Skip to content

Commit

Permalink
ldms-notify enhancements needed for production convenience.
Browse files Browse the repository at this point in the history
While not always calculable, the duration of tracked pids is of great interest
and rather hard to compute downstream. This adds the duration field to
process end messages and a format option to suppress it for sites not ready
to migrate to the expanded schema.

Unlike ldmsd, the status of ldms-notify cannot be determined remotely,
so this also adds a heartbeat message, which by default is not generated.
  • Loading branch information
baallan authored and tom95858 committed Feb 18, 2024
1 parent 5c88137 commit 9dd76c4
Show file tree
Hide file tree
Showing 3 changed files with 275 additions and 71 deletions.
11 changes: 7 additions & 4 deletions ldms/scripts/examples/linux_proc_sampler
Original file line number Diff line number Diff line change
Expand Up @@ -67,12 +67,15 @@ cat << EOF > $LDMSD_RUN/metrics.input
}
EOF
rm -f $LOGDIR/json*.log
#valgrind -v --tool=drd --log-file=$LOGDIR/vg.netlink.txt ${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json.log --exclude-dir-path= --exclude-short-path= --exclude-programs --track-dir=${LDMSD_RUN}/ldms-netlink-tracked &
#valgrind -v --leak-check=full --track-origins=yes --trace-children=yes --log-file=$LOGDIR/vg.netlink.txt ${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json.log --exclude-dir-path= --exclude-short-path= --exclude-programs --track-dir=${LDMSD_RUN}/ldms-netlink-tracked &
${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json.log --exclude-dir-path= --exclude-short-path= --exclude-programs --track-dir=${LDMSD_RUN}/ldms-netlink-tracked -x -e exec,clone,exit &
drd="valgrind -v --tool=drd --log-file=$LOGDIR/vg.netlink.drd.txt --trace-cond=yes --trace-fork-join=yes"
memcheck="valgrind -v --leak-check=full --track-origins=yes --trace-children=yes --log-file=$LOGDIR/vg.netlink.memcheck.txt --keep-debuginfo=yes --malloc-fill=3b"
#${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json.log --exclude-dir-path= --exclude-short-path= --exclude-programs --track-dir=${LDMSD_RUN}/ldms-netlink-tracked &

${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json.log --exclude-dir-path= --exclude-short-path= --exclude-programs --track-dir=${LDMSD_RUN}/ldms-netlink-tracked -x -e exec,clone,exit -L $LOGDIR/nl.log --heartbeat 1 -v 0 &

# uncomment next one to test duplicate handling
#${BUILDDIR}/sbin/ldms-netlink-notifier --port=61061 --auth=none --reconnect=1 -D 30 -r -j $LOGDIR/json2.log --exclude-dir-path= --exclude-short-path= --exclude-programs &
VGARGS="--tool=drd --suppressions=/scratch1/baallan/ovis/ldms/scripts/examples/linux_proc_sampler.drd.supp"
VGARGS="--tool=drd --trace-cond=yes --trace-fork-join=yes"
VGARGS="--leak-check=full --track-origins=yes --trace-children=yes --show-leak-kinds=definite --time-stamp=yes --keep-debuginfo=yes --malloc-fill=3b"
#vgon
LDMSD 1
Expand Down
Loading

0 comments on commit 9dd76c4

Please sign in to comment.