amtm amtm 4.9 - the Asuswrt-Merlin Terminal Menu, June 30, 2024 (locked thread)

[Viktor Jaep]

I went to install RTRMON via AMTM via the RT option. It responded with:

RTRMON requires the Entware repository
installed. Enter ep to install Entware now.

So I entered EP and it found a compatible entware package on my attached USB drive:

Select device to install Entware to

1. /tmp/mnt/xxxxxUSB
Found compatible previous Entware
installation on this device.

Select device [1-1 e=Exit] 1

Running device checks on /tmp/mnt/xxxxxUSB
Device checks passed
_____________________________________________

This device contains a compatible Entware
installation, select what to do.

/tmp/mnt/xxxxxUSB

1. Reuse previous Entware installation.
This requires rebooting this router
after completion.
2. New Entware installation
3. Return to device selection

Enter selection [1-3 e=Exit]

Why does it think I need to install entware when there is already a compatible entware installation found? It doesn't appear I should need to do anything with respect to entware. Or does entware need to exist in a specific location (maybe /OPT/BIN)?

I was looking at the AMTM code to determine what would kick off that message, and if I'm not mistaken, if it can't find /opt/bin/opkg, then it assumes entware is not installed. I hope someone here can help guide you on what the right steps to take are in this case?

 

[ColinTaylor]

There must be a very simple reason for the difference between these numbers, I mean "Available 990.7M" vs. "Available space 2.41 GB". What's that?

As you have demonstrated this is not a problem with amtm which is showing the correct information. So perhaps not a question suited for this thread. In my case the information shown in the GUI is also correct. However, it is not dynamically updated. So if you copy data to the USB drive the space remaining is not correct until you refresh that page.

 

[thelonelycoder]

Why does it think I need to install entware when there is already a compatible entware installation found? It doesn't appear I should need to do anything with respect to entware. Or does entware need to exist in a specific location (maybe /OPT/BIN)?

No need to get bold with me, I can read normally formatted text just fine.

With that aside, let me explain how Entware works.

Entware is installed into a folder named entware on the selected attached storage. During setup this folder is mounted as /opt.
During each reboot the file /jffs/scripts/post-mount kicks in if an attached drive is found.
In that post-mount file I place a source link to /jffs/addons/amtm/mount-entware.mod which then remounts the entware folder as /opt.
If the file post-mount does not contain the link to the mount-entware.mod file or either of these are missing, Entware will not be available on the router.

This is exactly what happened to you. I helpfully detect that previous installation of Entware and offer it to re-use it during setup.
Makes sense, right? You should thank me for being so helpful instead of screaming at me.

 

[Wrkdbf_Guy]

I only bolded my text so as to distinguish it from copy/pasted text. No shouting intended. Sorry.

 

[thelonelycoder]

I only bolded my text so as to distinguish it from copy/pasted text. No shouting intended. Sorry.

No problem, I hope you understand now what and why amtm did what it does.

 

[JGrana]

This is exactly what happened to you. I helpfully detect that previous installation of Entware and offer it to re-use it during setup.
Makes sense, right? You should thank me for being so helpful instead of screaming at me.

This scenario has happened to me in the past. It was a welcome site to see it detected and even better that AMTM cleaned up my mess.

I do thank you.

BTW, Like the new Avatar @thelonelycoder

 

[thelonelycoder]

This scenario has happened to me in the past. It was a welcome site to see it detected and even better that AMTM cleaned up my mess.

I do thank you.

BTW, Like the new Avatar @thelonelycoder

Yours looks like an abandoned Netgear router, pushed aside to make way for an ASUS model.

 

[JGrana]

Indeed. I need to make a new one ;-)

 

[thelonelycoder]

amtm 4.4 is now available

What's new in amtm 4.4
- Adds support for scMerlin v2.5.0 j3 fork by @thelonelycoder. Existing installations will show an extra menu entry to use this fork, new installation use it by default. Thanks to @Martinski for the updates. See new scMerlin thread. Use the SSH UI to update scMerlin after running the shown command.
- Updated content of shared amtm whitelist.
- Corrects wrong EHLO (Extended Hello) string when mailing an attachment with curl. Thanks GitHub user tazio for reporting.
- Adds disk check dc notification to amtm UI if the word error is found during the last check. Additionally an email can be sent if that happens. Configure it in dc. Thanks to GitHub user SirFireNewt for the suggestion.
- Adds support for Entware-backports Repo by @maurer for armv7sf-k2.6 based routers like the RT-AC68U or RT-AC1900P. Existing installations can add the repo in ep, new installations use it by default.
- Corrects scraping the Dynamic LED control lc sunset/sunrise time from the weather.com website. They have changed their JavaScript code.

 

[vlord]

amtm 4.4 is now available

What's new in amtm 4.4
- Adds support for scMerlin v2.5.0 j3 fork by @thelonelycoder. Existing installations will show an extra menu entry to use this fork, new installation use it by default. Thanks to @Martinski for the updates. See new scMerlin thread. Use the SSH UI to update scMerlin after running the shown command.
- Updated content of shared amtm whitelist.
- Corrects wrong EHLO (Extended Hello) string when mailing an attachment with curl. Thanks GitHub user tazio for reporting.
- Adds disk check dc notification to amtm UI if the word error is found during the last check. Additionally an email can be sent if that happens. Configure it in dc. Thanks to GitHub user SirFireNewt for the suggestion.
- Adds support for Entware-backports Repo by @maurer for armv7sf-k2.6 based routers like the RT-AC68U or RT-AC1900P. Existing installations can add the repo in ep, new installations use it by default.
- Corrects scraping the Dynamic LED control lc sunset/sunrise time from the weather.com website. They have changed their JavaScript code.

