From bbceb48f0fe8b1fc2604fd3fcd347e143cab600a Mon Sep 17 00:00:00 2001 From: Richard Wareing Date: Tue, 29 Apr 2014 21:05:35 -0700 Subject: Halo Replication feature for AFR translator Summary: Halo Geo-replication is a feature which allows Gluster or NFS clients to write locally to their region (as defined by a latency "halo" or threshold if you like), and have their writes asynchronously propagate from their origin to the rest of the cluster. Clients can also write synchronously to the cluster simply by specifying a halo-latency which is very large (e.g. 10seconds) which will include all bricks. In other words, it allows clients to decide at mount time if they desire synchronous or asynchronous IO into a cluster and the cluster can support both of these modes to any number of clients simultaneously. There are a few new volume options due to this feature: halo-shd-latency: The threshold below which self-heal daemons will consider children (bricks) connected. halo-nfsd-latency: The threshold below which NFS daemons will consider children (bricks) connected. halo-latency: The threshold below which all other clients will consider children (bricks) connected. halo-min-replicas: The minimum number of replicas which are to be enforced regardless of latency specified in the above 3 options. If the number of children falls below this threshold the next best (chosen by latency) shall be swapped in. New FUSE mount options: halo-latency & halo-min-replicas: As descripted above. This feature combined with multi-threaded SHD support (D1271745) results in some pretty cool geo-replication possibilities. Operational Notes: - Global consistency is gaurenteed for synchronous clients, this is provided by the existing entry-locking mechanism. - Asynchronous clients on the other hand and merely consistent to their region. Writes & deletes will be protected via entry-locks as usual preventing concurrent writes into files which are undergoing replication. Read operations on the other hand should never block. - Writes are allowed from _any_ region and propagated from the origin to all other regions. The take away from this is care should be taken to ensure multiple writers do not write the same files resulting in a gfid split-brain which will require resolution via split-brain policies (majority, mtime & size). Recommended method for preventing this is using the nfs-auth feature to define which region for each share has RW permissions, tiers not in the origin region should have RO perms. TODO: - Synchronous clients (including the SHD) should choose clients from their own region as preferred sources for reads. Most of the plumbing is in place for this via the child_latency array. - Better GFID split brain handling & better dent type split brain handling (i.e. create a trash can and move the offending files into it). - Tagging in addition to latency as a means of defining which children you wish to synchronously write to Test Plan: - The usual suspects, clang, gcc w/ address sanitizer & valgrind - Prove tests Reviewers: jackl, dph, cjh, meyering Reviewed By: meyering Subscribers: ethanr Differential Revision: https://phabricator.fb.com/D1272053 Tasks: 4117827 Change-Id: I694a9ab429722da538da171ec528406e77b5e6d1 Signed-off-by: Kevin Vigor Reviewed-on: http://review.gluster.org/16099 NetBSD-regression: NetBSD Build System Smoke: Gluster Build System CentOS-regression: Gluster Build System Reviewed-by: Shreyas Siravara --- libglusterfs/src/latency.c | 12 ++++++++++-- libglusterfs/src/mem-types.h | 1 + libglusterfs/src/timespec.c | 12 ++++++++++++ libglusterfs/src/timespec.h | 3 +++ libglusterfs/src/xlator.h | 1 + 5 files changed, 27 insertions(+), 2 deletions(-) (limited to 'libglusterfs/src') diff --git a/libglusterfs/src/latency.c b/libglusterfs/src/latency.c index 611615949fa..3399cc7c297 100644 --- a/libglusterfs/src/latency.c +++ b/libglusterfs/src/latency.c @@ -21,6 +21,7 @@ #include "statedump.h" #include "libglusterfs-messages.h" +static int gf_set_fop_from_fn_pointer_warning; void gf_set_fop_from_fn_pointer (call_frame_t *frame, struct xlator_fops *fops, void *fn) { @@ -108,8 +109,15 @@ gf_set_fop_from_fn_pointer (call_frame_t *frame, struct xlator_fops *fops, void fop = GF_FOP_READDIRP; else if (fops->getspec == *(fop_getspec_t *)&fn) fop = GF_FOP_GETSPEC; - else - fop = -1; + else if (fops->ipc == *(fop_ipc_t *)&fn) + fop = GF_FOP_IPC; + else { + fop = GF_FOP_NULL; + GF_LOG_OCCASIONALLY(gf_set_fop_from_fn_pointer_warning, + "latency", + GF_LOG_WARNING, + "Unknown FOP type"); + } frame->op = fop; } diff --git a/libglusterfs/src/mem-types.h b/libglusterfs/src/mem-types.h index afa52d8bc45..fc7bf9e5996 100644 --- a/libglusterfs/src/mem-types.h +++ b/libglusterfs/src/mem-types.h @@ -168,6 +168,7 @@ enum gf_common_mem_types_ { /*lock migration*/ gf_common_mt_lock_mig, gf_common_mt_pthread_t, + gf_common_ping_local_t, gf_common_mt_end }; #endif diff --git a/libglusterfs/src/timespec.c b/libglusterfs/src/timespec.c index f7b2bea2f30..903303d1380 100644 --- a/libglusterfs/src/timespec.c +++ b/libglusterfs/src/timespec.c @@ -60,3 +60,15 @@ void timespec_adjust_delta (struct timespec *ts, struct timespec delta) ts->tv_sec += ((ts->tv_nsec + delta.tv_nsec) / 1000000000); ts->tv_sec += delta.tv_sec; } + +void timespec_sub (const struct timespec *begin, const struct timespec *end, + struct timespec *res) +{ + if (end->tv_nsec < begin->tv_nsec) { + res->tv_sec = end->tv_sec - begin->tv_sec - 1; + res->tv_nsec = end->tv_nsec + 1000000000 - begin->tv_nsec; + } else { + res->tv_sec = end->tv_sec - begin->tv_sec; + res->tv_nsec = end->tv_nsec - begin->tv_nsec; + } +} diff --git a/libglusterfs/src/timespec.h b/libglusterfs/src/timespec.h index f37194b97cf..9c393ee7166 100644 --- a/libglusterfs/src/timespec.h +++ b/libglusterfs/src/timespec.h @@ -20,5 +20,8 @@ void timespec_now (struct timespec *ts); void timespec_adjust_delta (struct timespec *ts, struct timespec delta); +void timespec_sub (const struct timespec *begin, + const struct timespec *end, + struct timespec *res); #endif /* __INCLUDE_TIMESPEC_H__ */ diff --git a/libglusterfs/src/xlator.h b/libglusterfs/src/xlator.h index 70e6f0a108d..571dce2aded 100644 --- a/libglusterfs/src/xlator.h +++ b/libglusterfs/src/xlator.h @@ -927,6 +927,7 @@ struct _xlator { gf_loglevel_t loglevel; /* Log level for translator */ + int64_t client_latency; /* for latency measurement */ fop_latency_t latencies[GF_FOP_MAXVALUE]; -- cgit .git/diff/?h=v3.3.1qa2&id=74383e3ec6f8244b3de9bf14016452498c1ddcf0'>Diffstat
-rw-r--r--cli/src/cli-cmd-volume.c10
-rw-r--r--cli/src/cli.c5
-rw-r--r--cli/src/cli.h9
-rw-r--r--tests/bugs/glusterd/bug-1446172-brick-mux-reset-brick.t79
-rw-r--r--tests/include.rc7
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-brick-ops.c3
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-messages.h10
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-op-sm.c14
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-replace-brick.c3
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-reset-brick.c79
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-snapshot.c2
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-utils.c44
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-utils.h12
-rw-r--r--xlators/mgmt/glusterd/src/glusterd-volume-ops.c2
14 files changed, 194 insertions, 85 deletions
diff --git a/cli/src/cli-cmd-volume.c b/cli/src/cli-cmd-volume.c
index 61b6b2090a9..41995791cdd 100644
--- a/cli/src/cli-cmd-volume.c
+++ b/cli/src/cli-cmd-volume.c
@@ -2087,6 +2087,7 @@ cli_cmd_volume_reset_brick_cbk (struct cli_state *state,
if (!frame)
goto out;
+
ret = cli_cmd_volume_reset_brick_parse (words, wordcount, &options);
if (ret) {
@@ -2095,6 +2096,15 @@ cli_cmd_volume_reset_brick_cbk (struct cli_state *state,
goto out;
}
+ if (state->mode & GLUSTER_MODE_WIGNORE_PARTITION) {
+ ret = dict_set_int32 (options, "ignore-partition", _gf_true);
+ if (ret) {
+ gf_log ("cli", GF_LOG_ERROR, "Failed to set ignore-"
+ "partition option");
+ goto out;
+ }
+ }
+
CLI_LOCAL_INIT (local, words, frame, options);
if (proc->fn) {
diff --git a/cli/src/cli.c b/cli/src/cli.c
