ive still had no reply to my email i sent them after giving them screen shots and logs![]()
ive still had no reply to my email i sent them after giving them screen shots and logs![]()
As of sometime this morning (San Jose time) the Astra/US plant is basically failed.
It appears that the upload/posting/propagation continued for awhile (that is still a bit under review) but the downloading/leeching has pretty much ground to a halt, or very intermittent at best, with minimal requested connections (or the number of those going up/down/up/down you get the picture).
It isn't the internet connection (or route) to the plant (I have the ability with vpn to come at them from various directions, as see through the trace routes).
I haven't re-traced the 'gaps' that were seen in the last couple of weeks, but I'll do that now and see what gives (if I can get any nzbcc connections that hold up, of course).
Over the day into the next (now) things slowly, and I mean S L O W L Y improved, where even now it's still bouncing off zero as far as pure throughput downloading.
That 'semi-dead-zone' that is between things posted (approx.) mid-October to around December 1 still exists. About a week ago it looked like it was beginning to improve a bit (completion rate up to around 95% or so which is still pretty bad) but it's down into the mid-80's or worse now.
But the base throughput is, as I said, still dragging quite a bit, at best now is running at about 70% of the usual rate, with dips down to zero every few seconds or so.
Guess I'll see what give in another 12+ hours or so.
Generally flipping between the EU and US servers fills in broken files. Although it keeps getting worse. A few years back I never had to use the EU server.
No matter WHICH plant is closest to one, you should always utilize the 'opposite' as your first fills, your downloader should take care of that automatically if you set it up right.
Then of course use a block provider (Blocknews etc.) as a third possibility. Actually, as far as Astra is concerned, folks in Europe seem to report more problems with the Amsterdam plant than the San Jose one. My checking shows that Amsterdam seems to loose about a percentage point or two from the US/SJ. Of course, your mileage may vary depending on what you're trying to get and where (perhaps) it was posted to (a US vr Euro plant).
Astraweb is a pile of turd they don't give shit anymore they refuse to acknowledge problems and fix correctly everything with them is band aid and problems keep coming back.
Binary group count is a joke now when comparing to other providers.
They refuse to add new groups claiming they don't exist but in realty they don't want to deal with more take-downs.
Even up-loaders are reverting to other providers because their servers are too unreliable everyone else should to the same.
Last edited by myself12345; 01-18-2015 at 02:54 PM.
I don't think there is anyone else who tracks Astraweb/US as close or as completely as I do, and while there is a bit of reality in your statement, the one area that I've NEVER seen any problems with their US plant is in posting/propagation. This in over at least 10+ years.
Now, there have been wobbles in the actual connections while posting, to be sure, just as there is in downloading/leeching. As far as adding 'new' groups I don't know. I don't think there are any 'new' groups I've tripped across in ages that had any decent amount of traffic. And as far as takedowns, I repeat what I've said many times here, that in those 10+ years of use (Giganews and Supernews were my major usenet connections for at least the 10 years previous to that, so we're talking about 20 years ago) that I've NEVER had a single posted file that has been removed by any provider, and in the last 8+ years I've transferred (posted) at least 20TB every year since 'day zero', which is when most plants stopped 'rolling off' any data, i.e., around Aug2008. And yes, I've saved every nzb generated.
I do 'retention studies' on Astra/Giga/Blocknews every month or so, going all the way back to those early days. I sometime spot 'weak' retention in the plants that are data sensitive. Right now, Astra/US is still having a problem with data in the Sept-December 2014 time frame, that doesn't seem to be going away, and was some of the reason why this particular thread was started. I along with several others have complained and it doesn't seem to be being fixed or even looked at.
Then again, a lot if not all of these data 'fades' would be minor to inconsequential if folks would simply follow a few basic rules on the included percentage of parity, which I outlined previously in a reply a couple pages back, as lots of folks seem to believe that usenet operates perfectly (both transmission and storage) and that providers that aren't 100% are, as you put it, dung.
Well, then, move to other providers at 3+ times the cost. Or get a good block provider like most folks, and be done with it.
Things may be improving for Astraweb. Over the last month I've been getting pretty much a 100% successful retrieval,including a 10Gb zip which needed no pars or fills.
Bookmarks