I'm using newshosting running Forte Agent 8 on Windows 10.

I'm trying to figure out the source of this problem when using Forte
Agent. I recently rebuilt windows 10 from scratch as I was having
some real issues with my PC. I copied over my old Forte Agent
preferences from my old installation to the new installation.
Everything seems to be working fine except for nto being able to
retrieve older text posts or download older binarie posts. Under Task
Manager>>Active Tasks it shows an orange triangle with three dots
underneath.

The odd thing is that if i create a nzb file and use Sabnzbd to
retrieve the contents I have no issues, so that makes me think that
the problem does not relate to newshosting but either some setting in
Windows or Agent.

Any thoughts?

Thanks
Jigger

On Tue, 02 May 2023 20:02:13 -0400, Jigger Pine wrote:

> I'm using newshosting running Forte Agent 8 on Windows 10.
>
> I'm trying to figure out the source of this problem when using Forte
> Agent. I recently rebuilt windows 10 from scratch as I was having
> some real issues with my PC. I copied over my old Forte Agent
> preferences from my old installation to the new installation.
> Everything seems to be working fine except for nto being able to
> retrieve older text posts or download older binarie posts. Under Task
> Manager>>Active Tasks it shows an orange triangle with three dots
> underneath.

An orange triangle with three dots underneath is a task queued.
Which means the task is waiting for something to complete before
it will start.

See: Help >> Index >> Icons Used in Agent >> Task Manager Window

> The odd thing is that if i create a nzb file and use Sabnzbd to
> retrieve the contents I have no issues, so that makes me think that
> the problem does not relate to newshosting but either some setting in
> Windows or Agent.
>
> Any thoughts?

1. I use APN which uses the same Usenet backbone as newshosting.
I have had problems retrieving new bodies after downloading
several older bodies on the same connection. I have not seen
it the other way around, but I suppose it could happen.

So my first suggestion would be, before downloading older bodies
first close all connections to the server (Action >> Go Offline)
and then let Agent open new connections to download those older
bodies.

2. Something simple to check.

With newshosting, it is more efficient if Agent requests message
bodies by Message-ID. So make sure the Agent server setting in
'Request message bodies by Message-ID" is turned ON for newshosting.

screen-shot: <http://img4.imagetitan.com/img.php?image=26_6b835ip5__agent-download-by-mid.png>

This only applies to some news servers but not to other news servers.

3. Find what the task is queued waiting for, in Agent.

(a) If there are other tasks running, maybe try again when there
are no other tasks running.

(b) If there are other tasks running, then the question is what
is this task queued waiting for?
A debug log could help find out.

3.1 With Agent closed, edit AGENT.INI and in the
[Debugging] section change VerboseLog=0 to
VerboseLog=1.

3.2 Open the debug log window ("Tools >> Task Log")
*before* you get older bodies. Leave the window open
while getting older bodies.

3.3 Use an Agent command to get those older bodies.

3.4 Watch the debug log window to see what is going on.
Agent's requests to the server are marked with a '<' prefix.
The server's responses to Agent are marked with a '>' prefix.

Unfortunately you cannot copy and paste from the debug log
window. But if you need it, there is a way to log the same
information to a file "agent.log".

--
Kind regards
Ralph

Leorna a hwæthwugu; ðeah ðe þine gesælða forlætan, ne forlætt þe no þin cræft.

Ralph

Thanks for your guidance.

I did have 'Request message bodies by Message-ID' turned on so that
was not issue.

I turned on verbose debug and is just says, as an example:
2013-05-03 2:27:13 PM [0 None: 0 0] Retrieving 1 body for
alt.binaries.sounds.lossless.indie

There are no tasks before it holding it up. And sometimes but not
always it is followed by these commands:
2013-05-03 2:27:45 PM [0 NNTP: 8 925] Connection closed by server
2013-05-03 2:27:45 PM [0 NNTP: 8 925] Delete Channel
2013-05-03 2:27:45 PM [0 NNTP: 8 900] Connection closed by server
2013-05-03 2:27:45 PM [0 NNTP: 8 900] Delete Channel
2013-05-03 2:27:45 PM [0 NNTP: 8 893] Connection closed by server
2013-05-03 2:27:45 PM [0 NNTP: 8 893] Delete Channel
2013-05-03 2:27:45 PM [0 NNTP: 8 938] Connection closed by server
2013-05-03 2:27:45 PM [0 NNTP: 8 938] Delete Channel
2013-05-03 2:27:45 PM [0 NNTP: 8 889] Connection closed by server
2013-05-03 2:27:45 PM [0 NNTP: 8 889] Delete Channel

Even after these commands, the task remains in the queue and is not
deleted. While this task is in the queue waiting I am able to
retrieve other bodies that are newer from the same group as well as
from other groups.

