Blocknews corrupting

Within the last couple of days, Blocknews(US), which is used by many as a 'fills' server, has begun to corrupt it's file/database.

It appears that the further back the parts are, the more corrupted they are, to the point where things older than 3 months or so are virtually non-existent.

They were notified of this, but then their email server is blocking any attempt to receive nzb's (bouncing back and file attachment), so I guess the upshot is they don't care it's falling down flat on it's face.

As is usual with these usenet failures, Giganews (as far as I can tell) is running 100%.

Hm. I have had trouble retrieving articles a year old, thought it was just bad nzb. Hopefully this is not the case.

The eu server had no probs, was just half the speed from northam.

They finally actually took a deep look and found a 'problem', unknown as to what. It appears to have solved it.

I went back one year to an nzb that was checked/verified 100% at the time, and it checked out again.

Apparently these folks don't keep a very close watch on their systems, relying on automated alarms of some type. 'Back in the day' I did various checks on usenet plants and helped develop check/alarm programs that would have caught problems such as this. But that was 30 yrs ago, guess such things have not remained in the toolkit/memory of todays plant operators.

Instead of starting a new thread, decided to 'append' here with the latest news on Blocknews US, in that their database is starting to corrupt yet again.

It appears to be slight so far, and again seems to be confined to the US server, so anyone should use the EU server as an additional backup. A ticket has been opened, and hopefully as I found a major error in their US plant less than a month ago, that they'll 'hop to it' this time around.

ADD: It looks like I caught the plant right at the edge of a massive failure; some 6+ hours later, I'm getting massive numbers of 'not found' blocks; eu server as well as others (Giga) perfectly fine.

Blocknews jumped on this pretty fast, it all appears to now be completely fixed. Would of course be nice if it didn't happen at all, and they instituted some more on-going checks and such so they didn't have to rely on users to give them heads up on their own system(s).

I had noticed the older a file, the more chance it was corrupted. Good thing they worked to fix it. Haven't used them in a while though.