On Sun, 23 Mar 1997 sthaug@nethelp.no wrote: > > And the performance I'm talking is not just that it doesn't have the > > information available and have to go and get it, that wouldn't explain why > > either 'list active' or 'list active <newsgroup>' takes 20-100 times as > > long as the new INN 1.5.1 server on the same machine, especially not if > > you did exactly the same operation 1 minute ago!! > > ... but this one I have a problem understanding. Either of these commands > are usually instantaneous against my nntpcache servers. If it's been too Yes, that what I expected too... I don't have a nntpcache to test it against, but I just used tcpdump to watch the information-flow between trn4 and our INN server. Here's an abbreviated transcript of the *per group* conversation, ignoring the beginning and ending parts: <- list active newsgroup-name -> 215: list: -> newsgroup-name <high> <low> <flag> This is repeated for each newsgroup in the screen. I suspect it tries to avoid a full 'list active' which can be a long operation on a loaded server. I can't be sure that it's this that is the bottleneck, not without a full trace against a nntpcache server with timestamps on the entries, but I suspect it *strongly*. After all, you could *see* it walk through the newsgroups... Note that 3 seconds for 50 lines means that it took 60 milliseconds per query. This is much to small time to be noticible by a mere human, so it would *appear* to be near instantaneous if one tried to look at it without knowing where to look.... There's no way a fairly well-equipped Pentium-Pro 200 should need that kind of time to answer to that query, and INN certainly hasn't got any problems with this. -- Torbjörn Lindgren Karenlyst Allè 5, N-0277 Oslo, Norway Phone: +47 22420102 E-mail: tl@funcom.com If Santa ever DID deliver presents on Christmas Eve, he's dead now.
