On Wed, 21 Jan 2009 17:28:47 +0000, Matthew Toseland wrote: > On Wednesday 21 January 2009 03:01, Dennis Nezic wrote: > > On Mon, 19 Jan 2009 23:52:26 +0100, bqz69 wrote: > > > > > > > > Now my freenet is running on my fit-pc - has been running > > > > > properly the last week. > > > > > > > > > > I did following (I am using ubuntu 8.04): > > > > > > > > > > 1. Created a text file /etc/cron.allow containing my username. > > > > > > > > That might explain why your wrapper's wrapper (your crontab-run > > > > script) wasn't working. Though you don't have to manually > > > > specify a cron.allow file... you can just delete it, and it > > > > allows everyone by default, unless they're mentioned in > > > > cron.deny. > > > > > > > > > 2. Inserted following line in /etc/crontab: > > > > > > > > > > @hourly myusername ~/Freenet/run.sh start > > > > > > > > > > (Probably not necessary) > > > > > > > > This one, the system-cron file, is necessary. The second one is > > > > useless, I believe. Cron never checks > > > > ~/.crontab--only /etc/crontab, and /var/spool/cron/crontabs, > > > > for individual users. > > > > > > > > > 3. Created a text file ~/.crontab with the following line: > > > > > > > > > > @hourly myusername ~/Freenet/run.sh start > > > > > > > > > > The freenet system seems however to very sensitive, and stops > > > > > when I do some other work, and then I have to restart freenet, > > > > > but as a mini-freenet server just serving data, it seems to > > > > > work well. > > > > > > > > Interesting. And bad! :). How sure are you that the crashes > > > > occur when you're doing other work on the system? > > > > > > Ok, not so sure, just tried again and freenet did not stop this > > > time. > > > > > > > (Is it wishful thinking? ;). > > > > What is the "nice" value for freenet's java process? > > > > > > It is 10 > > > > > > > (You can check > > > > it via the "top" command.) I had mine at a brutal 20 (the lowest > > > > priority of all my processes on my system), and toad suggested > > > > that this may have been the cause of my crashes. I have raised > > > > it's priority now, and will continue to test. Though I am > > > > skeptical. Crashing should not happen. Ever! > > > > > > > > > Thanks everybody so far, for your help > > > > > > > > "So far". I'm sure we haven't heard the end of this one :). > > > > Mine just "crashed" recently. Actually, it shuts itself down pretty > > cleanly, after outputting the following age-old messages: > > > > Restarting node: PacketSender froze for 3 minutes! > > Exiting on deadlock. > > Restarting node: MessageCore froze for 3 minutes! > > (USM deadlock) > > Goodbye. > > > > In general, I do notice that freenet bogs down my 1.2GHz machine > > quite a bit. Could that be the cause of these freezes/deadlocks? > > Can't it be less cpu/mem intensive? If I recall correctly, it does > > run smoothly for the first few hours, then slowly grinds itself > > (apparently) and my box to an unbearable crawl. > > > > Maybe we could make those messages more informative? For example, > > before having the node shut itself down, have it dump it's list of > > threads or queues or whatever. > > Give it more memory. If you can't give it more memory, throw the box > out the window and buy a new one. If you can't do that wait for the > db4o branch.
My main point in my last post was a suggestion to have the error message more informative. As another example, have it output it's memory/cpu usage before it shuts itself down, in the case of the deadlock I mentioned. Also, why is there such a high requirement?? Why on earth is 100MB memory not enough? If it can't allocate any more memory, it should wait or throttle itself. Restricting freenet to the latest unecessary super-computers is dumb. (It really should be developed on a 486.) When is the db4o stuff expected to be released? > Seriously, EVERY time I have investigated these sorts of issues the > answer has been either that it is showing constant Full GC's because > it has slightly too little memory, or that there is external CPU > load. Are you absolutely completely totally > 100000000000000000000000000% sure that that is not the problem? > AFAICS there are two posters here, and just because one of them is > sure that the problem isn't memory doesn't necessarily mean that the > other one's problems are not due to memory?? There are reports on FMS of people with gigs of ram, and powerful machines, with crashing nodes. Though, I'm not sure if it's the same problem, as my node hasn't really crashed this time--it just shut itself down. (Before I would get JVM hung errors, without any clean shut downs.)