index 574ee0eea33..cd016514cab 100644
--- a/cli/src/cli.c
+++ b/cli/src/cli.c
@@ -348,6 +348,11 @@ cli_opt_parse (char *opt, struct cli_state *state)
return 0;
}
+ if (strcmp (opt, "wignore-partition") == 0) {
+ state->mode |= GLUSTER_MODE_WIGNORE_PARTITION;
+ return 0;
+ }
+
if (strcmp (opt, "wignore") == 0) {
state->mode |= GLUSTER_MODE_WIGNORE;
return 0;
diff --git a/cli/src/cli.h b/cli/src/cli.h
index 60492c1b9fc..1e92d675199 100644
--- a/cli/src/cli.h
+++ b/cli/src/cli.h
@@ -56,10 +56,11 @@ typedef enum {
MAX
} values;
-#define GLUSTER_MODE_SCRIPT (1 << 0)
-#define GLUSTER_MODE_ERR_FATAL (1 << 1)
-#define GLUSTER_MODE_XML (1 << 2)
-#define GLUSTER_MODE_WIGNORE (1 << 3)
+#define GLUSTER_MODE_SCRIPT (1 << 0)
+#define GLUSTER_MODE_ERR_FATAL (1 << 1)
+#define GLUSTER_MODE_XML (1 << 2)
+#define GLUSTER_MODE_WIGNORE (1 << 3)
+#define GLUSTER_MODE_WIGNORE_PARTITION (1 << 4)
#define GLUSTERD_GET_QUOTA_LIST_MOUNT_PATH(abspath, volname, path) do { \
diff --git a/tests/bugs/glusterd/bug-1446172-brick-mux-reset-brick.t b/tests/bugs/glusterd/bug-1446172-brick-mux-reset-brick.t
new file mode 100644
index 00000000000..e6aaaa4e87c
--- /dev/null
+++ b/tests/bugs/glusterd/bug-1446172-brick-mux-reset-brick.t
@@ -0,0 +1,79 @@
+#!/bin/bash
+
+. $(dirname $0)/../../include.rc
+. $(dirname $0)/../../traps.rc
+. $(dirname $0)/../../volume.rc
+
+cleanup;
+
+function count_up_bricks {
+ $CLI --xml volume status | grep '<status>1' | wc -l
+}
+
+function count_brick_processes {
+ pgrep glusterfsd | wc -l
+}
+
+TEST glusterd
+
+TEST $CLI volume set all cluster.brick-multiplex on
+push_trapfunc "$CLI volume set all cluster.brick-multiplex off"
+push_trapfunc "cleanup"
+
+TEST $CLI volume create $V0 $H0:$B0/${V0}{0,1}
+TEST $CLI volume create $V1 $H0:$B0/${V1}{0,1}
+
+TEST $CLI volume start $V0
+TEST $CLI volume start $V1
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 4 count_up_bricks
+EXPECT 1 count_brick_processes
+
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;
+# Create files
+for i in {1..5}
+do
+ echo $i > $M0/file$i.txt
+done
+
+TEST $CLI volume reset-brick $V0 $H0:$B0/${V0}1 start
+
+EXPECT_WITHIN $PROCESS_DOWN_TIMEOUT 3 count_up_bricks
+EXPECT 1 count_brick_processes
+
+# Negative case with brick killed but volume-id xattr present
+TEST ! $CLI volume reset-brick $V0 $H0:$B0/${V0}1 $H0:$B0/${V0}1 commit
+
+# reset-brick commit force should work and should bring up the brick
+TEST $CLI volume reset-brick $V0 $H0:$B0/${V0}1 $H0:$B0/${V0}1 commit force
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 4 count_up_bricks
+EXPECT 1 count_brick_processes
+
+TEST glusterfs --volfile-id=$V1 --volfile-server=$H0 $M1;
+# Create files
+for i in {1..5}
+do
+ echo $i > $M1/file$i.txt
+done
+
+TEST $CLI volume reset-brick $V1 $H0:$B0/${V1}1 start
+
+EXPECT_WITHIN $PROCESS_DOWN_TIMEOUT 3 count_up_bricks
+EXPECT 1 count_brick_processes
+
+# Simulate reset disk
+for i in {1..5}
+do
+ rm -rf $B0/${V1}1/file$i.txt
+done
+
+setfattr -x trusted.glusterfs.volume-id $B0/${V1}1
+setfattr -x trusted.gfid $B0/${V1}1
+
+# Test reset-brick commit. Using CLI_IGNORE_PARTITION since normal CLI uses
+# the --wignore flag that essentially makes the command act like "commit force"
+TEST $CLI_IGNORE_PARTITION volume reset-brick $V1 $H0:$B0/${V1}1 $H0:$B0/${V1}1 commit
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 4 count_up_bricks
+EXPECT 1 count_brick_processes
diff --git a/tests/include.rc b/tests/include.rc
index 9e6cea4b487..d92361006eb 100644
--- a/tests/include.rc
+++ b/tests/include.rc
@@ -90,7 +90,12 @@ LOGDIR=$(gluster --print-logdir)
statedumpdir=`gluster --print-statedumpdir`; # Default directory for statedump
CLI="gluster --mode=script --wignore";
-CLI_NO_FORCE="gluster --mode-script";
+CLI_NO_FORCE="gluster --mode=script";
+
+# CLI_IGNORE_PARTITION makes sure that the warning related to bricks being on
+# root partition is ignored while running the command in a "no force" mode
+CLI_IGNORE_PARTITION="gluster --mode=script --wignore-partition"
+
_GFS () {
glusterfs "$@"
local mount_ret=$?
