Scribe and aimesh node

[maghuro]

I've an idea.
I'd love to receive my aimesh node (rt-ac86u) logs on my main router. And using scribe, put them in a separate window.

Is that possible? I know there's some kind of server/slave log thingy but I don't know where to start.

---
Also, my qnap has the same log server thing. I'd be great if I could also receive them on my router and put them in a separate window.

Thanks in advance for who kindly helps.

 

[ScottW]

It is entirely possible -- I have mine setup that way. The aimesh node forwards all log entries to the router, which uses syslog-ng (scribe) to organize them.

As best I recall, here's how I did it.

I installed Merlin on the aimesh node, then SSH'd to the node and enabled scripts:
nvram set jffs2_scripts="1"
nvram commit

I created a services-start script on the aimesh node with the following content (this kills syslogd and restarts it with forwarding to the router):
#!/bin/sh
PID=`ps | grep "syslogd" | grep -v "grep" | awk '{print $1}'`
'kill "$PID"`
/sbin/syslogd -m 0 -S -O /tmp/syslog.log -s 256 -l 6 -L -R 192.168.1.1
# (substitute your router's IP above if not 192.168.1.1)

Then I changed /opt/etc/syslog-ng.conf to process the remote source messages from the node:
add a new line source net { network( transport(udp) ); };
below the existing source(src); line, add a new line with source(net);

Now go to /opt/etc/syslog-ng.d, and add a source(net); line to any of the files you want to process the messages from aimesh.

Hope that helps.

 

[elorimer]

It's easier than that. There are several threads here on it.

On each, set your main router as a remote server destination. On the main router, create a source (net), include that in the main cycle, and then create a configuration triad to separate by node.

 

[elorimer]

Scribe - Logging Mesh nodes - Scribe 3.03

So, followed the instructions here: https://www.snbforums.com/threads/feature-to-support-aimesh-node-s-log.69868/ But I'm pretty sure I messed something up doing this as the logs from the mesh nodes (MasterBedroom and Bedroom) as well as the router are all intertwined both in UIScribe as well...

www.snbforums.com

 

[maghuro]

It is entirely possible -- I have mine setup that way. The aimesh node forwards all log entries to the router, which uses syslog-ng (scribe) to organize them.

As best I recall, here's how I did it.

I installed Merlin on the aimesh node, then SSH'd to the node and enabled scripts:
nvram set jffs2_scripts="1"
nvram commit

I created a services-start script on the aimesh node with the following content (this kills syslogd and restarts it with forwarding to the router):
#!/bin/sh
PID=`ps | grep "syslogd" | grep -v "grep" | awk '{print $1}'`
'kill "$PID"`
/sbin/syslogd -m 0 -S -O /tmp/syslog.log -s 256 -l 6 -L -R 192.168.1.1
# (substitute your router's IP above if not 192.168.1.1)

Then I changed /opt/etc/syslog-ng.conf to process the remote source messages from the node:
add a new line source net { network( transport(udp) ); };
below the existing source(src); line, add a new line with source(net);

Now go to /opt/etc/syslog-ng.d, and add a source(net); line to any of the files you want to process the messages from aimesh.

Hope that helps.

Thanks for your detailed help!

I followed step by step, made some minimal adjustments for my particular case, and it's working flawlessly.
Is it possible to do the same for a QNAP?

@elorimer thank you also! However, believe if you want, I don't have a spare USB stick to install entware on aimesh node

 

[elorimer]

You don't need entware on a node.

ssh into the node, change the nvram variable for the remote logger, commit, and that's it.

 

[maghuro]

You don't need entware on a node.

ssh into the node, change the nvram variable for the remote logger, commit, and that's it.

Wow so I read wrong on that thread.

Despite of that, the solution provided earlier did the trick for me....
If I had more than one node, I'd do that way. But as I only have one, and don't planing to have another one, the first solution was good

Thank you both!

 

[elorimer]

Yes, the solution in the thread is I think more complicated than necessary, even if you are doing heavy discarding of messages from the node. The main router can handle a lot more than it has to.