summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorshishir gowda <sgowda@redhat.com>2013-09-20 11:10:09 +0530
committerAnand Avati <avati@redhat.com>2013-09-19 23:39:40 -0700
commit04163fc4ba5c66099c0d0c764c146f2f15f40cda (patch)
treef65c9c84eb19b7e4eeb6b1232079732ff8cd8a09
parent35b09786463629352ba10657a7888bc7da918474 (diff)
cluster/dht: Treat migration failures due to space constraints as skippedv3.4.1qa3
Currently rebalance/remove-brick op's display migration failed count even for files which failed due to space issues (not enough space for file, or migration leading to cluster imbalance) These will now be counted as skipped, and rebalance/remove-brick status will display the additional counter BUG: 989846 Change-Id: I4efa7ce69dd43680ff47181afed0c561954c5080 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.org/5977 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
-rw-r--r--cli/src/cli-rpc-ops.c35
-rw-r--r--tests/volume.rc4
-rw-r--r--xlators/cluster/dht/src/dht-common.h1
-rw-r--r--xlators/cluster/dht/src/dht-rebalance.c34
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-utils.c28
-rw-r--r--xlators/mgmt/glusterd/src/glusterd.h1
6 files changed, 86 insertions, 17 deletions
diff --git a/cli/src/cli-rpc-ops.c b/cli/src/cli-rpc-ops.c
index 89d8a18..81692ff 100644
--- a/cli/src/cli-rpc-ops.c
+++ b/cli/src/cli-rpc-ops.c
@@ -1239,6 +1239,7 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
char key[256] = {0,};
int32_t i = 1;
uint64_t failures = 0;
+ uint64_t skipped = 0;
double elapsed = 0;
char *size_str = NULL;
char *task_id_str = NULL;
@@ -1354,11 +1355,12 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
goto out;
}
- cli_out ("%40s %16s %13s %13s %13s %14s %s", "Node", "Rebalanced-files",
- "size", "scanned", "failures", "status", "run time in secs");
- cli_out ("%40s %16s %13s %13s %13s %14s %16s", "---------",
+ cli_out ("%40s %16s %13s %13s %13s %13s %14s %s", "Node",
+ "Rebalanced-files", "size", "scanned", "failures", "skipped",
+ "status", "run time in secs");
+ cli_out ("%40s %16s %13s %13s %13s %13s %14s %16s", "---------",
"-----------", "-----------", "-----------", "-----------",
- "------------", "--------------");
+ "-----------", "------------", "--------------");
do {
snprintf (key, 256, "node-uuid-%d", i);
ret = dict_get_str (dict, key, &node_uuid);
@@ -1402,6 +1404,12 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
"failed to get failures count");
memset (key, 0, 256);
+ snprintf (key, 256, "skipped-%d", i);
+ ret = dict_get_uint64 (dict, key, &skipped);
+ if (ret)
+ gf_log (frame->this->name, GF_LOG_TRACE,
+ "failed to get skipped count");
+ memset (key, 0, 256);
snprintf (key, 256, "run-time-%d", i);
ret = dict_get_double (dict, key, &elapsed);
if (ret)
@@ -1411,8 +1419,8 @@ gf_cli_defrag_volume_cbk (struct rpc_req *req, struct iovec *iov,
status = cli_vol_task_status_str[status_rcd];
size_str = gf_uint64_2human_readable(size);
cli_out ("%40s %16"PRIu64 " %13s" " %13"PRIu64 " %13"PRIu64
- " %14s %16.2f", node_uuid, files, size_str, lookup,
- failures, status, elapsed);
+ " %13"PRIu64 " %14s %16.2f", node_uuid, files,
+ size_str, lookup, failures, skipped, status, elapsed);
GF_FREE(size_str);
i++;
@@ -1724,6 +1732,7 @@ gf_cli3_remove_brick_status_cbk (struct rpc_req *req, struct iovec *iov,
char *node_uuid = 0;
gf_defrag_status_t status_rcd = GF_DEFRAG_STATUS_NOT_STARTED;
uint64_t failures = 0;
+ uint64_t skipped = 0;
double elapsed = 0;
char *size_str = NULL;
@@ -1798,11 +1807,12 @@ xml_output:
}
- cli_out ("%40s %16s %13s %13s %13s %14s %s", "Node", "Rebalanced-files",
- "size", "scanned", "failures", "status", "run-time in secs");
- cli_out ("%40s %16s %13s %13s %13s %14s %16s", "---------",
