[Jeff Hatch]

Hello Max,

When using the cat command in your example it looks like there is a space missing between ‘cat’ and ‘/etc/issue’. I get the same error when issuing the same thing. With a space after ‘cat’ it will display the contents of the argument ‘/etc/issue’.

admin@mtcdt:~$ cat /etc/issue
     _        _____     ____
    / \\      | ____|   |  _ \\
   / _ \\     |  _|     | |_) |
  / ___ \\  _ | |___  _ |  __/_
 /_/   \\_\\(_)|_____|(_)|_|  (_)

MultiTech Systems Application Enablement Platform with mLinux GNU/Linux

mLinux 6.0.2 \n \l

Version: 6.0.1
Date: 2022-09-01T18:08:36

Jeff

[Jeff Hatch]

Hello Stephen,

If this Access Point is running mPower and not mLinux, it will not have a default username and password when powered up initially. It will be in commissioning mode and you will need to login through the Web UI and create the initial administrative user and corresponding password.

Thank You,

Jeff

[Jeff Hatch]

Hello Gabriel,

The Lora Network Server package for mLinux 6.0.1 can be found at:

In the package for the network server the init script should also be present.

Thank You,

Jeff

[Jeff Hatch]

Hello Gabriel,

Is your device running mLinux? What version of software is running? What kind of device is it? There are mLinux feeds with packages for pretty much everything “mLinux” at https://www.multitech.net/mlinux/feeds/

Thank You,

Jeff

[Jeff Hatch]

Hello Rina,

Can you check to make sure that the Start script in your application is executable and anything that it is going to try to execute is executable by root? Make sure that once it is installed that the Start script and other executables have execute permissions set on the files.

Jeff

[Jeff Hatch]

Hello Daniel,

Do you have access to the debug console set up? If not, I will have to get into the office and get my Conduit 300 device. I think you can interrupt the boot and drop into u-boot if u-boot is working. If you do successfully drop into u-boot you can try the following commands:

The following commands will do the trick.
=> setenv partition_swap_f 1
=> boot

The Conduit 300 has two boot partitions. The first command causes the boot process to attempt to boot the alternative partition. If the rootfs partition hasn’t been messed up on that boot partition you may be able to boot into mPower.

Otherwise, the device may need to be RMA’ed if you cannot boot into u-boot since at that point it takes the manufacturing process to reflash the device.

Thank You,

Jeff

[Jeff Hatch]

Hello Daniel,

The Conduit 300 will only run code signed by Multitech. There is a root of trust that consists of a key that is burned into the CPU and cannot be altered.

Do you have bitbake layers that support the Conduit 300 device? It is not the same Atmel CPU. It is a TI Sitara, so they are different generations of instruction sets.

I don’t believe that there is any bsp support within the mLinux release available publicly.

Thank You,

Jeff

[Jeff Hatch]

Hello Dániel,

I think your proposal should work in theory. However, I would warn that the reason the logs are not persistent on the device is that when under heavy logging load, the NAND flash can start having issues that can cause writes to fail and various other negative affects.

If the system is not logging heavily it may work fine.

Thank You,

Jeff

[Jeff Hatch]

Hello Rolf,

That probably will work also (I think I used that option several years ago, but it has been quite a while) to install the dashboard code in /var/persistent. I am not sure how exactly it will integrated with the Node-RED, but there will be some way to associate the Node-RED package with the dashboard if you need to.

The /var/persistent directory is definitely where you want things to be so that they don’t get deleted on firmware upgrade. Since that isn’t a standard directory that packages may look into, there may be more tweaking to get things to work out of that directory with the dashboard package.

Jeff

[Jeff Hatch]

Hello Rolf,

One thing I did a long time ago, so I’m not sure how it will work today, is to use “npm install ” where folder is the location of the actual npm package and it creates a link in the “current project” which I think is the default npm install location. Hopefully it would work such that you would put the dashboard package in a directory with more space and a sym link would get created to it in the project where you want it to install.

