summaryrefslogtreecommitdiffstats
path: root/xlators/cluster/ec/src/ec-heald.c
diff options
context:
space:
mode:
authorXavi Hernandez <jahernan@redhat.com>2018-02-21 17:47:37 +0100
committerAmar Tumballi <amarts@redhat.com>2018-03-14 03:12:27 +0000
commit7f81067f4522f973e98aa5abbb4d2028da2a2e6f (patch)
treecdcfa501a27a2c5bc472495e27b2d37d35fdc068 /xlators/cluster/ec/src/ec-heald.c
parentfe52fc33d00eafe7d52ffff1b2dab846374f1d4a (diff)
cluster/ec: avoid delays in self-heal
Self-heal creates a thread per brick to sweep the index looking for files that need to be healed. These threads are started before the volume comes online, so nothing is done but waiting for the next sweep. This happens once per minute. When a replace brick command is executed, the new graph is loaded and all index sweeper threads started. When all bricks have reported, a getxattr request is sent to the root directory of the volume. This causes a heal on it (because the new brick doesn't have good data), and marks its contents as pending to be healed. This is done by the index sweeper thread on the next round, one minute later. This patch solves this problem by waking all index sweeper threads after a successful check on the root directory. Additionally, the index sweep thread scans the index directory sequentially, but it might happen that after healing a directory entry more index entries are created but skipped by the current directory scan. This causes the remaining entries to be processed on the next round, one minute later. The same can happen in the next round, so the heal is running in bursts and taking a lot to finish, specially on volumes with many directory levels. This patch solves this problem by immediately restarting the index sweep if a directory has been healed. Change-Id: I58d9ab6ef17b30f704dc322e1d3d53b904e5f30e BUG: 1547662 Signed-off-by: Xavi Hernandez <jahernan@redhat.com>
Diffstat (limited to 'xlators/cluster/ec/src/ec-heald.c')
-rw-r--r--xlators/cluster/ec/src/ec-heald.c27
1 files changed, 21 insertions, 6 deletions
diff --git a/xlators/cluster/ec/src/ec-heald.c b/xlators/cluster/ec/src/ec-heald.c
index b4fa6f87189..a703379a59b 100644
--- a/xlators/cluster/ec/src/ec-heald.c
+++ b/xlators/cluster/ec/src/ec-heald.c
@@ -184,8 +184,19 @@ ec_shd_index_purge (xlator_t *subvol, inode_t *inode, char *name)
int
ec_shd_selfheal (struct subvol_healer *healer, int child, loc_t *loc)
{
- return syncop_getxattr (healer->this, loc, NULL, EC_XATTR_HEAL, NULL,
- NULL);
+ int32_t ret;
+
+ ret = syncop_getxattr (healer->this, loc, NULL, EC_XATTR_HEAL, NULL,
+ NULL);
+ if ((ret >= 0) && (loc->inode->ia_type == IA_IFDIR)) {
+ /* If we have just healed a directory, it's possible that
+ * other index entries have appeared to be healed. We put a
+ * mark so that we can check it later and restart a scan
+ * without delay. */
+ healer->rerun = _gf_true;
+ }
+
+ return ret;
}
@@ -472,11 +483,15 @@ ec_shd_index_healer_spawn (xlator_t *this, int subvol)
}
void
-ec_selfheal_childup (ec_t *ec, int child)
+ec_shd_index_healer_wake(ec_t *ec)
{
- if (!ec->shd.iamshd)
- return;
- ec_shd_index_healer_spawn (ec->xl, child);
+ int32_t i;
+
+ for (i = 0; i < ec->nodes; i++) {
+ if (((ec->xl_up >> i) & 1) != 0) {
+ ec_shd_index_healer_spawn(ec->xl, i);
+ }
+ }
}
int