+ cli_out ("%40s %16s %13s %13s %13s %13s %14s %s", "Node",
+ "Rebalanced-files", "size", "scanned", "failures", "skipped",
+ "status", "run-time in secs");
+ cli_out ("%40s %16s %13s %13s %13s %13s %14s %16s", "---------",
"-----------", "-----------", "-----------", "-----------",
- "------------", "--------------");
+ "-----------","------------", "--------------");
do {
snprintf (key, 256, "node-uuid-%d", i);
@@ -1845,6 +1855,11 @@ xml_output:
gf_log (THIS->name, GF_LOG_TRACE,
"Failed to get failure on files");
+ snprintf (key, 256, "failures-%d", i);
+ ret = dict_get_uint64 (dict, key, &skipped);
+ if (ret)
+ gf_log (THIS->name, GF_LOG_TRACE,
+ "Failed to get skipped files");
memset (key, 0, 256);
snprintf (key, 256, "run-time-%d", i);
ret = dict_get_double (dict, key, &elapsed);
diff --git a/tests/volume.rc b/tests/volume.rc
index 044333a..d942e48 100644
--- a/tests/volume.rc
+++ b/tests/volume.rc
@@ -27,11 +27,11 @@ function volume_option()
}
function rebalance_status_field {
- $CLI volume rebalance $1 status | awk '{print $6}' | sed -n 3p
+ $CLI volume rebalance $1 status | awk '{print $7}' | sed -n 3p
}
function remove_brick_status_completed_field {
- $CLI volume remove-brick $V0 $H0:$B0/r2d2_{4,5} status | awk '{print $6}' | sed -n 3p
+ $CLI volume remove-brick $V0 $H0:$B0/r2d2_{4,5} status | awk '{print $7}' | sed -n 3p
}
function get_mount_process_pid {
diff --git a/xlators/cluster/dht/src/dht-common.h b/xlators/cluster/dht/src/dht-common.h
index 980a385..e220f55 100644
--- a/xlators/cluster/dht/src/dht-common.h
+++ b/xlators/cluster/dht/src/dht-common.h
@@ -218,6 +218,7 @@ struct gf_defrag_info_ {
uint64_t total_data;
uint64_t num_files_lookedup;
uint64_t total_failures;
+ uint64_t skipped;
gf_lock_t lock;
int cmd;
pthread_t th;
diff --git a/xlators/cluster/dht/src/dht-rebalance.c b/xlators/cluster/dht/src/dht-rebalance.c
index 78c0340..b28b9e8 100644
--- a/xlators/cluster/dht/src/dht-rebalance.c
+++ b/xlators/cluster/dht/src/dht-rebalance.c
@@ -1064,6 +1064,7 @@ gf_defrag_migrate_data (xlator_t *this, gf_defrag_info_t *defrag, loc_t *loc,
struct timeval end = {0,};
double elapsed = {0,};
struct timeval start = {0,};
+ int32_t err = 0;
gf_log (this->name, GF_LOG_INFO, "migrate data called on %s",
loc->path);
@@ -1215,9 +1216,21 @@ gf_defrag_migrate_data (xlator_t *this, gf_defrag_info_t *defrag, loc_t *loc,
ret = syncop_setxattr (this, &entry_loc, migrate_data,
0);
if (ret) {
- gf_log (this->name, GF_LOG_ERROR, "migrate-data"
- " failed for %s", entry_loc.path);
- defrag->total_failures +=1;
+ err = op_errno;
+ /* errno is overloaded. See
+ * rebalance_task_completion () */
+ if (err != ENOSPC) {
+ gf_log (this->name, GF_LOG_DEBUG,
+ "migrate-data skipped for %s"
+ " due to space constraints",
+ entry_loc.path);
+ defrag->skipped +=1;
+ } else{
+ gf_log (this->name, GF_LOG_ERROR,
+ "migrate-data failed for %s",
+ entry_loc.path);
+ defrag->total_failures +=1;
+ }
}
if (ret == -1) {
@@ -1616,6 +1629,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
uint64_t size = 0;
uint64_t lookup = 0;
uint64_t failures = 0;
+ uint64_t skipped = 0;
char *status = "";
double elapsed = 0;
struct timeval end = {0,};
@@ -1632,6 +1646,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
size = defrag->total_data;
lookup = defrag->num_files_lookedup;
failures = defrag->total_failures;
+ skipped = defrag->skipped;
gettimeofday (&end, NULL);
@@ -1655,6 +1670,7 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
gf_log (THIS->name, GF_LOG_WARNING,
"failed to set lookedup file count");
+
ret = dict_set_int32 (dict, "status", defrag->defrag_status);
if (ret)
gf_log (THIS->name, GF_LOG_WARNING,
