyz]

Well did a factory reset this weekend working on a small issue I was having ( turns out not the router) and noticed that the Max TCP connections were reset back to 16387 again not the 300,000 as default that was addressed in this build (or 13j). Something apparently still off at lease for the AC66.

Same here on n66, after Factory RESET,max connection is set at 15967 and normally its 300000( like Asus and Merlin)

 

[john9527]

Yes, both SSIDs have special characters. Other than alphanumerics are these * < / - @ among the SSIDs.
The SSIDs for 2.4ghz and 5ghz are actually the same but in reverse order. They are also either max length(32 charcters?) or close to it. If that is the case it does not affect the 5G portion of the display.

Same SSIDs in V10 and to my knowledge prior versions of fork displays correctly on System Log>Wireless Log.

As far as 2.4ghz client names they are mostly unnamed and therefore just show the MAC ID. The ones that are named are simple 2-3 alphanumeric characters. Example: RBR, RLR, WP, XB1

It's the < and / (slash if it follows <) that are causing the problem. When I added the color coding, it converted from a simple textbox to HTML div sections so I could embed HTML tags. HTML tags have the form <tag>.....</tag> See the potential problem when it tries to interpret part of your SSID as a start or end tag? There are of course ways to code around it, but quite frankly it's just not worth the effort because there are many places in the code besides just my change that are effected. For example, the code uses < and > as field separators in some nvram variables.

Also, you may be getting hit other ways. You say the 5GHz SSID is the same as the 2.4 except in reverse....except that in your screen capture, the 5GHz SSID has been reset to the default of Asus5G (must have choked on it in another part of the code, as I mentioned as a possibility).

The entry of special characters is already blocked for client names and so forth, but we missed blocking the SSID. The next release will prevent the entry of these characters as part of the SSID.

This falls into the category of that old joke....

Patient-to-Doctor: Doctor, Doctor....It hurts when I do this!
Doctor-to-Patient: Well, then don't do that!

 

[derrickrozay]

What is the difference between Merlins official FW and a fork of it?

Also does this fork allow policy based VPN routing?

 

[Deleted member 22229]

What is the difference between Merlins official FW and a fork of it?

Also does this fork allow policy based VPN routing?

[Fork] Update for 374.43 available (V13E1)

 

[Badlandz]

It's the < and / (slash if it follows <) that are causing the problem. When I added the color coding, it converted from a simple textbox to HTML div sections so I could embed HTML tags. HTML tags have the form <tag>.....</tag> See the potential problem when it tries to interpret part of your SSID as a start or end tag? There are of course ways to code around it, but quite frankly it's just not worth the effort because there are many places in the code besides just my change that are effected. For example, the code uses < and > as field separators in some nvram variables.

Also, you may be getting hit other ways. You say the 5GHz SSID is the same as the 2.4 except in reverse....except that in your screen capture, the 5GHz SSID has been reset to the default of Asus5G (must have choked on it in another part of the code, as I mentioned as a possibility).

The entry of special characters is already blocked for client names and so forth, but we missed blocking the SSID. The next release will prevent the entry of these characters as part of the SSID.

This falls into the category of that old joke....

Patient-to-Doctor: Doctor, Doctor....It hurts when I do this!
Doctor-to-Patient: Well, then don't do that!

Your level of support is outstanding. A very small price to pay, I will simply change the SSID to avoid that conflit. This community at SNB is impressive.
Thanks John.

 

[RMerlin]

It's the < and / (slash if it follows <) that are causing the problem. When I added the color coding, it converted from a simple textbox to HTML div sections so I could embed HTML tags. HTML tags have the form <tag>.....</tag> See the potential problem when it tries to interpret part of your SSID as a start or end tag? There are of course ways to code around it, but quite frankly it's just not worth the effort because there are many places in the code besides just my change that are effected. For example, the code uses < and > as field separators in some nvram variables.

If you want to start escaping some of these, I recently added a new function to merlin.js which converts < and > into HTML entities (I needed it for the new system log pages, as hostnames could be set as "<unknown>").

 

[jlschwartz]

LATEST RELEASE: Update-13E1
14-July-2015
Merlin fork 374.43_2-13E1j9527
Download http://1drv.ms/1uChm3J
============================

New release is up! And hopefully will finally resolve some long standing issues.

The biggest items

• Updated OpenSSL to 1.0.2d to pick up a high sev security fix
• Fixed connectivity issues for PPPoE/IPv6
• Fixed updating to the new DH size on MIPS routers
• Fixed impact to IPv6 download speed with QoS enabled

And a few more...please take the time to read thru the first post for a complete list.

Enjoy!

I may be the only one experiencing this, but I find this release to be not as good as the last. I have an RT-N66U router and I'm not using any special setup. I find the range is not as good and the router seems to drop the wireless signal at random making the reliability an issue. As a result I went back to version 12. Just thought I should report this. Thanks for all of your work on this.

 

[zekesdad]

Ive had to manually reboot the router about once a day for past two builds because of losing internet access. V11 would last for weeks, but these past two havent so much. No special vpn or anything like that being used, only have a usb drive attached.

 

[john9527]

I may be the only one experiencing this, but I find this release to be not as good as the last. I have an RT-N66U router and I'm not using any special setup. I find the range is not as good and the router seems to drop the wireless signal at random making the reliability an issue. As a result I went back to version 12. Just thought I should report this. Thanks for all of your work on this.

Sorry, nothing I can do here....there is absolutely no change in the wireless subsystem between any of the fork releases.

 

[john9527]

Ive had to manually reboot the router about once a day for past two builds because of losing internet access. V11 would last for weeks, but these past two havent so much. No special vpn or anything like that being used, only have a usb drive attached.

You are so far the only one to report anything like this....anything in the system log? Without data I have nothing to look at.

 

[jlschwartz]

Sorry, nothing I can do here....there is absolutely no change in the wireless subsystem between any of the fork releases.

I understand. Yet, when I return to the previous version my system runs noticeably better. Just an anomaly in my router I guess. Thanks again for all of your work.

 

[Deleted member 22229]

I may be the only one experiencing this, but I find this release to be not as good as the last. I have an RT-N66U router and I'm not using any special setup. I find the range is not as good and the router seems to drop the wireless signal at random making the reliability an issue. As a result I went back to version 12. Just thought I should report this. Thanks for all of your work on this.

Just to try it i unhooked my 68R and installed my old N66U and updated to the latest build V-13E1 and did some testing and i can say with certainty there was no difference in signal,range or speeds with the new build and not one dissconnect. This must be something in your environment causing the change.

 

[mobileman88]

Ive had to manually reboot the router about once a day for past two builds because of losing internet access. V11 would last for weeks, but these past two havent so much. No special vpn or anything like that being used, only have a usb drive attached.

I have never experienced this with any of the versions. What model of router are you using? Are you able to generate some log files?

 

[wiz]

I may be the only one experiencing this, but I find this release to be not as good as the last. I have an RT-N66U router and I'm not using any special setup. I find the range is not as good and the router seems to drop the wireless signal at random making the reliability an issue. As a result I went back to version 12. Just thought I should report this. Thanks for all of your work on this.

I have a RT_N66u too, and 13e1 range is comparable to 12, no issues here.

 

[zekesdad]

I have never experienced this with any of the versions. What model of router are you using? Are you able to generate some log files?

absolutely.im on the rt-ac68u. what logs do you want?

edit: heres the syslog