Topics

VE7CC - Spot Source


Curt Bowles
 

Anyone else having issues with VE7CC…

NO changes have been made to the setup under “SpotCollector/Config/Spot Sources” : 

SpotCollector has been unable to connect to VE7CC for the last 24 hours or so.

Keep getting this msg:

Winsock Error # 10060 - The attempt to connect timed out @ 16-Oct-20 1119


Host Address: dxc.ve7cc.net

Port: 23

Caption: VE7CC

Plus my user name and password 

--
Curt Bowles
VE3ZN


Charles N2SO
 

From CCusers



On Fri, Oct 16, 2020 at 7:24 AM Curt Bowles <curt.bowles@...> wrote:

Anyone else having issues with VE7CC…

NO changes have been made to the setup under “SpotCollector/Config/Spot Sources” : 

SpotCollector has been unable to connect to VE7CC for the last 24 hours or so.

Keep getting this msg:

Winsock Error # 10060 - The attempt to connect timed out @ 16-Oct-20 1119


Host Address: dxc.ve7cc.net

Port: 23

Caption: VE7CC

Plus my user name and password 

--
Curt Bowles
VE3ZN



--
73, Charles N2SO
 

--
73, Charles N2SO


Curt Bowles
 

Charles N2SO

Excuse my ignorance but what does "From CCusers" mean...

Nothing else in you msg
--
Curt Bowles
VE3ZN


Charles N2SO
 

Sorry about that. The image did not go through. VE7CC appears to be down. Possible weather problems. It was suggested to try another node like: N4ZKF-1 (+CW Skimmer)
Should be the same feed.

On Fri, Oct 16, 2020, 8:00 AM Curt Bowles <curt.bowles@...> wrote:
Charles N2SO

Excuse my ignorance but what does "From CCusers" mean...

Nothing else in you msg
--
Curt Bowles
VE3ZN


--
73, Charles N2SO


Curt Bowles
 

Hi Charles

That worked just fine this time...thanks for the feedback... appreciated.

73
--
Curt Bowles
VE3ZN


John Battin
 

I had the same THING FOR ABOUT 6 HOURS

 

Sent from Mail for Windows 10

 

From: Curt Bowles
Sent: Friday, October 16, 2020 6:24 AM
To: DXLab@groups.io
Subject: [DXLab] VE7CC - Spot Source

 

Anyone else having issues with VE7CC…

NO changes have been made to the setup under “SpotCollector/Config/Spot Sources” : 

SpotCollector has been unable to connect to VE7CC for the last 24 hours or so.

Keep getting this msg:

Winsock Error # 10060 - The attempt to connect timed out @ 16-Oct-20 1119

 

Host Address: dxc.ve7cc.net

Port: 23

Caption: VE7CC

Plus my user name and password 

--
Curt Bowles
VE3ZN

 


Charles N2SO
 

FB Curt. I'm glad that worked out. If you want skimmer spots just enter the command "set/skimmer" or if you use VE7CC software program, just check the skimmer box in the configuration window. Then re-start Spotcollector.

On Fri, Oct 16, 2020 at 9:31 AM Curt Bowles <curt.bowles@...> wrote:
Hi Charles

That worked just fine this time...thanks for the feedback... appreciated.

73
--
Curt Bowles
VE3ZN



--
73, Charles N2SO
 

--
73, Charles N2SO


Dave AA6YQ
 

Why is anyone interposing the "VE7CC user" program between SpotCollector and spot sources?

       73,

            Dave, AA6YQ


Henk Remijn PA5KT
 

It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.

Also useful when spotcollector is not used. Most of the time during contests, which is 95% of my operating.

I use Spotcollector for award chasing.

73 Henk PA5KT

Op 16-10-2020 om 18:53 schreef Dave AA6YQ:

Why is anyone interposing the "VE7CC user" program between SpotCollector and spot sources?

       73,

            Dave, AA6YQ


Dave AA6YQ
 

+ AA6YQ comments below
It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.