Thank You,

Jeff

[Jeff Hatch]

Hello Rolf,

You have a device running mPower 5.3.3. You can add things to the root partition, however, that will be deleted on upgrade due to the fact that an overlayfs is being used. In summary, the overlayfs is a “delta” of all the changes added/written to the root partition. The actual root partition is read-only. The overlayfs is something you’re better off letting the system handle.

There should be a /var/persistent directory on the device. That directory is created in the same partition as the overlayfs, but it is handled separately and not deleted on upgrade. In that partition you will see the Node-RED application installed if you have installed it. That is where you could theoretically install the dashboard also.

Thank You,

Jeff

[Jeff Hatch]

Hello Rolf,

1) Please confirm you are using mLinux and tell us which version.
2) If the device is running mPower, there will be a different way to handle this also depending on the version.

Thank You,

Jeff

[Jeff Hatch]

Hello,

Normally the registration is complete by the time the check you see in the log is done. The delay is being caused either due to the Telit Cellular radio, Cellular signal quality or the provider side.

The mPower code waits 10s between checks for successful registration by the radio. Registration with the network is taking ~2 minutes to complete in this case. I am in America, but in my experience registration does usually only take around one to three seconds. In my experience, the behavior you’re seeing is probably something signal or network related.

If you need additional assistance with this, I would recommend contacting Multitech support at https://support.multitech.com

Jeff

[Jeff Hatch]

Hello Martin,

I figured out how mLinux may be thwarting your editing via script. There is a script “/etc/init.d/config” that creates links to all the configuration files since I believe that /etc is not writable on some systems (I am not sure why mLinux is behaving in this manner other than on mLinux upgrades the /var/config directory and partition there don’t get overwritten). Anyway, I believe that your script is actually overwriting the link and not the file the link points to, ie. instead of editing the file /etc/network/interfaces->/var/config/network/interfaces, it is reading the resolved linked file “/var/config/network/interfaces”, editing it, and then writing to /etc/network/interfaces creating a new file. This works as expected.

However, upon reboot, the /etc/init.d/config script runs and since /etc/network/interfaces is not a link it blows /etc/network/interfaces away and creates the link described in the previous paragraph.

To address this try editing the file in /var/config/network/interfaces instead. That way it should persist.

Thank You,

Jeff

[Jeff Hatch]

Hello Martin,

Could you post the code that you are trying to do this with, both the “replacing” code and the sed code? Maybe someone can help you fix the code if there is something incorrect.

Jeff

[Jeff Hatch]

Hello,

There is also an Ethernet interface that could be configured as a WAN with a default route and DNS. If Cellular is disabled and the Ethernet interface is configured as WAN with default route and DNS, the Ethernet interface will get used. Assuming that the network server data has to go over the Internet (usually it does for most applications).

Jeff

[Jeff Hatch]

Hello,

The Multitech Conduit Access Point (AP) does not have a WiFi module from what I know. The AP for that device refers to the LoRa WAN access point functionality.

Is the case device case blue or white?

Jeff

[Jeff Hatch]

Hello,

Please contact Mutitech Support at https://support.multitech.com/ for help with this.

Thank You,

Jeff

[Jeff Hatch]

HEllo,

Please talk with Multitech support at https://support.multitech.com if you would like to get a radio firmware update for the LE910-NA1 from Multitech.

Thank You,

Jeff

[Jeff Hatch]

Hello Denis,

A feature improvement request has been created. When this change would be made available is TBD.

Thank You,

Jeff

[Jeff Hatch]

Hello Rafael,

What I mean by “not supported” is that Multitech does not test downgrading mPower firmware. It may work, or you may run into issues where you have to factory reset the device.

Typically, downgrades of versions that are close together, ie. 5.3.3 -> 5.3.0 will work close to all the time. In the case you have experienced, downgrading between 5.3.3 to 5.2.1, there are changes to the configuration database collections that is causing an incompatibility between the configurations, and can only be resolved by a factory reset.