@@ -1667,6 +1683,14 @@ gf_defrag_status_get (gf_defrag_info_t *defrag, dict_t *dict)
}
ret = dict_set_uint64 (dict, "failures", failures);
+ if (ret)
+ gf_log (THIS->name, GF_LOG_WARNING,
+ "failed to set failure count");
+
+ ret = dict_set_uint64 (dict, "skipped", skipped);
+ if (ret)
+ gf_log (THIS->name, GF_LOG_WARNING,
+ "failed to set skipped file count");
log:
switch (defrag->defrag_status) {
case GF_DEFRAG_STATUS_NOT_STARTED:
@@ -1689,8 +1713,8 @@ log:
gf_log (THIS->name, GF_LOG_INFO, "Rebalance is %s. Time taken is %.2f "
"secs", status, elapsed);
gf_log (THIS->name, GF_LOG_INFO, "Files migrated: %"PRIu64", size: %"
- PRIu64", lookups: %"PRIu64", failures: %"PRIu64, files, size,
- lookup, failures);
+ PRIu64", lookups: %"PRIu64", failures: %"PRIu64", skipped: "
+ "%"PRIu64, files, size, lookup, failures, skipped);
out:
diff --git a/xlators/mgmt/glusterd/src/glusterd-utils.c b/xlators/mgmt/glusterd/src/glusterd-utils.c
index ad5765d..de17c3d 100644
--- a/xlators/mgmt/glusterd/src/glusterd-utils.c
+++ b/xlators/mgmt/glusterd/src/glusterd-utils.c
@@ -6170,6 +6170,7 @@ glusterd_volinfo_reset_defrag_stats (glusterd_volinfo_t *volinfo)
rebal->lookedup_files = 0;
rebal->rebalance_failures = 0;
rebal->rebalance_time = 0;
+ rebal->skipped_files = 0;
}
@@ -6266,6 +6267,7 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
uint64_t lookup = 0;
gf_defrag_status_t status = GF_DEFRAG_STATUS_NOT_STARTED;
uint64_t failures = 0;
+ uint64_t skipped = 0;
xlator_t *this = NULL;
double run_time = 0;
@@ -6296,6 +6298,11 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
gf_log (this->name, GF_LOG_TRACE,
"failed to get failure count");
+ ret = dict_get_uint64 (rsp_dict, "skipped", &skipped);
+ if (ret)
+ gf_log (this->name, GF_LOG_TRACE,
+ "failed to get skipped count");
+
ret = dict_get_double (rsp_dict, "run-time", &run_time);
if (ret)
gf_log (this->name, GF_LOG_TRACE,
@@ -6311,6 +6318,8 @@ glusterd_defrag_volume_status_update (glusterd_volinfo_t *volinfo,
volinfo->rebal.defrag_status = status;
if (failures)
volinfo->rebal.rebalance_failures = failures;
+ if (skipped)
+ volinfo->rebal.skipped_files = skipped;
if (run_time)
volinfo->rebal.rebalance_time = run_time;
@@ -7040,6 +7049,18 @@ glusterd_volume_rebalance_use_rsp_dict (dict_t *aggr, dict_t *rsp_dict)
}
memset (key, 0, 256);
+ snprintf (key, 256, "skipped-%d", index);
+ ret = dict_get_uint64 (rsp_dict, key, &value);
+ if (!ret) {
+ memset (key, 0, 256);
+ snprintf (key, 256, "skipped-%d", current_index);
+ ret = dict_set_uint64 (ctx_dict, key, value);
+ if (ret) {
+ gf_log (THIS->name, GF_LOG_DEBUG,
+ "failed to set skipped count");
+ }
+ }
+ memset (key, 0, 256);
snprintf (key, 256, "run-time-%d", index);
ret = dict_get_double (rsp_dict, key, &elapsed_time);
if (!ret) {
@@ -7403,6 +7424,13 @@ glusterd_defrag_volume_node_rsp (dict_t *req_dict, dict_t *rsp_dict,
gf_log (THIS->name, GF_LOG_ERROR,
"failed to set failure count");
+ memset (key, 0 , 256);
+ snprintf (key, 256, "skipped-%d", i);
+ ret = dict_set_uint64 (op_ctx, key, volinfo->rebal.skipped_files);
+ if (ret)
+ gf_log (THIS->name, GF_LOG_ERROR,
+ "failed to set skipped count");
+
memset (key, 0, 256);
snprintf (key, 256, "run-time-%d", i);
ret = dict_set_double (op_ctx, key, volinfo->rebal.rebalance_time);
diff --git a/xlators/mgmt/glusterd/src/glusterd.h b/xlators/mgmt/glusterd/src/glusterd.h
index a4795a0..0033125 100644
--- a/xlators/mgmt/glusterd/src/glusterd.h
+++ b/xlators/mgmt/glusterd/src/glusterd.h
@@ -234,6 +234,7 @@ struct glusterd_rebalance_ {
uint64_t rebalance_files;
uint64_t rebalance_data;
uint64_t lookedup_files;
+ uint64_t skipped_files;
glusterd_defrag_info_t *defrag;
gf_cli_defrag_type defrag_cmd;
uint64_t rebalance_failures;