+ SpotCollector provides the ability to pre-filter cluster spots. See

<https://www.dxlabsuite.com/dxlabwiki/SpotDatabasePrefilter>

+ With SpotCollector running, the VE7CC user application adds no value, but forces incoming spots to be processed twice, unnecessarily increasing CPU loading and memory consumption.

           73,

                   Dave, AA6YQ

 


iain macdonnell - N6ML
 

On Fri, Oct 16, 2020 at 2:23 PM Dave AA6YQ <@AA6YQ> wrote:

+ AA6YQ comments below

It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.

+ SpotCollector provides the ability to pre-filter cluster spots. See

<https://www.dxlabsuite.com/dxlabwiki/SpotDatabasePrefilter>

+ With SpotCollector running, the VE7CC user application adds no value, but forces incoming spots to be processed twice, unnecessarily increasing CPU loading and memory consumption.
That's not really accurate - (AFAIK - it's been years since I've used
it) filters configured by the (VE7)CC User program are effected *on
the cluster node*, so the spots don't even traverse the network
(internet). This may make a significant difference to someone with
limited bandwidth, and would not consume CPU/memory on the client PC.
Those filters can be configured using commands/macros, but the CC User
tool provides a nice GUI.

73,

~iain / N6ML


Dave AA6YQ
 

* More AA6YQ comments below


It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.

+ SpotCollector provides the ability to pre-filter cluster spots. See

<https://www.dxlabsuite.com/dxlabwiki/SpotDatabasePrefilter>

+ With SpotCollector running, the VE7CC user application adds no value, but forces incoming spots to be processed twice, unnecessarily increasing CPU loading and memory consumption.
That's not really accurate - (AFAIK - it's been years since I've used
it) filters configured by the (VE7)CC User program are effected *on
the cluster node*, so the spots don't even traverse the network
(internet). This may make a significant difference to someone with
limited bandwidth, and would not consume CPU/memory on the client PC.
Those filters can be configured using commands/macros, but the CC User
tool provides a nice GUI. 

* The CPU and memory consumed by running the CC User app significantly exceed the cost of conveying ~100 byte spots over the network every few seconds. 

* Is there something wrong with SpotCollector's pre-filter GUI? There's a screen shot in the article cited above.

       73,

              Dave, AA6YQ


iain macdonnell - N6ML
 

On Fri, Oct 16, 2020 at 3:04 PM Dave AA6YQ <@AA6YQ> wrote:

* More AA6YQ comments below


It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.

+ SpotCollector provides the ability to pre-filter cluster spots. See

<https://www.dxlabsuite.com/dxlabwiki/SpotDatabasePrefilter>

+ With SpotCollector running, the VE7CC user application adds no value, but forces incoming spots to be processed twice, unnecessarily increasing CPU loading and memory consumption.
That's not really accurate - (AFAIK - it's been years since I've used
it) filters configured by the (VE7)CC User program are effected *on
the cluster node*, so the spots don't even traverse the network
(internet). This may make a significant difference to someone with
limited bandwidth, and would not consume CPU/memory on the client PC.
Those filters can be configured using commands/macros, but the CC User
tool provides a nice GUI.

* The CPU and memory consumed by running the CC User app significantly exceed the cost of conveying ~100 byte spots over the network every few seconds.
Curious how you quantified the cost of each, and what the actual results were?

I would have guessed that the network consumption would be more than
"~100 bytes every few seconds", especially if skimmer spots are
included.

I'm very lucky to have gigabit fiber internet to my home. I'm guessing
that some DXLab users have somewhat less than that, and some of those
users may have PCs with adequate capacity to run the CC User client (I
don't really it being especially bloated (but again, it's been years
since I used it).


