Author Topic: We need to make major updates  (Read 1535 times)

Offline david_ritterhous

  • Prolific Indexer
  • ****
  • Posts: 240
  • Helpful: +6/-0
We need to make major updates
« on: 2017-07-18, 04:46:56 am »
I mentioned this about 8 months ago regarding the messageID issues.  Now it is in full swing.
https://www.reddit.com/r/usenet/comments/6nldvn/astrawebsupernewsnewsdemon_all_downloads_failing/

A quote from Giganews regarding the issues:
Code: [Select]
Looks like you hit the nail on the head. This is the message I got from Giganews support:
"Recently we have seen reports from customers about posts failing to complete when downloaded from certain groups. These groups include, but may not be limited to a.b.teevee and a.b.etc. The general consensus based on the complaints is that this has affected some posts in those groups since 15 July 2017. Newsreaders may report "430: No such article" or "Article not found." Upon investigation into these complaints, we have seen a few posters to these groups that are using invalid message IDs when posting. This causes the content to fail to download from our servers and is not something we can correct. We have ensured the issue is not with our servers, but with the formatting of the message IDs from the posters.
On a more technical note, we follow 1036 (https://www.ietf.org/rfc/rfc1036.txt). This is a standard for newsgroup message IDs that we have always followed. Anything that does not conform with this standard will show as incomplete and fail to be retrieved from our servers.
If you are experiencing an issue outside of the groups mentioned above and have some examples of message IDs or NZB files you would like us to scan and check on our servers to see if they are affected - feel free to respond and attach them so we can investigate."

Any plans to update nzedb to handle these kinds of posts?

Offline speckled

  • Newbie
  • *
  • Posts: 1
  • Helpful: +0/-0
Re: We need to make major updates
« Reply #1 on: 2017-07-18, 08:08:53 am »
Do these articles actually exist on their servers (can still be retrieved by article number) or are they being discarded entirely? Either way I'm not sure what nZEDb can do about it, because the nzb format itself is Message-ID based.