FIXED - Server priority not working in 6.52B2

[CoolStream]

The server I prefer to use is set @ 1 in the server priority settings. However it is the last one in the list. At the top of the list is a server I would prefer to use only if files can't be found on the preferred server which has only 7-day retention. I have given it a value of 8.

This has always worked in other versions, so am wondering if there is a bug. (settings still advise that the higher the number, the less likely the server will be used)

[Quade]

Server order in the list doesn't matter. The only thing that matters is the priority. "Typical Retention" will influence this too. In that a post older than "Typical Retention" for a server will be immediately failed and fall over to the next server. 430 errors on the first server are enough to trigger the second server.

What makes you think it's not working? I actually was looking at this yesterday and it seemed to work fine.

[CoolStream]

OK, to recap. Server A has limited retention but is free to use. I give it top priority (1) and set the Typical Retention in the 'Really Advanced' section to 7 days. Server B has years of retention but costs to use. I therefore give it a higher priority value (so that I only use it if the posts no longer exist on Server A). I have been using these settings for years without problem.

What I am noticing now is that if I try to download something older than 7 days, instead of defaulting to server B, Newsbin continues to attempt the download from server A (and logging just fills up with 'message not on this server' messages).

If I manually disable server A, downloading from Server B starts immediately and download speed jumps from kbs to MBs. It is the need to manually intervene that makes me think there is a problem or that I have managed to mess up my settings at some point.

[Quade]

Do you see it trying to download from server A?

You might want to give 6.52B3 a try.

[CoolStream]

OK, I'll give it a try and report back

[CoolStream]

Apologies for the delay.

I have just tested B3 with a post made over 1000 days ago (so no way that the 7-day retention server should think it had the cojones to attempt it).

RESULT:

- with both servers enabled, Server A appears to take precedence resulting in incoming traffic made up of 'nothing found' messages. Incoming data rate low k's

- with server A disabled, server B kicks in and files downloading the files immediately. Incoming data rate MAXIMUM.



I am wondering if the change to using a new 'Really Advanced' page in the 'Server Options' has broken something

[CoolStream]

Server order in the list doesn't matter. The only thing that matters is the priority. "Typical Retention" will influence this too. In that a post older than "Typical Retention" for a server will be immediately failed and fall over to the next server. 430 errors on the first server are enough to trigger the second server.

What makes you think it's not working? I actually was looking at this yesterday and it seemed to work fine.

I appreciate that this is a bit of an old thread, however the problem still exists. BIG OUCH! I ran out of credit on my backup server while downloading a heap of stuff that was less than 24 hours old because of this. All of the material existed on my main server (which typically has a 7-day retention). This arrangement has been ideal for me over the years and has only manifested itself in v6.

I'm bumping this thread to see if I am the only one who has found this issue.

If you need, Quade, I could made a video for you (I'm still using B3 as per your suggestion)

[Quade]

Are you using an older version? If yes, I'd upgrade before anything else. RC1 is probably going to be posted pretty soon.

[CoolStream]

Thanks for the quick response. I'm using 6.52B3 as per your suggestion.

Fingers crossed for RC1, although I doubt it will make any difference if you weren't able to replicate the problem.

[Quade]

Pretty sure the latest is B6

viewtopic.php?f=44&t=36026

[CoolStream]

Sorry for the delay in giving feedback.

RC1 appears to have fixed the problem. Well done and many thanks, guys

[Ribble]

Server "XS" is last one on my list. NB 6.52 3289 tries to connect to "news.astraweb.com" instead of "free.xsusenet.com"

[00:33:49] HIGH Open Post - AstrawebUS ARTICLE <1233501439.45932.33@news.astraweb.com> | 220 0 <1233501439.45932.33@news.astraweb.com> article
[00:33:52] HIGH Open Post - XS ARTICLE <1233501441.93168.4@news.astraweb.com> | 430 No such article

Wanted to post this before I lost the information. Now will look to see if a new version fixed this.
Thanks
---

Looks Like 3289 is latest.

Made a new "Last server" and errors started again. In fact, download speed dropped to zero.

[00:46:21] HIGH Open Post - XS ARTICLE <1233501449.62233.23@news.astraweb.com> | 430 No such article
[00:46:21] HIGH Failed to retrieve Article from Server:XS ARTICLE <1233501449.62233.23@news.astraweb.com> | 430 No such article
[00:46:22] HIGH Open Post - XS ARTICLE <1233501449.67489.25@news.astraweb.com> | 430 No such article
[00:46:22] HIGH Failed to retrieve Article from Server:XS ARTICLE <1233501449.67489.25@news.astraweb.com> | 430 No such article
[00:47:22] ERROR Error: WSARecv Socket Error Code: 10060 A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.

[00:47:22] HIGH Open Post - XS ARTICLE <1233501449.70026.26@news.astraweb.com> |
[00:47:23] HIGH Open Post - AstrawebUS ARTICLE <1233501449.70026.26@news.astraweb.com> | 220 0 <1233501449.70026.26@news.astraweb.com> article

Maybe I am reading these errors wrong.

Thanks.

[itimpi]

Server "XS" is last one on my list. NB 6.52 3289 tries to connect to "news.astraweb.com" instead of "free.xsusenet.com"

[00:33:49] HIGH Open Post - AstrawebUS ARTICLE <1233501439.45932.33@news.astraweb.com> | 220 0 <1233501439.45932.33@news.astraweb.com> article
[00:33:52] HIGH Open Post - XS ARTICLE <1233501441.93168.4@news.astraweb.com> | 430 No such article

Wanted to post this before I lost the information. Now will look to see if a new version fixed this.
Thanks

Those messages indicate that the string news.atraweb.com is part of the message-Id field (possibly because that is where the item was originally uploaded) - not that Newsbin is using that server.

[Ribble]

Thanks, they were posted to astraweb...

"NNTP-Posting-Host: b3744f08.news.astraweb.com"

So I was reading the error message incorrectly.

[Quade]

Yeah, this happens all the time. People see Astra and they think Newsbin is talking to Astra.