* Is there something wrong with SpotCollector's pre-filter GUI? There's a screen shot in the article cited above.
I think it's more about where the filtering is effected (on the client
or on the server). My point about the CC User GUI is that it provides
a nice interface for managing *server-side* filtering

73,

~iain / N6ML


Dave AA6YQ
 

% more AA6YQ comments below

Curious how you quantified the cost of each, and what the actual results were?

I would have guessed that the network consumption would be more than
"~100 bytes every few seconds", especially if skimmer spots are included.

I'm very lucky to have gigabit fiber internet to my home. I'm guessing that some DXLab users have somewhat less than that, and some of those users may have PCs with adequate capacity to run the CC User client (I don't really it being especially bloated (but again, it's been years since I used it).

% A DX spot is 75 bytes. At this moment, telnet.reversebeacon.net is reporting 8600 spots per hour, which is 2.4 spots per second, which is 180 bytes per second. No one's internet connection will be strained by this. No one's internet connection would be strained by 10X that bandwidth.

% With CC user running, every incoming spot that survives the cluster filters must be received, processed, and retransmitted to SpotCollector. These tasks consume CPU and memory resources, but add no value. I don't know whether the added time delay on incoming spots is significant.

% At most, CC user should only be started when there's a desire to modify cluster filter settings; then it should be terminated so that SpotCollector can interact with its spot sources directly. But I doubt that the internet bandwidth savings achieved by "filtering at the cluster" will be noticeable.

73,

Dave, AA6YQ


iain macdonnell - N6ML
 

On Fri, Oct 16, 2020 at 4:45 PM Dave AA6YQ <@AA6YQ> wrote:

% more AA6YQ comments below

Curious how you quantified the cost of each, and what the actual results were?

I would have guessed that the network consumption would be more than
"~100 bytes every few seconds", especially if skimmer spots are included.

I'm very lucky to have gigabit fiber internet to my home. I'm guessing that some DXLab users have somewhat less than that, and some of those users may have PCs with adequate capacity to run the CC User client (I don't really it being especially bloated (but again, it's been years since I used it).

% A DX spot is 75 bytes. At this moment, telnet.reversebeacon.net is reporting 8600 spots per hour, which is 2.4 spots per second, which is 180 bytes per second. No one's internet connection will be strained by this. No one's internet connection would be strained by 10X that bandwidth.
Is that really true? Maybe it is, and I'm way behind the times. It
used to be that there were lots of remote areas where broadband
service is not available, so choices (still) included options like
dial-up and satellite. Maybe that's all changed. Interested to hear
from others on this.


% With CC user running, every incoming spot that survives the cluster filters must be received, processed, and retransmitted to SpotCollector. These tasks consume CPU and memory resources, but add no value. I don't know whether the added time delay on incoming spots is significant.
I would imagine that the delay would be measured in milliseconds. I
know that you're quick to jump on the spots, but I'd be surprised if
milliseconds makes a difference :)


% At most, CC user should only be started when there's a desire to modify cluster filter settings; then it should be terminated so that SpotCollector can interact with its spot sources directly. But I doubt that the internet bandwidth savings achieved by "filtering at the cluster" will be noticeable.
It's been too long since I used it to recall if there are any settings
that are only effective for the session in which they are selected.
Even if they are, if someone wants to make frequent changes, it's
convenient to have the client running all of the time.

I'm not saying that everyone should be using this model - I'm just not
convinced that there aren't cases where it does make sense.....

73,

~iain / N6ML


Jim N7US
 

Once you set the filters at the node using CC User, you can close it and thereafter connect directly to the node with SpotCollector.  I have several profiles set up at VE7CC and AE5E for different filters using different SSIDs (N7US, N7US-20, etc.) for different purposes.  N7US-20, for instance, gives me CW spots for the contest bands from K/VE, and I use that with N1MM+.

My standard profile, N7US, gives me spots from K/VE for 160-10M and from other stations in W9 for 6M, all excluding FT4/8.

In the original days of AK1A's DX PacketCluster (I was a sysop for many years), unfiltered spots were known as "The fire hose."