I'm guessing that the program is having difficulty sending the command
to the server or is sending incorrect data to the server which is why
it hangs or closes the connection. The only thing I changed on the
new install was the data folder location and turning on Megajoin.
Since then I have retrieved new headers, deleted headers I did not
want and purged and compressed the groups to save space and now it
seems that items prior to Nov 2019 cannot be retrieved in the groups I
had originally been using.

I thought maybe the message ID info within the data folders had become
corrupt so I ran a couple of more tests.

As an additional test, I added a new group, retrieved all headers and
had no problems retrieving bodies as far back as 2008. So I think
that narrows the problem to something in the original groups.

I reset everything to default and ran Repair Folders on one of the
groups but I was still unable to retrieve older posts.

Do you have any further insights based on this information?

If not, since the rebuild was only a couple of months ago, I'm going
to just copy over from the old build and override the agent data
folder in the new build. If that fails to solve the issue, I guess I
will need to rebuild the newsgroups I had from scratch.

Thanks

Jigger

On Wed, 03 May 2023 20:53:34 +1200, Ralph Fox <-rf-nz-@-.invalid>
wrote:

>On Tue, 02 May 2023 20:02:13 -0400, Jigger Pine wrote:
>
>> I'm using newshosting running Forte Agent 8 on Windows 10.
>>
>> I'm trying to figure out the source of this problem when using Forte
>> Agent. I recently rebuilt windows 10 from scratch as I was having
>> some real issues with my PC. I copied over my old Forte Agent
>> preferences from my old installation to the new installation.
>> Everything seems to be working fine except for nto being able to
>> retrieve older text posts or download older binarie posts. Under Task
>> Manager>>Active Tasks it shows an orange triangle with three dots
>> underneath.
>
>An orange triangle with three dots underneath is a task queued.
>Which means the task is waiting for something to complete before
>it will start.
>
>See: Help >> Index >> Icons Used in Agent >> Task Manager Window
>
>
>> The odd thing is that if i create a nzb file and use Sabnzbd to
>> retrieve the contents I have no issues, so that makes me think that
>> the problem does not relate to newshosting but either some setting in
>> Windows or Agent.
>>
>> Any thoughts?
>
>
> 1. I use APN which uses the same Usenet backbone as newshosting.
> I have had problems retrieving new bodies after downloading
> several older bodies on the same connection. I have not seen
> it the other way around, but I suppose it could happen.
>
> So my first suggestion would be, before downloading older bodies
> first close all connections to the server (Action >> Go Offline)
> and then let Agent open new connections to download those older
> bodies.
>
>
> 2. Something simple to check.
>
> With newshosting, it is more efficient if Agent requests message
> bodies by Message-ID. So make sure the Agent server setting in
> 'Request message bodies by Message-ID" is turned ON for newshosting.
>
> screen-shot: <http://img4.imagetitan.com/img.php?image=26_6b835ip5__agent-download-by-mid.png>
>
> This only applies to some news servers but not to other news servers.
>
>
> 3. Find what the task is queued waiting for, in Agent.
>
> (a) If there are other tasks running, maybe try again when there
> are no other tasks running.
>
> (b) If there are other tasks running, then the question is what
> is this task queued waiting for?
> A debug log could help find out.
>
> 3.1 With Agent closed, edit AGENT.INI and in the
> [Debugging] section change VerboseLog=0 to
> VerboseLog=1.
>
> 3.2 Open the debug log window ("Tools >> Task Log")
> *before* you get older bodies. Leave the window open
> while getting older bodies.
>
> 3.3 Use an Agent command to get those older bodies.
>
> 3.4 Watch the debug log window to see what is going on.
> Agent's requests to the server are marked with a '<' prefix.
> The server's responses to Agent are marked with a '>' prefix.
>
> Unfortunately you cannot copy and paste from the debug log
> window. But if you need it, there is a way to log the same
> information to a file "agent.log".

