From df932c6a2038245e7a9684f62e586360d92cdc95 Mon Sep 17 00:00:00 2001 From: Patrick Seidensal Date: Thu, 26 Mar 2020 22:49:57 +0100 Subject: [PATCH] monitoring: alert for prediction of disk and pool fill up broken Fixes: https://tracker.ceph.com/issues/44776 Signed-off-by: Patrick Seidensal (cherry picked from commit 6935dc55921ef533e939719bad7dc12324450bbb) Conflicts: monitoring/prometheus/alerts/ceph_default_alerts.yml --- .../prometheus/alerts/ceph_default_alerts.yml | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/monitoring/prometheus/alerts/ceph_default_alerts.yml b/monitoring/prometheus/alerts/ceph_default_alerts.yml index ba7764d32e9c6..4c37d7f9c9b0a 100644 --- a/monitoring/prometheus/alerts/ceph_default_alerts.yml +++ b/monitoring/prometheus/alerts/ceph_default_alerts.yml @@ -186,15 +186,10 @@ groups: Node {{ $labels.instance }} experiences packet errors > 1 packet/s on interface {{ $labels.device }}. - # predict fs fill-up times - - alert: storage filling + - alert: storage filling up expr: | - ( - ( - node_filesystem_free_bytes / deriv(node_filesystem_free_bytes[2d]) - * on(instance) group_left(nodename) node_uname_info - ) <= 5 - ) > 0 + predict_linear(node_filesystem_free_bytes[2d], 3600 * 24 * 5) * + on(instance) group_left(nodename) node_uname_info < 0 labels: severity: warning type: ceph_default @@ -219,10 +214,9 @@ groups: - alert: pool filling up expr: | ( - ( - (ceph_pool_max_avail - ceph_pool_stored) / deriv(ceph_pool_max_avail[2d]) - ) * on(pool_id) group_right ceph_pool_metadata <= 5 - ) > 0 + predict_linear(ceph_pool_stored[2d], 3600 * 24 * 5) >= + ceph_pool_max_avail + ) * on(pool_id) group_right(name) ceph_pool_metadata labels: severity: warning type: ceph_default -- 2.39.5