]> git-server-git.apps.pok.os.sepia.ceph.com Git - ceph.git/commit
mgr/devicehealth: fix telemetry stops sending device reports after 48 hours 33346/head
authorYaarit Hatuka <yaarit@redhat.com>
Mon, 27 Jan 2020 13:57:55 +0000 (08:57 -0500)
committerYaarit Hatuka <yaarit@redhat.com>
Sat, 15 Feb 2020 01:29:14 +0000 (20:29 -0500)
commit21d78c0b87ffabf13e1db91e8321859d6195444d
tree26858f0858c841bd1289a972cef590cfba77cd7e
parent70fa3cf2341ec018076fe41fb23a930ff18f2560
mgr/devicehealth: fix telemetry stops sending device reports after 48 hours

Telemetry module fetches device metrics which were scraped in the last
"telemetry interval"*2 (=48 hours by default) by calling
_get_device_metrics() with min_sample. _get_device_metrics() fetches the
metrics from omap and breaks on the first one that is older than
min_sample. But because it fetched in ascending order (from oldest to
newest) it was breaking on the first one it received, if it was older
than the interval above. We need to pass min_sample to get_omap_vals()
so it will start fetching from that value.

Fixes: https://tracker.ceph.com/issues/43837
Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>
(cherry picked from commit 5f7e4a980a73e8cacb2c9bde47d822a32fb8c440)
src/pybind/mgr/devicehealth/module.py