Further testing reveals the problem existed before the new build.
Original build data files suffer from the same problem.

Jigger

On Wed, 03 May 2023 15:29:54 -0400, Jigger Pine
<jigger.pine@bluesinthenight.com> wrote:

>Ralph
>
>Thanks for your guidance.
>
>I did have 'Request message bodies by Message-ID' turned on so that
>was not issue.
>
>I turned on verbose debug and is just says, as an example:
>2013-05-03 2:27:13 PM [0 None: 0 0] Retrieving 1 body for
>alt.binaries.sounds.lossless.indie
>
>There are no tasks before it holding it up. And sometimes but not
>always it is followed by these commands:
>2013-05-03 2:27:45 PM [0 NNTP: 8 925] Connection closed by server
>2013-05-03 2:27:45 PM [0 NNTP: 8 925] Delete Channel
>2013-05-03 2:27:45 PM [0 NNTP: 8 900] Connection closed by server
>2013-05-03 2:27:45 PM [0 NNTP: 8 900] Delete Channel
>2013-05-03 2:27:45 PM [0 NNTP: 8 893] Connection closed by server
>2013-05-03 2:27:45 PM [0 NNTP: 8 893] Delete Channel
>2013-05-03 2:27:45 PM [0 NNTP: 8 938] Connection closed by server
>2013-05-03 2:27:45 PM [0 NNTP: 8 938] Delete Channel
>2013-05-03 2:27:45 PM [0 NNTP: 8 889] Connection closed by server
>2013-05-03 2:27:45 PM [0 NNTP: 8 889] Delete Channel
>
>Even after these commands, the task remains in the queue and is not
>deleted. While this task is in the queue waiting I am able to
>retrieve other bodies that are newer from the same group as well as
>from other groups.
>
>I'm guessing that the program is having difficulty sending the command
>to the server or is sending incorrect data to the server which is why
>it hangs or closes the connection. The only thing I changed on the
>new install was the data folder location and turning on Megajoin.
>Since then I have retrieved new headers, deleted headers I did not
>want and purged and compressed the groups to save space and now it
>seems that items prior to Nov 2019 cannot be retrieved in the groups I
>had originally been using.
>
>I thought maybe the message ID info within the data folders had become
>corrupt so I ran a couple of more tests.
>
>As an additional test, I added a new group, retrieved all headers and
>had no problems retrieving bodies as far back as 2008. So I think
>that narrows the problem to something in the original groups.
>
>I reset everything to default and ran Repair Folders on one of the
>groups but I was still unable to retrieve older posts.
>
>Do you have any further insights based on this information?
>
>If not, since the rebuild was only a couple of months ago, I'm going
>to just copy over from the old build and override the agent data
>folder in the new build. If that fails to solve the issue, I guess I
>will need to rebuild the newsgroups I had from scratch.
>
>Thanks
>
>Jigger
>
>
>
>On Wed, 03 May 2023 20:53:34 +1200, Ralph Fox <-rf-nz-@-.invalid>
>wrote:
>
>>On Tue, 02 May 2023 20:02:13 -0400, Jigger Pine wrote:
>>
>>> I'm using newshosting running Forte Agent 8 on Windows 10.
>>>
>>> I'm trying to figure out the source of this problem when using Forte
>>> Agent. I recently rebuilt windows 10 from scratch as I was having
>>> some real issues with my PC. I copied over my old Forte Agent
>>> preferences from my old installation to the new installation.
>>> Everything seems to be working fine except for nto being able to
>>> retrieve older text posts or download older binarie posts. Under Task
>>> Manager>>Active Tasks it shows an orange triangle with three dots
>>> underneath.
>>
>>An orange triangle with three dots underneath is a task queued.
>>Which means the task is waiting for something to complete before
>>it will start.
>>
>>See: Help >> Index >> Icons Used in Agent >> Task Manager Window
>>
>>
>>> The odd thing is that if i create a nzb file and use Sabnzbd to
>>> retrieve the contents I have no issues, so that makes me think that
>>> the problem does not relate to newshosting but either some setting in
>>> Windows or Agent.
>>>
>>> Any thoughts?
>>
>>
>> 1. I use APN which uses the same Usenet backbone as newshosting.
>> I have had problems retrieving new bodies after downloading
>> several older bodies on the same connection. I have not seen
>> it the other way around, but I suppose it could happen.
>>
>> So my first suggestion would be, before downloading older bodies
>> first close all connections to the server (Action >> Go Offline)
>> and then let Agent open new connections to download those older
>> bodies.
>>
>>
>> 2. Something simple to check.
>>
>> With newshosting, it is more efficient if Agent requests message
>> bodies by Message-ID. So make sure the Agent server setting in
>> 'Request message bodies by Message-ID" is turned ON for newshosting.
>>
>> screen-shot: <http://img4.imagetitan.com/img.php?image=26_6b835ip5__agent-download-by-mid.png>
>>
>> This only applies to some news servers but not to other news servers.
>>
>>
>> 3. Find what the task is queued waiting for, in Agent.
>>
>> (a) If there are other tasks running, maybe try again when there
>> are no other tasks running.
>>
>> (b) If there are other tasks running, then the question is what
>> is this task queued waiting for?
>> A debug log could help find out.
>>
>> 3.1 With Agent closed, edit AGENT.INI and in the
>> [Debugging] section change VerboseLog=0 to
>> VerboseLog=1.
>>
>> 3.2 Open the debug log window ("Tools >> Task Log")
>> *before* you get older bodies. Leave the window open
>> while getting older bodies.
>>
>> 3.3 Use an Agent command to get those older bodies.
>>
>> 3.4 Watch the debug log window to see what is going on.
>> Agent's requests to the server are marked with a '<' prefix.
>> The server's responses to Agent are marked with a '>' prefix.
>>
>> Unfortunately you cannot copy and paste from the debug log
>> window. But if you need it, there is a way to log the same
>> information to a file "agent.log".

