Report on Tues's Net - Please read - of interest to everyone
Thanks for your interest, And your participation last Tues. I am breaking this little report down into sections. This is the first. You have to know that the interest in the topic of discussion is much higher than anyone might have predicted. _What happened During the net last Tues.._ Let me say that this was originally configured for 60-70 nodes to connect. Should not have been_any_ issues. That is what I was expecting with the unknown of personal/Pi nodes that are quite numerous now if you have not looked. Being a repeater or a personal Pi, ...it's a stream of the same size . A guideline for connecting was published. Those that connected in advance adhered to it. I made one balancing move in the beginning before the net and nobody even knew it happened. This was to keep nodes on par with their potential bandwidth. I had some options to balance even further. At some point near the beginning of the net, about the time Steve/N4IRS was having issues with the digital input and time-out on his own node, ...well...some took that as a node drop-out and began to switch connections from the planed path. At this same time, we had a influx of connections (while net was ongoing) LATE COMERS ! Expected some, but there were many. And most were not adhering to the connection plans. Within that part of the mess, a node had been connected up who came with 50+ nodes in a private network. I could not identify this node for the duration of the net. No way to see who is spewing large chucks of iax data on our network. It's not going to show in your allmon. This unidentified network normally runs with all their elemetry "OFF" (found out later) Most of you know how telemetry works in the iax2 and know your connected brings that data to your network. Same as they will receive yours. But this one node had telemetry data beyond belief and in loop. They don't know it, it's turned off on their systems. With most of our nodes are using telemetry, it became a additional overload load for Allison just trying to speak it. And the telemetry list stacked-up beyond comprehension. And very quickly. With all of the cpu activity and iax activity, I could not get a command in edge-wise. At this point I don't know where the iax data is coming from or that it is a major factor in the eventual overload of 29999. I had figured that much of it was coming from one node because it was the same stuff over and over. It's a funny story how I discovered the who but I'll save that for some other dicussion. I have some very strong feelings about this - 'running a large network without any telemetry and connecting to other large networks"you better be sure your network is clean, but I'll keep that under control for now. Everyone should realize that when you connect to others, you bring any problems your system has to the network you connect to. Even if it is not noticeable to you when connected to anything. I take full responsibility for this failure. But I want to share some of the blame on some others. Those that will not get with the plan,_especially when late._ And those that will not stick to the plan making matters worse. Contingency's were in place but you can not make changes when everyone is moving to who knows where. And that iax data stream kept me from changing anything anyway. I needed some direct communication with Mike and Steve to make them aware of things during the net process and pause them until we made adjustments. But this came about very suddenly and snowballed. . Normally we would account for some of these late connections but there was a large number of them. -- Most of you can match that from what you seen and heard to understand this better. But on my next note to group, I will post my plans for the next net. They are very strong. I will post by Saturday/Sun. _Then some new thoughts for discussion going forward in this new age of personal connectable network systems_. We have twice as many nodes now as just 3-4 years ago. It's still growing at that rate++. The older days when repeater systems were the dominate bunch are over. Repeater nodes were pretty much predictable. Personal nodes are not. _We need to have this discussion before we grow to big for it and things get to messy to keep the good perception we recieve from others at this point.._ Watch for my next post on this issue and you will see clearly how I am going to deal with these issues going forward. It may be just one example of many to follow of how to avoid these problems in the future. And know that every one of these issues will be fully addressed in the next net. Since we can no longer judge the size of interest and participation with any real degree of accuracy anymore. Thanks for reading the whole thing ! I welcome responsible comments at this time. But I would ask you to wait and see the plan going forward. These nets that 'anyone produces' are going to get larger and we need a plan. It is my desire to involve as many going forward as we can. Because others will need the practice of knowing how to properly configure and run such a net structure that continues to grow. And it is growing rapidly. The more that have a feel for this, the better we all are. Myself included. ...mike/kb8jnm
I could not use my node that night, so I connected via Echolink as a published option. I was on early, but was booted twice when it began. I gave up. Disappointed. I wonder if borrowing the Alaska net or some other might have worked. Cheers Bill - KB9WKR Sent from my iPad
On Jun 21, 2018, at 3:05 PM, Mike <mm@midnighteng.com> wrote:
Thanks for your interest, And your participation last Tues. I am breaking this little report down into sections. This is the first.
