From: Sage Weil Date: Thu, 27 Dec 2012 19:12:33 +0000 (-0800) Subject: osd: drop 'osd recovery max active' back to previous default (5) X-Git-Tag: v0.56~13 X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=82c71716f715cdfd54c81716496df905ea31cbcb;p=ceph.git osd: drop 'osd recovery max active' back to previous default (5) Having this too large means that queues get too deep on the OSDs during backfill and latency is very high. In my tests, it also meant we generated a lot of slow recovery messages just from the recovery ops themselves (no client io). Keeping this at the old default means we are no worse in this respect than argonaut, which is a safe position to start from. Signed-off-by: Sage Weil --- diff --git a/src/common/config_opts.h b/src/common/config_opts.h index 9c62c1958b33..55db517b055e 100644 --- a/src/common/config_opts.h +++ b/src/common/config_opts.h @@ -337,7 +337,7 @@ OPTION(osd_default_data_pool_replay_window, OPT_INT, 45) OPTION(osd_preserve_trimmed_log, OPT_BOOL, false) OPTION(osd_auto_mark_unfound_lost, OPT_BOOL, false) OPTION(osd_recovery_delay_start, OPT_FLOAT, 15) -OPTION(osd_recovery_max_active, OPT_INT, 20) +OPTION(osd_recovery_max_active, OPT_INT, 5) OPTION(osd_recovery_max_chunk, OPT_U64, 8<<20) // max size of push chunk OPTION(osd_recovery_forget_lost_objects, OPT_BOOL, false) // off for now OPTION(osd_max_scrubs, OPT_INT, 1)