The time to upgrade to 3.0 was quite some time ago.

1.0.0 is a problem just waiting to happen.  The issues with stability seen in 1.1.x exist in 1.0.0, it is just the mesh traffic on those networks is generally smaller, but get enough nodes on the network and those systems are expected to crash.

In addition security flaws in 1.0.0 are also a reason to upgrade.

In regards to your comments on patching, there will always be new items to patch, I'm finding issues still today that existed in 1.0.0 and before. These are not your average radio, they are computers running dedicated software sets. One should not expect to throw a node up and just ignore it. While mountain top nodes may not need the newest code even they have needs to get updated over time, user nodes at houses and field locations (served parties) really do have a much higher need to be on newest firmware as each iteration brings new features and makes the user experience better.

Reboot is probably the best/easiest way.

I ran across this issue documented when working two other tickets (this one has been created as BBHN->ticket:78 ) as its overall a flaw in that older services do not get deleted from the backend files which causes issues like this.

Hey there guys, Glen (k6gse) from NALCO gave a couple of appearances at the local clubs around here and sounds like some interesting stuff. I've got a bullet and have an 8dB gain omni in the mail and will probably have some time next weekend to see what's visible from here.

I've got some decent LOS over Hayward and that's San Mateo over the bay. San Mateo will probably be too far without a directional though.

Is there anyone in my area here? (I'd look on the map but it doesn't appear to be available, the site here says it's private.) Should have some sort of a report next weekend, if anyone has stuff that isn't powered on let me know and we can try to set something up.

Dear Members,

Since I am Chicken when it comes to upgrades and my 27 node BBHN network is still on V1.0...

I am curious if Version 3.0 is considered to be stable enough to move our network ? ...

We are stable now and I want to be stable when I move versions.   I keep checking the

Board but don't see a lot of comments either way.  So, I am Curious and want a an idea

about what others think before I jump into upgrading.  Some of my nodes are on mountain

tops which may or may not be easy to get to this time of year .   So, any comments Yea or Neh?

Upgrading this BBHN is a tremendous job and I only want to it if it is worth it. 

73 Dave WA5DJJ

Dear Development group,

Since I am leading the charge of the Las Cruces, NM BBHN and I am a Chicken when it comes to

upgrades, (I only want to do it once in a while, We are still on 1.0 ),  Is 3.0 stable enough for us

to move our whole BBHN to this version?  I have been watching this board for comments on version

3.0 and have not heard much.   Since this is going to be a lot of work for our group to get done

since we have 27 nodes in some very diverse places that are sometimes hard to get to,  I don't want

to upgrade and then either have to drop back or play patch this and patch that to get it to work again.

So, what is the opinion?   Is it a go or a NO Go... ?

73 Dave WA5DJJ

You could use the following line with a loop in a script  or have 'cron' run it directly:

(date; echo '/links' | nc mesh_node 2006 | sed -e '1,5d') >> /tmp/stat.dat

Port 2006 is listened to by 'olsrd' and has some tidbits of info.   Change '/links' to '/all' and run manually to see what is available.   The stat.dat file can later be processed by 'awk' or whatever to extract the data desired. The data captured here is from the same source the mesh status page displays.

    -Rusty-

No program that I know of. But you can easily script something to make a remote connection and make a local copy of a status screen at regular intervals.

Two key components will be a time based job scheduler called cron. And you can use Lynx a command-line text based web browser to perform a screen dump.

http://en.wikipedia.org/wiki/Cron

http://en.wikipedia.org/wiki/Lynx_%28web_browser%29

Another option is to use an expect script that makes hourly ssh connects and issues the iwconfig command, etc.

No prerequisites (I would told ya if there were some, dewd :-). All the webpages are generated by CGI files on both the WRT54G's and Ubiquiti's and are written in Perl. Certainly there's no need to install another Perl interpreter. It's simply install package and play!

Is there a program (or something) that can periodically record the Mesh Status Page information?

I would like to monitor over a period of time just how the LQ of "Current Neighbors" goes up and down.

Especially if we have fog, rain, high winds, etc.

Thanks much.

Is there a program (or something) that can periodically record the Mesh Status Page information?

I would like to monitor over a period of time just how the LQ of "Current Neighbors" goes up and down.

Especially if we have fog, rain, high winds, etc.

Thanks much.

What prerequisites are there for this to work (on Ubuiquiti)? I assume I need to install a perl parser, but not sure which one. Any tips?

Thanks for the ipk!

What experience are you having with M9, M2, and M5 in the Salt Lake County area?

What experience are you having with M9, M2, and M5 in the Salt Lake County area?

What experience are you having with M9, M2, and M5 in the Salt Lake County area?

What experience are you having with M9, M2, and M5 in the Salt Lake County area?

What experience are you having with M9, M2, and M5 in the Salt Lake County area?

We're doing some testing with over a dozen nodes across three mesh islands connected via tunnel. So far so good, things work great.

However there is one thing I've noticed. When services are updated or deleted, the other nodes on the network don't notice this change until after a reboot.  New services are noticed quickly, but existing services changing or being removed doesn't get updated on the other nodes until they are rebooted.

Is there a way to force a node to refresh it's understanding of the network without reboot?

Thanks!

We're doing some testing with over a dozen nodes across three mesh islands connected via tunnel. So far so good, things work great.

However there is one thing I've noticed. When services are updated or deleted, the other nodes on the network don't notice this change until after a reboot.  New services are noticed quickly, but existing services changing or being removed doesn't get updated on the other nodes until they are rebooted.

Is there a way to force a node to refresh it's understanding of the network without reboot?

Thanks!