> > Hi there. I'm using nntpcache-2.3.2 on a Linux (Redhat 5.0) machine. > I've turned nocem support on, and I'm seeing regular crashes after > several nocem messages are processed. When I first fired it up, it > read around 10K messages from the server before it started flailing. > It now looks like this in the logs: > > Feb 18 14:18:03 davinci nntpcache-nocem[24234]: nocem.c:93: <180298043942@CREDWBCWWI> already clasified as 'spam' > Feb 18 14:18:03 davinci nntpcache-nocem[24234]: nocem.c:93: <228419961473.7823455999@This.Site> already clasified as 'spam' > Feb 18 14:18:03 davinci nntpcache-nocem[24234]: nocem.c:93: <412292837380.2200314278@celebsxxx.com> already clasified as 'spam' > Feb 18 14:18:03 davinci nntpcache-nocem[24234]: nocem.c:93: <34B31937.2E11@xxx.muncies.not> already clasified as 'spam' > Feb 18 14:18:03 davinci nntpcache-nocem[24234]: nntpcache.c:207:Input/output error: SIGSEGV! > > It looks like one nocem message is read every time, since every time > the nocem thingy wakes up, when it crashes there is one more > article in var/nntpcache/news.server/news/lists/filters. > > Has anybody else seen this before? Any particular places I should > start poking at? > > Thanks. I haven't seen this before. My only suggestions in terms of a possible fix is to rm cache.map and nntpcache.stats (which may have become corrupted ... somehow) or try 2.3.2.1. If you know how to use gdb, grab the core file and issue a bt. Cheers, Julian.
