This one time, at band camp, Paolo Lucente wrote: >Hey Wim, > >On Tue, Aug 23, 2005 at 12:02:23PM -0700, Wim Kerkhoff wrote: >> Here's an interesting one that I just noticed in dmesg on a Linux 2.6.10 >> traffic accounting machine. Not sure if it's caused by pmacctd, network >> conditions, or a bad kernel... but a bit scary anyhow :-) > >Googling a bit, seems like such message is caused by a memory 'shortage' >(ok, it's not a rocket: you already know this :-) ). Any chance this really >happened ? Do you notice anything strange in the size of the running pmacctd >processes ? I would just confirm that none of the functions (even the cache >related ones) are part of the daemon code. Let me know whether things evolve >someway.
I've seen these also, maybe one or two a day; the thing that strikes me as odd is that the machine is never short of memory. There's usually 50% swap usage and full ram, but nothing that I'd expect to cause pmacctd to run out of memory.