Jim N7US
Sent from Samsung tablet


-------- Original message --------
From: "iain macdonnell - N6ML via groups.io" <ar@...>
Date: 10/16/20 6:19 PM (GMT-06:00)
To: DXLab@groups.io
Subject: Re: [DXLab] VE7CC - Spot Source

On Fri, Oct 16, 2020 at 3:04 PM Dave AA6YQ <aa6yq@...> wrote:
>
> * More AA6YQ comments below
>
>
> It is en easy program to create filters which can be sent to the cluster. That limits bandwidth.
>
> + SpotCollector provides the ability to pre-filter cluster spots. See
>
> <https://www.dxlabsuite.com/dxlabwiki/SpotDatabasePrefilter>
>
> + With SpotCollector running, the VE7CC user application adds no value, but forces incoming spots to be processed twice, unnecessarily increasing CPU loading and memory consumption.
> That's not really accurate - (AFAIK - it's been years since I've used
> it) filters configured by the (VE7)CC User program are effected *on
> the cluster node*, so the spots don't even traverse the network
> (internet). This may make a significant difference to someone with
> limited bandwidth, and would not consume CPU/memory on the client PC.
> Those filters can be configured using commands/macros, but the CC User
> tool provides a nice GUI.
>
> * The CPU and memory consumed by running the CC User app significantly exceed the cost of conveying ~100 byte spots over the network every few seconds.

Curious how you quantified the cost of each, and what the actual results were?

I would have guessed that the network consumption would be more than
"~100 bytes every few seconds", especially if skimmer spots are
included.

I'm very lucky to have gigabit fiber internet to my home. I'm guessing
that some DXLab users have somewhat less than that, and some of those
users may have PCs with adequate capacity to run the CC User client (I
don't really it being especially bloated (but again, it's been years
since I used it).


> * Is there something wrong with SpotCollector's pre-filter GUI? There's a screen shot in the article cited above.

I think it's more about where the filtering is effected (on the client
or on the server). My point about the CC User GUI is that it provides
a nice interface for managing *server-side* filtering

73,

    ~iain / N6ML




Dave AA6YQ
 

$ more AA6YQ comments below


% A DX spot is 75 bytes. At this moment, telnet.reversebeacon.net is reporting 8600 spots per hour, which is 2.4 spots per second, which is 180 bytes per second. No one's internet connection will be strained by this. No one's internet connection would be strained by 10X that bandwidth.
Is that really true? Maybe it is, and I'm way behind the times. It used to be that there were lots of remote areas where broadband service is not available, so choices (still) included options like dial-up and satellite. Maybe that's all changed. Interested to hear from others on this.

$ 1800 bytes/sec would not strain a dialup connection. When SpotCollector was first released in 2001, many users employed dialup connections.

% With CC user running, every incoming spot that survives the cluster filters must be received, processed, and retransmitted to SpotCollector. These tasks consume CPU and memory resources, but add no value. I don't know whether the added time delay on incoming spots is significant.
I would imagine that the delay would be measured in milliseconds. I know that you're quick to jump on the spots, but I'd be surprised if milliseconds makes a difference :)

$ Without a network analyzer or access to source code, there's no way to know how much delay is being inserted.


% At most, CC user should only be started when there's a desire to modify cluster filter settings; then it should be terminated so that SpotCollector can interact with its spot sources directly. But I doubt that the internet bandwidth savings achieved by "filtering at the cluster" will be noticeable.
It's been too long since I used it to recall if there are any settings that are only effective for the session in which they are selected.
Even if they are, if someone wants to make frequent changes, it's convenient to have the client running all of the time.

I'm not saying that everyone should be using this model - I'm just not convinced that there aren't cases where it does make sense.....

$ The addition of "pre-filtering" to SpotCollector is relatively recent. I suspect that anyone still using CC User with SpotCollector is doing so because they aren't aware of SpotCollector's new capability, not because they've determined that CC User's CPU and memory consumption are justified by the small reduction in internet bandwidth.