Hope that helps.

Jeff

[Jeff Hatch]

Hello Rafael,

For the device that you downgraded, and things were acting badly, I recommend, if you can, factory resetting the device. That will be tricky, though because it appears that the device is arbitrarily rebooting constantly from your description. If you can get it to stay up long enough to press the reset button for >30s, you might be able to get it to reset to factory defaults. It may also be possible to get it to factory default through the API. Downgrading is not a supported use case since the API and the configuration database have to change going forward, and at times, is incompatible with past versions.

I think it should be safe to say that your app should run fine (you’ll need to test it). If your application is working as a custom application using the Node-RED application version of node.js which I think it is, then the primary dependency is on the Node-RED custom application. Currently we have not had to rev the Node-RED custom application yet to run on the different 5.3.3+ versions of mPower. The risk is that if you upgrade to a future mPower version at some point you may it may break and need the Node-RED app updated. Unfortunately, that is hard to predict.

Thank You,

Jeff

[Jeff Hatch]

Hello Lawrence,

On the Conduit 300 there is a partition located in the file system at /var/persistent. This is a much larger partition, and it has the bonus that it does not get erased on upgrade, much like the /var/config. The intent of the /var/persistent area is to provide a larger area than /var/config.

Thank You,

Jeff

[Jeff Hatch]

Hello M.K.

What are the permissions on the binary file of the application that you are trying to execute? Also, the status.json file needs to be created by and updated by the running application. Is your application creating it?

Thank You,

Jeff

[Jeff Hatch]

Hello Manuel and Mr. Mentor,

At this time that feature is not supported. I have added a feature/enhancement request for this to be added to mPower.

Thank You,

Jeff

[Jeff Hatch]

Hello Marco,

On comment on the INPUT rules: we are not sure that you need to specify the out interface on the rules. I am not positive, but that seems like it might cause issues. In the iptables man page for the -o (–out-interface) option it really only specifies using that option on FORWARD, OUTPUT, and POSTROUTING chains. See https://linux.die.net/man/8/iptables and specifically:

-o, --out-interface [!] name
Name of an interface via which a packet is going to be sent (for packets entering the FORWARD, OUTPUT and POSTROUTING chains). When the "!" argument is used before the interface name, the sense is inverted. If the interface name ends in a "+", then any interface which begins with this name will match. If this option is omitted, any interface name will match.

I am not sure exactly what effect specifying the output interface on the input rule would have in this situation.

Thank You,

Jeff

[Jeff Hatch]

There are some options for restarting the app. The app-manager program can for one. The “app-manager –command restart …” command actually runs the Start script with the restart argument.

If you want to skip using app-manager you could use the “Start restart” directly, but app-manager then may not accurately capture the status of the application.

So, from within your application you could use a call to app-manager to restart, making sure to background app-manager.

[Jeff Hatch]

Hello Tom,

The logging from AppManager appears to be normal in the sense that the events you see are just the “appmanager –status” running to report the application status. Is your script generating a pid file?

On the MQTT and Mosquito front. When I have worked with MQTT and Mosquitto, I have run into cases where the MQTT connection will timeout or disconnect regularly. When using libmosquitto with C++, the library seemed to have a reconnect behavior.

In your script do you check to verify that you are still connected and maintain the connection when it becomes disconnected? That is one possibility that would cause your app to stop working.

Thank You,

Jeff

[Jeff Hatch]

Hello,

What version of code is your MTCDT running?

Thank You,

Jeff

[Jeff Hatch]

Hello Johny,

Please create a portal case at https://support.multitech.com where they can probably help you with this. Performing SAM-BA requires access to the board (opening the case).

Is it possible to do a factory reset with these devices by pressing the reset button? You could try holding the reset button >30s and see if the devices default and are accessible on 192.168.2.1.

Thank You,

Jeff

[Author]

Posts