650B15-3007 posting errors

[Ribble]

650B15-3007 Win7 AMD64

Is there a way to increase retries when posting or to cause Newsbin to try posting the missed files again? Headsers get somewhat confusing if I simply repost, for example, all of "File03.mkv.008" or all of "File03.mkv.009" .

Examples of slime trail after posting complete:

D:\NewsbinData\Posting\File03.mkv.008.post.2
D:\NewsbinData\Posting\File03.mkv.008.post.23
D:\NewsbinData\Posting\File03.mkv.008.post.25
D:\NewsbinData\Posting\File03.mkv.008.post.3
D:\NewsbinData\Posting\File03.mkv.009.post.1
D:\NewsbinData\Posting\File03.mkv.009.post.6
D:\NewsbinData\Posting\File03.mkv.009.post.8

---
While I am on this subject...

Can you add a leading "0" to make
D:\NewsbinData\Posting\File03.mkv.009.post.1
into
D:\NewsbinData\Posting\File03.mkv.009.post.01
so it sorts better

[Quade]

Not easily on the leading zero. Not impossible but it would require new code.

What kind of errors are causing this? I'd thinking posting again with the same subject would fill in the chunks.

[Ribble]

Problem isn't the errors (chunks that did not post), but that Newsbin knows it did not post the chunks. So, looking for a way to make Newsbin post what it did not post.

Examples:
File0123.mkv.001 posted maybe 1 out of 28 chunks
File0123.mkv.002 posted maybe 3 out of 28 chunks
File0123.mkv.003 posted maybe 1 out of 28 chunks
The rest of File0123.mkv. 004 through 020 and all par2's posted OK

The downloaded headers comment field shows:
[001/200] for the original posts
[1/3] for repost of the same files which makes the headers different.

Why does the comment field even show {xxx/yyy}. It adds nothing that I can see and no "xxx" is ever greater than "1" or "001" anyway. "yyy" always shows the number of posts that were queued and not the number of posts in each file. So, the "yyy" also causes different headers on reposts.

The reposted files should work for Quickpar, though I have not tried to test it.

Since right click properties doesn't work, it is hard to see what is actually in the files posted.

As for what the errors actually are, I would have to know what to look for in the logging file. Probably transmission errors.

Again, since Newsbin already knows what chunks it didn't post, is there a way to increase the number of posting retries or a way to make Newsbin retry the failed posts.

The leading "0" thing just adds to the confusion. My opinion is that computers should always generate leading "0" when the computer converts from hexadecimal to decimal. Or, don't convert and name it using the hex instead of convoluted decimal.

[Ribble]

Oh hell, I went this far, and it all looks like bitching anyway.

Binsearch shows all the files that I posted.

Newsbin does not show all the files that I posted.

Someone mentioned elsewhere that Newsbin is not showing all the headers. Confirmed. If Binsearch sees the headers posted on Astraweb, then Newsbin should also see them as the headers are definitely there on Astraweb.
---

Will watch for "BogusMaxspool" errors in logging list. Closing Newsbin and restarting again now shows the files (headers) that I posted.

[Quade]

Oh hell, I went this far, and it all looks like bitching anyway.

It looks like lack of focus to me. I'd like to help you but, you can't stay on task long enough to for me figure out what you're doing. There's nothing rocket sciency about this process. If we can stay focused, what's going on should be obvious.

1 - You post and watch the logging tab for errors.

2 - If you see any errors you can tell me what they are then maybe I can make code changes.

3 - Then you download headers and see if you can see your files.

4 - If you want search to see the files, it might be 15 minutes before they show up, assuming you're using a group we catalog. If you're using an off the wall group we don't catalog, you tell us and we can add it. We add groups whenever people ask us to.

PM me an example of something you've posted and what group you posted it to. I'll look at it.

Try this

1 - hit the pause button

2 - Select all the downloads, pause them all using the pause button on the download list.

3 - Unpause Newsbin - downloads should stay paused

4 - Post a couple files.

5 - Watch the logging tab.

To trouble-shoot, you need to do one thing at a time.

[Ribble]

Thanks Quade.

Starting over.

For posting, Is there a way to make Newsbin try harder to post what I am posting?

To keep it simple, the answer may be "no".

[Quade]

You have no control of it. Then again, I've never seen a posting fail so, there hasn't been a need for the user to control it.

[Ribble]

Set my upload server to one connection. A little slower but no errors now. My guess, about 3% slower.

Had been set to 7 connections.

[Quade]

If you determine what the errors were from the logs, I can look at perhaps improving it. I need to know why it's failing before I can fix it.

[Ribble]

Had one missed chunk after I went to bed. It was around this time, but I have trouble seeing my own posts to look for what isn't there. These are the first few lines of the log - I think that I cleared the log before going to bed so I would expect a lot more would be logged before this. In other words, This likely isn't the errors.

I do have a Wordpad copy of the log, if I knew what to look for.


[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] ERROR Error: WSASend Socket Error Code: 10038 An operation was attempted on something that is not a socket.

[02:22:32] HIGH ssl.astraweb.com None |340 340 <523fddfb$0$53771$c3e8da3$14a0410e@news.astraweb.com> (desired)
[02:22:33] HIGH ssl.astraweb.com |340 340 <523fde19$0$49508$c3e8da3$fdf4f6af@news.astraweb.com> (desired)
[02:22:47] HIGH ssl.astraweb.com None |240 <523fde19$0$49508$c3e8da3$fdf4f6af@news.astraweb.com> Article posted
[02:22:47] HIGH ssl.astraweb.com |340 340 <523fde27$0$49508$c3e8da3$fdf4f6af@news.astraweb.com> (desired)
--------

EDIT:
Checked download headers and the error happened between 02:22 and 02:31. Aside of the above errors, there is nothing abnormal in the log until 03:18 when Newsbin began a download that was queued to start after the upload was near completion.

So much has changed since I started this thread and the one error out of perhaps 400 chunks posted really doesn't mean too much anymore. My original question was essentially "why doesn't Newsbin retry a chunk when it knows that chunk wasn't posted by looking at the slime trail (leftovers) in "D:\NewsbinData\Posting\".

I tried to point out that because of the [xxx/yyy] comment field, manually reposting a chunk is difficult.