You have to know that the interest in the topic of discussion is much higher than anyone might have predicted.
What happened During the net last Tues..
Let me say that this was originally configured for 60-70 nodes to connect. Should not have been any issues. That is what I was expecting with the unknown of personal/Pi nodes that are quite numerous now if you have not looked.
Being a repeater or a personal Pi, ...it's a stream of the same size . A guideline for connecting was published. Those that connected in advance adhered to it.
I made one balancing move in the beginning before the net and nobody even knew it happened.
This was to keep nodes on par with their potential bandwidth. I had some options to balance even further. At some point near the beginning of the net, about the time Steve/N4IRS was having issues with the digital input and time-out on his own node, ...well...some took that as a node drop-out and began to switch connections from the planed path.
At this same time, we had a influx of connections (while net was ongoing) LATE COMERS ! Expected some, but there were many. And most were not adhering to the connection plans.
Within that part of the mess, a node had been connected up who came with 50+ nodes in a private network. I could not identify this node for the duration of the net. No way to see who is spewing large chucks of iax data on our network. It's not going to show in your allmon. This unidentified network normally runs with all their elemetry "OFF" (found out later) Most of you know how telemetry works in the iax2 and know your connected brings that data to your network. Same as they will receive yours. But this one node had telemetry data beyond belief and in loop. They don't know it, it's turned off on their systems. With most of our nodes are using telemetry, it became a additional overload load for Allison just trying to speak it.
And the telemetry list stacked-up beyond comprehension. And very quickly.
With all of the cpu activity and iax activity, I could not get a command in edge-wise. At this point I don't know where the iax data is coming from or that it is a major factor in the eventual overload of 29999. I had figured that much of it was coming from one node because it was the same stuff over and over. It's a funny story how I discovered the who but I'll save that for some other dicussion. I have some very strong feelings about this - 'running a large network without any telemetry and connecting to other large networks"you better be sure your network is clean, but I'll keep that under control for now.
Everyone should realize that when you connect to others,
you bring any problems your system has to the network you connect to. Even if it is not noticeable to you when connected to anything. I take full responsibility for this failure. But I want to share some of the blame on some others.
Those that will not get with the plan, especially when late. And those that will not stick to the plan making matters worse.
Contingency's were in place but you can not make changes when everyone is moving to who knows where.
And that iax data stream kept me from changing anything anyway. I needed some direct communication with Mike and Steve to make them aware of things during the net process and pause them until we made adjustments. But this came about very suddenly and snowballed.
. Normally we would account for some of these late connections but there was a large number of them.
-- Most of you can match that from what you seen and heard to understand this better.
But on my next note to group, I will post my plans for the next net. They are very strong. I will post by Saturday/Sun. Then some new thoughts for discussion going forward in this new age of personal connectable network systems.
We have twice as many nodes now as just 3-4 years ago. It's still growing at that rate++. The older days when repeater systems were the dominate bunch are over. Repeater nodes were pretty much predictable. Personal nodes are not. We need to have this discussion before we grow to big for it and things get to messy to keep the good perception we recieve from others at this point..
Watch for my next post on this issue and you will see clearly how I am going to deal with these issues going forward.
It may be just one example of many to follow of how to avoid these problems in the future. And know that every one of these issues will be fully addressed in the next net. Since we can no longer judge the size of interest and participation with any real degree of accuracy anymore. Thanks for reading the whole thing ! I welcome responsible comments at this time. But I would ask you to wait and see the plan going forward. These nets that 'anyone produces' are going to get larger and we need a plan.
It is my desire to involve as many going forward as we can. Because others will need the practice of knowing how to properly configure and run such a net structure that continues to grow. And it is growing rapidly. The more that have a feel for this, the better we all are. Myself included. ...mike/kb8jnm
_______________________________________________ App_rpt-users mailing list App_rpt-users@lists.allstarlink.org http://lists.allstarlink.org/cgi-bin/mailman/listinfo/app_rpt-users
To unsubscribe from this list please visit http://lists.allstarlink.org/cgi-bin/mailman/listinfo/app_rpt-users and scroll down to the bottom of the page. Enter your email address and press the "Unsubscribe or edit options button" You do not need a password to unsubscribe, you can do it via email confirmation. If you have trouble unsubscribing, please send a message to the list detailing the problem.
participants (2)
-
Mike -
Te Bussell