On Wed, 03 May 2023 15:29:54 -0400, Jigger Pine wrote:

> I turned on verbose debug and is just says, as an example:
> 2013-05-03 2:27:13 PM [0 None: 0 0] Retrieving 1 body for
> alt.binaries.sounds.lossless.indie
>
> There are no tasks before it holding it up. And sometimes but not
> always it is followed by these commands:
> 2013-05-03 2:27:45 PM [0 NNTP: 8 925] Connection closed by server
> 2013-05-03 2:27:45 PM [0 NNTP: 8 925] Delete Channel
> 2013-05-03 2:27:45 PM [0 NNTP: 8 900] Connection closed by server
> 2013-05-03 2:27:45 PM [0 NNTP: 8 900] Delete Channel
> 2013-05-03 2:27:45 PM [0 NNTP: 8 893] Connection closed by server
> 2013-05-03 2:27:45 PM [0 NNTP: 8 893] Delete Channel
> 2013-05-03 2:27:45 PM [0 NNTP: 8 938] Connection closed by server
> 2013-05-03 2:27:45 PM [0 NNTP: 8 938] Delete Channel
> 2013-05-03 2:27:45 PM [0 NNTP: 8 889] Connection closed by server
> 2013-05-03 2:27:45 PM [0 NNTP: 8 889] Delete Channel

This indicates that 5 *other* connections are being closed. Perhaps
because you had been doing other downloads and their connections have
reached the expiry time configured either at the news server or in
Agent's server setting "Disconnect from the server after...".

What would be more useful is to show what the attempt to download the
older body is doing. Start from no activity and get the debug log
from the very beginning where Agent logs on to the server. Include
Agent's considering what to do about downloading that older body after
it has logged on, and any commands Agent sends to the server.

> I'm guessing that the program is having difficulty sending the command
> to the server or is sending incorrect data to the server which is why
> it hangs or closes the connection.

The debug log would show whether Agent is sending incorrect data to the
server for retrieving older bodies. If you have the part where Agent
sends data to the server.

If Agent is having difficulty, perhaps Agent had too many connections open?
Try limiting the number of connections to 8 or less. Also try 4 or less.

