On Thursday 23 October 2003 21:36, Doug Bostrom wrote: > Greetings, > > Having responded to Matthew's request for a few people to set up a test > network of NGR servers, I'm now wondering how the one I've started can be > more useful. I've been using it to retrieve various random things, and > when I have a few moments from time to time I'll insert something into the > network. > > Running the server and exercising it a little is all fine,and dandy, but > what about the information from error messages and the log output that's > currently being dumped unwatched into a file on the host here? I'm split > heavily on a bunch of other work, and I'll likely miss anything interesting > that might be coughed up. So, an idea: Would it be completely stupid to > capture the log output periodically, grep it for stuff suggested as > interesting material by the developers, then gzip and mail the result to > some address where Matthew or whoever is responsible for all this can have > a look at it? Sure, there's quite a bit in the logs but by removing all the > unexceptional stuff maybe something would remain that would be useful if > only you could see it. This could all be done effortlessly by a script. In > fact, such a script could fashioned so that the items to search for could > be retrieved along w/ the daily snapshot jar retrieval. > > Whatever helps get things working better- if you want the entire log for > every day I'll be happy to send it. It just seems like a shame to be > stabbing around in the dark (so to speak, I'm sure you're not actually) > when there's all this info available if only you could see it. > > Given the nature of the NGR test network, it's hard to think that such a > scheme would be a dreadful security compromise or somehow expose the > network to danger. > > I mention all of this in part because a few hours ago while I was busy with > a conference call I noticed the console on the NGR box here spit out a > dire warning having to with a serious JVM error or somesuch. Unfortunately > I couldn't do a thing about it at the time and now I can't find it in the > log, though it's presumably in there someplace. > > Over and out.
Better yet, insert it onto the network. Just have a script where Frost or something inserts it into a channel. That would give you plenty of traffic, new stuff to request, and make the information available. Make a few categories: Errors, Normal, Discussion etc. I'll try to get a unstable node up tomorrow. _______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
