Hide Forgot
Description of problem: numa_faults.stp example (which is run as part of the testsuite) hangs the box when run. Version-Release number of selected component (if applicable): systemtap-1.3-4.el4 How reproducible: always Steps to Reproduce: 1. stap -v -c 'sleep 0.2' /usr/share/systemtap/testsuite/systemtap.examples/memory/numa_faults.stp Actual results: the box stops responding immediatelly after 'Pass 5: starting run.' and I cannot ping or ssh it again: # stap -v -c 'sleep 0.2' /usr/share/systemtap/testsuite/systemtap.examples/memory/numa_faults.stp Pass 1: parsed user script and 72 library script(s) using 50656virt/20304res/2240shr kb, in 210usr/20sys/351real ms. Pass 2: analyzed script: 3 probe(s), 6 function(s), 2 embed(s), 4 global(s) using 108448virt/37488res/3164shr kb, in 150usr/350sys/6515real ms. Pass 3: translated to C into "/tmp/stap5kUO48/stap_90ec3fc1e332cc51011297f58409f44c_3422.c" using 107260virt/39820res/5428shr kb, in 410usr/90sys/550real ms. Pass 4: compiled C into "stap_90ec3fc1e332cc51011297f58409f44c_3422.ko" in 1170usr/120sys/2925real ms. Pass 5: starting run. Starting pagefault counters Expected results: no badness Additional info: this appeared before the death of the system in /var/log/messages: Jan 13 11:19:38 hp-bl260cg5-01 kernel: Unable to handle kernel paging request at 00000000000018f0 RIP: Old 4.8 systemtap does not even compile this systemtap script, so it is not a real regression, but if it could be avoided, it would be better
This could be a problem with some tapset embedded-C, or perhaps some basic ia64 kprobes problem. On account of systemtap still being a tech-preview on rhel4/ia64, I propose to WONTFIX this for now.
(same applies for x86-64; kprobes are not super robust on rhel4).
Development Management has reviewed and declined this request. You may appeal this decision by reopening this request.