73,

Dave, AA6YQ


Dave AA6YQ
 

+ AA6YQ comments below

Once you set the filters at the node using CC User, you can close it and thereafter connect directly to the node with SpotCollector. I have several profiles set up at VE7CC and AE5E for different filters using different SSIDs (N7US, N7US-20, etc.) for different purposes. N7US-20, for instance, gives me CW spots for the contest bands from K/VE, and I use that with N1MM+.

My standard profile, N7US, gives me spots from K/VE for 160-10M and from other stations in W9 for 6M, all excluding FT4/8.

+ Please try using SpotCollector's pre-filtering mechanism instead of CC User, and see if you notice any difference. On the "Spot Sources" tab of SpotCollector's Configuration window, click the "Pre-filter" button in the lower-right corner.

In the original days of AK1A's DX PacketCluster (I was a sysop for many years), unfiltered spots were known as "The fire hose."

+ What seemed like a fire hose then is a trickle now.

73,

Dave, AA6YQ


iain macdonnell - N6ML
 

On Fri, Oct 16, 2020 at 5:29 PM Dave AA6YQ <@AA6YQ> wrote:

$ more AA6YQ comments below


% A DX spot is 75 bytes. At this moment, telnet.reversebeacon.net is reporting 8600 spots per hour, which is 2.4 spots per second, which is 180 bytes per second. No one's internet connection will be strained by this. No one's internet connection would be strained by 10X that bandwidth.
Is that really true? Maybe it is, and I'm way behind the times. It used to be that there were lots of remote areas where broadband service is not available, so choices (still) included options like dial-up and satellite. Maybe that's all changed. Interested to hear from others on this.

$ 1800 bytes/sec would not strain a dialup connection. When SpotCollector was first released in 2001, many users employed dialup connections.
There was no CW Skimmer in 2001. I'm starting to see obvious
robo-feeds of FT8 (etc.) spots now too.

I suppose we'll have to agree to disagree on the potential impact to a
marginal internet connection in a remote location.


% With CC user running, every incoming spot that survives the cluster filters must be received, processed, and retransmitted to SpotCollector. These tasks consume CPU and memory resources, but add no value. I don't know whether the added time delay on incoming spots is significant.
I would imagine that the delay would be measured in milliseconds. I know that you're quick to jump on the spots, but I'd be surprised if milliseconds makes a difference :)

$ Without a network analyzer or access to source code, there's no way to know how much delay is being inserted.
Contesters make extensive use of the CC User model (inserted infront
of their context logging software) - or at least they did when I was
involved a while back. If anyone's more sensitive to spot latency than
DXers, it'd be contesters.


% At most, CC user should only be started when there's a desire to modify cluster filter settings; then it should be terminated so that SpotCollector can interact with its spot sources directly. But I doubt that the internet bandwidth savings achieved by "filtering at the cluster" will be noticeable.
It's been too long since I used it to recall if there are any settings that are only effective for the session in which they are selected.
Even if they are, if someone wants to make frequent changes, it's convenient to have the client running all of the time.

I'm not saying that everyone should be using this model - I'm just not convinced that there aren't cases where it does make sense.....

$ The addition of "pre-filtering" to SpotCollector is relatively recent. I suspect that anyone still using CC User with SpotCollector is doing so because they aren't aware of SpotCollector's new capability, not because they've determined that CC User's CPU and memory consumption are justified by the small reduction in internet bandwidth.
Perhaps. I don't have enough data to attempt a definitive determination on that.

73,

~iain / N6ML


Dave AA6YQ
 

+ AA6YQ comments below

Contesters make extensive use of the CC User model (inserted infront of their context logging software) - or at least they did when I was involved a while back. If anyone's more sensitive to spot latency than DXers, it'd be contesters.

+ Have they measured the added latency and decided it was acceptable, or do they just employ CC User because they have no alternative?

73,

Dave, AA6YQ