Hi J., I glanced through the FAQ, but I didn't find a reference to the problem I reported earlier (never ending expires on a system upgraded from 2.3.3 to 2.4.0b2). Yesterday, I upgraded two servers from NNTPcache 2.3.3 to 2.4.0b2 for Y2K reasons, retaining the cache history and spool area. Basically, all I did after installing was update nntpcache.config, leaving the rest as it was. Later that day, I noticed that the spool (full before) had been emptied out, and that expire kept on running and restarting, not allowing the spool to be filled any more. I tried some different settings for minblocksfree and mininodesfree, but that didn't make a difference. Finally, I totally wiped out /usr/local/var/nntpcache, and rebooted the system. All went well from that moment on. The situation was identical for two separate systems, both running FreeBSD 3.3-STABLE. The most interesting part was that the expire stats were almost identical for both systems, whereby the reported percentage of usedblocks was nowhere near the figure reported by df -ki ... not to mention the freaky value of maxArtAge .. newsreader2 nntpcache-expire[6794]: expire.c:212: usedblocks = 8.5%, usedinodes = 0.1%, maxArtAge = 57s, expire ended newsreader1 nntpcache-expire[10872]: expire.c:212: usedblocks = 8.4%, usedinodes = 0.1%, maxArtAge = 57s, expire ended P.S.: maybe I'm a bit too early waving the green flag, because the spool has hardly had the chance to get filled, but I think the situation described above is not going to be repeated. If so, I'll get back to you. In the meantime: let me know if this rings a bell in any way .. -- - Ben C. O. Grimm ----------------- Ben.Grimm@wirehub.net - - Wirehub! Internet Engineering - http://www.wirehub.net/ - - Wirehub! Backbone --- http://doema.wirehub.net/wirehub/ - - Private Ponderings ------- http://libertas.wirehub.net/ -
