UK based. Perl, XML/HTTP, SAP, Debian hacker.
Sometime ago we had a strange problem on a Red Hat box where running top would result in a segfault in top and you get a coredump. Yesterday we had it again, and this time with a little help from Google I found out why. There is a bug in the procps package in Red Hat Linux 7.x, where on a dual CPU system after 250 days top will segfault. You can run top under a strace and it's okay, but otherwise it always dies with a division by zero error.
A quick hunt round Red Hat's site finds a later version of the package for Red Hat Linux 8.0, but I'm not sure how compatible Red Hat 7 and 8 are...
build your own (Score:2)
Re:build your own (Score:2)
Hadn't thought of that. Seems to work okay with the newly built version...
-- "It's not magic, it's work..."