osd: Set default max active recovery and backfill limits for mClock

Client ops are sensitive to the recovery load and must be carefully
set for osds whose underlying device is HDD. Tests revealed that
recoveries with osd_max_backfills = 10 and osd_recovery_max_active_hdd = 5
were still aggressive and overwhelmed client ops. The built-in defaults
for mClock are now set to:

    1) osd_recovery_max_active_hdd = 3
    2) osd_recovery_max_active_ssd = 10
    3) osd_max_backfills = 3

The above may be modified if necessary by setting
osd_mclock_override_recovery_settings option.

Fixes: https://tracker.ceph.com/issues/58529
Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>

diff --git a/src/osd/OSD.cc b/src/osd/OSD.cc
index 4fca8a48b87dadf4c6556effed95544daf52d86f..f964e4179fd5be78a49e37332f124cdc082c8e60 100644 (file)
--- a/src/osd/OSD.cc
+++ b/src/osd/OSD.cc
@@ -9987,9 +9987,9 @@ bool OSD::maybe_override_options_for_qos(const std::set<std::string> *changed)
       !unsupported_objstore_for_qos()) {
     static const std::map<std::string, uint64_t> recovery_qos_defaults {
       {"osd_recovery_max_active", 0},
-      {"osd_recovery_max_active_hdd", 10},
-      {"osd_recovery_max_active_ssd", 20},
-      {"osd_max_backfills", 10},
+      {"osd_recovery_max_active_hdd", 3},
+      {"osd_recovery_max_active_ssd", 10},
+      {"osd_max_backfills", 3},
     };
 
     // Check if we were called because of a configuration change