I selected j3u to migrate over to your fork but it’s still showing 2.4.1 after update. Now when I enter j3u, I get this:

j3u
/usr/sbin/amtm: line 434: case_j3u: not found
admin@GT-AX6000:/tmp/home/root

Update: Removing 2.4.1 and reinstalling 2.5.0 fixed the problem.

 

[alan6854321]

Hi there,
Everything's working OK, but I found all these lines in the 'services-start' script

cat /jffs/scripts/services-start
#!/bin/sh

/jffs/scripts/YazDHCP startup "$@" & # YazDHCP
/jffs/AGT/scripts/send-wan-usage-mqtt & # AGT
/jffs/AGT/scripts/log-reboot-complete & # AGT
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/addons/scmerlin.d/S95tailtaintdns start >/dev/null 2>&1 & # scMerlin - tailtaintdns
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$

 

[visortgw]

Hi there,
Everything's working OK, but I found all these lines in the 'services-start' script

cat /jffs/scripts/services-start
#!/bin/sh

/jffs/scripts/YazDHCP startup "$@" & # YazDHCP
/jffs/AGT/scripts/send-wan-usage-mqtt & # AGT
/jffs/AGT/scripts/log-reboot-complete & # AGT
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/addons/scmerlin.d/S95tailtaintdns start >/dev/null 2>&1 & # scMerlin - tailtaintdns
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$

It appears that an instance of the "/jffs/scripts/scmerlin startup & # scMerlin$" line is appended to services-start file every time that the script is invoked, whether from invoking scmerlin via CLI or a reboot, indicating an issue with the logic that checks for its existence.

 

[JGrana]

Yes, your right. Every CLI call adds another one.
I also lost all my other Addons Web tabs - the only one showing is scmerlin.
I’m going to reboot and see if they come back.

Post reboot - all Addon web pages are back.
As @visortgw said, scmerlin added another startup line to services-start

 

[Martinski]

Hi there,
Everything's working OK, but I found all these lines in the 'services-start' script

cat /jffs/scripts/services-start
#!/bin/sh

/jffs/scripts/YazDHCP startup "$@" & # YazDHCP
/jffs/AGT/scripts/send-wan-usage-mqtt & # AGT
/jffs/AGT/scripts/log-reboot-complete & # AGT
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/addons/scmerlin.d/S95tailtaintdns start >/dev/null 2>&1 & # scMerlin - tailtaintdns
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$
/jffs/scripts/scmerlin startup & # scMerlin$

This bug was introduced on 2022-Jan-05 with changes in PR #22:

v2.4.1 by jackyaz · Pull Request #22 · jackyaz/scMerlin

scMerlin allows you to easily control the most common services/scripts on your router. scMerlin also augments your router's WebUI with a Sitemap and dynamic submenus for the main left menu of Asuswrt-Merlin. - v2.4.1 by jackyaz · Pull Request #22 · jackyaz/scMerlin

github.com

But those changes never made it to the master branch so they're found only in the "2.4.1" version from @JackYaz's develop branch. When @thelonelycoder forked the repository, it inherited the bug which eventually made it into his "2.5.0" master version.

I've submitted a PR for the fix on @JackYaz's develop branch. There's no develop branch in @thelonelycoder's repository so I have not submitted a PR to fix the bug there.

 

[elorimer]

I see it was merged into the master 17 minutes ago. But it looks like most of the changes are already in 2.5, but not all the sed lines.

 

[Martinski]

I see it was merged into the master 17 minutes ago.

Yes, I see now that @Jack Yaz has already merged the PR and updated his master branch as well so all source files in both develop & master are now in sync.

But it looks like most of the changes are already in 2.5, but not all the sed lines.

Sorry, I don't understand what you mean by "not all the sed lines."

If you're referring to the bug fix, as of this writing the @thelonelycoder's master branch is not yet fixed so there would be differences in the 2 echo commands that had the error. The grep & sed commands in that same function are fine - no need to modify them.

 

[dave14305]

The grep & sed commands in that same function are fine - no need to modify them.

Ideally, the seds would be updated to cleanup the errant lines from the buggy version (i.e. look for optional trailing $ before EOL).

 

[Martinski]

Ideally, the seds would be updated to cleanup the errant lines from the buggy version (i.e. look for optional trailing $ before EOL).

Ah, yes; very good point. The code should clean up after itself.

 

[elorimer]

Does the change in the script (adding the $) apply to Auto_ServiceEvent as well as Auto_Startup? e.g., lines 478, 496 in the 2.5 script?

 

[thelonelycoder]

Yes, I see now that @Jack Yaz has already merged the PR and updated his master branch as well so all source files in both develop & master are now in sync.


Sorry, I don't understand what you mean by "not all the sed lines."

If you're referring to the bug fix, as of this writing the @thelonelycoder's master branch is not yet fixed so there would be differences in the 2 echo commands that had the error. The grep & sed commands in that same function are fine - no need to modify them.

So theoretically I could revert using my repo to @Jack Yaz and relieve me of that burden of being responsible