[PATCH] Remove logging every slow request details to monitors

 recent change https://tracker.ceph.com/issues/43975 was made to slow
 request logging to include detail on each operation in the cluster logs.
 With this change, detail for every slow request is always sent to the
 monitors and added to the cluster logs.

 This does not scale. Large, high-throughput clusters can overwhelm
 their monitors with spurious logs in the event of a performance issue.
 Disrupting the monitors can then cause further instability in the
 cluster.

 This SRU reverts the cluster logging of every slow request the osd is
 processing.

 The slow request clog change was added in nautilus (14.2.10) and
 octopus (15.2.0).

Signed-off-by: Gerald Yang <gerald.yang@canonical.com>
Gbp-Pq: Name disable-log-slow-requests.patch

diff --git a/src/osd/OSD.cc b/src/osd/OSD.cc
index 8dceddae384c4614adeccdb83057f7ce3c9070f6..373d12d6e6fd9bb39fd2ba18a9e363d343d21753 100644 (file)
--- a/src/osd/OSD.cc
+++ b/src/osd/OSD.cc
@@ -7780,7 +7780,6 @@ vector<DaemonHealthMetric> OSD::get_health_metrics()
            << " currently "
            << op.state_string();
         lgeneric_subdout(cct,osd,20) << ss.str() << dendl;
-        clog->warn() << ss.str();
        slow++;
        if (!oldest_op || op.get_initiated() < oldest_op->get_initiated()) {
          oldest_op = &op;