> As an additional test, I added a new group, retrieved all headers and
> had no problems retrieving bodies as far back as 2008. So I think
> that narrows the problem to something in the original groups.
>
> I reset everything to default and ran Repair Folders on one of the
> groups but I was still unable to retrieve older posts.
>
> Do you have any further insights based on this information?

We need to see what Agent wants to send to the server to retrieve the
older body. That is where the debug log comes in.

--
Kind regards
Ralph Fox

‘Þe bet þe be, þe bet þe byse;’ quoþ Hendyng.

Closed out of Agent. Started a new session. Attempted to retrieve an
old post. The only thing showing in the Log is as follows:

2023-05-05 10:04:52 PM [1 None: 0 0] Retrieving 1 body for
alt.binaries.sounds.lossless.1990s

No other activity in the log and the same Orange triangle with 3 dots.

Here is a summary of how I finally was able to get older bodies
retrieving again. However, not sure if this would work with groups
that have excessive headers.

I thought maybe if I get new headers in the group that will fix the
issue. With original retrieval still stuck in the queue, I upated the
folder headers without issue with the log starting as follows and
ending with task deleted.
2023-05-05 10:10:42 PM [0 NNTP 8 3] Create channel total=0 busy=0
priority=0 avail=0 connecting=0 retry=0 requesting=1

I attempted a second body retrieval. Same result as the first
attempt. Thus, no resolution, a second task stuck in Task Manager
with a similar log entry.
2023-05-05 10:11:31 PM [9 None: 0 0] Retrieving 1 body for
alt.binaries.sounds.lossless.1990s

With two Tasks in queue with the same Orange triangle with 3 dots, I
next attempted Get All Headers in Folder. This task completed
successfully with the log starting as follows and ending with task
deleted.
2023-05-05 10:24:58 PM [17 NNTP 8 3] Create channel total=0 busy=0
priority=0 avail=0 connecting=0 retry=0 requesting=1

Attempted a third retrieval. This time success. Stopped the two other
retrieval tasks and retried them. Again success.

So in my situation the problem was resolved by updated all headers.

Ralph, thanks for your guidance, assistance and patience.

J

On Thu, 04 May 2023 17:44:58 +1200, Ralph Fox <-rf-nz-@-.invalid>
wrote:

>On Wed, 03 May 2023 15:29:54 -0400, Jigger Pine wrote:
>
>> I turned on verbose debug and is just says, as an example:
>> 2013-05-03 2:27:13 PM [0 None: 0 0] Retrieving 1 body for
>> alt.binaries.sounds.lossless.indie
>>
>> There are no tasks before it holding it up. And sometimes but not
>> always it is followed by these commands:
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 925] Connection closed by server
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 925] Delete Channel
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 900] Connection closed by server
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 900] Delete Channel
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 893] Connection closed by server
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 893] Delete Channel
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 938] Connection closed by server
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 938] Delete Channel
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 889] Connection closed by server
>> 2013-05-03 2:27:45 PM [0 NNTP: 8 889] Delete Channel
>
>
>This indicates that 5 *other* connections are being closed. Perhaps
>because you had been doing other downloads and their connections have
>reached the expiry time configured either at the news server or in
>Agent's server setting "Disconnect from the server after...".
>
>What would be more useful is to show what the attempt to download the
>older body is doing. Start from no activity and get the debug log
>from the very beginning where Agent logs on to the server. Include
>Agent's considering what to do about downloading that older body after
>it has logged on, and any commands Agent sends to the server.
>
>
>> I'm guessing that the program is having difficulty sending the command
>> to the server or is sending incorrect data to the server which is why
>> it hangs or closes the connection.
>
>The debug log would show whether Agent is sending incorrect data to the
>server for retrieving older bodies. If you have the part where Agent
>sends data to the server.
>
>If Agent is having difficulty, perhaps Agent had too many connections open?
>Try limiting the number of connections to 8 or less. Also try 4 or less.
>
>
>> As an additional test, I added a new group, retrieved all headers and
>> had no problems retrieving bodies as far back as 2008. So I think
>> that narrows the problem to something in the original groups.
>>
>> I reset everything to default and ran Repair Folders on one of the
>> groups but I was still unable to retrieve older posts.
>>
>> Do you have any further insights based on this information?
>
>We need to see what Agent wants to send to the server to retrieve the
>older body. That is where the debug log comes in.