diff --git a/xlators/mgmt/glusterd/src/glusterd-brick-ops.c b/xlators/mgmt/glusterd/src/glusterd-brick-ops.c
index b1b13a49f71..21e3377b631 100644
--- a/xlators/mgmt/glusterd/src/glusterd-brick-ops.c
+++ b/xlators/mgmt/glusterd/src/glusterd-brick-ops.c
@@ -1883,7 +1883,8 @@ glusterd_op_stage_add_brick (dict_t *dict, char **op_errstr, dict_t *rsp_dict)
ret = glusterd_validate_and_create_brickpath (brickinfo,
volinfo->volume_id,
- op_errstr, is_force);
+ op_errstr, is_force,
+ _gf_false);
if (ret)
goto out;
diff --git a/xlators/mgmt/glusterd/src/glusterd-messages.h b/xlators/mgmt/glusterd/src/glusterd-messages.h
index 8cfc26ce905..e66107c3b24 100644
--- a/xlators/mgmt/glusterd/src/glusterd-messages.h
+++ b/xlators/mgmt/glusterd/src/glusterd-messages.h
@@ -41,7 +41,7 @@
#define GLUSTERD_COMP_BASE GLFS_MSGID_GLUSTERD
-#define GLFS_NUM_MESSAGES 600
+#define GLFS_NUM_MESSAGES 601
#define GLFS_MSGID_END (GLUSTERD_COMP_BASE + GLFS_NUM_MESSAGES + 1)
/* Messaged with message IDs */
@@ -4853,6 +4853,14 @@
*/
#define GD_MSG_GNFS_XLATOR_NOT_INSTALLED (GLUSTERD_COMP_BASE + 600)
+/*!
+ * @messageid
+ * @diagnosis
+ * @recommendedaction
+ *
+ */
+#define GD_MSG_PIDFILE_UNLINKING (GLUSTERD_COMP_BASE + 601)
+
/*------------*/
#define glfs_msg_end_x GLFS_MSGID_END, "Invalid: End of messages"
diff --git a/xlators/mgmt/glusterd/src/glusterd-op-sm.c b/xlators/mgmt/glusterd/src/glusterd-op-sm.c
index 3558ae91059..1a12b0e7160 100644
--- a/xlators/mgmt/glusterd/src/glusterd-op-sm.c
+++ b/xlators/mgmt/glusterd/src/glusterd-op-sm.c
@@ -6196,8 +6196,6 @@ glusterd_bricks_select_stop_volume (dict_t *dict, char **op_errstr,
glusterd_volinfo_t *volinfo = NULL;
glusterd_brickinfo_t *brickinfo = NULL;
glusterd_pending_node_t *pending_node = NULL;
- glusterd_conf_t *conf = THIS->private;
- char pidfile[PATH_MAX] = {0,};
ret = glusterd_op_stop_volume_args_get (dict, &volname, &flags);
if (ret)
@@ -6233,11 +6231,6 @@ glusterd_bricks_select_stop_volume (dict_t *dict, char **op_errstr,
*/
brickinfo->status = GF_BRICK_STOPPED;
brickinfo->started_here = _gf_false;
- GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo,
- brickinfo, conf);
- gf_log (THIS->name, GF_LOG_INFO,
- "unlinking pidfile %s", pidfile);
- (void) sys_unlink (pidfile);
}
}
@@ -6260,8 +6253,6 @@ glusterd_bricks_select_remove_brick (dict_t *dict, char **op_errstr,
glusterd_pending_node_t *pending_node = NULL;
int32_t command = 0;
int32_t force = 0;
- glusterd_conf_t *conf = THIS->private;
- char pidfile[PATH_MAX] = {0,};
ret = dict_get_str (dict, "volname", &volname);
@@ -6342,11 +6333,6 @@ glusterd_bricks_select_remove_brick (dict_t *dict, char **op_errstr,
*/
brickinfo->status = GF_BRICK_STOPPED;
brickinfo->started_here = _gf_false;
- GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo,
- brickinfo, conf);
- gf_log (THIS->name, GF_LOG_INFO,
- "unlinking pidfile %s", pidfile);
- (void) sys_unlink (pidfile);
}
i++;
}
diff --git a/xlators/mgmt/glusterd/src/glusterd-replace-brick.c b/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
index fb29c6efcfd..887c3496475 100644
--- a/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
+++ b/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
@@ -252,7 +252,8 @@ glusterd_op_stage_replace_brick (dict_t *dict, char **op_errstr,
if (gf_is_local_addr (host)) {
ret = glusterd_validate_and_create_brickpath (dst_brickinfo,
volinfo->volume_id,
- op_errstr, is_force);
+ op_errstr, is_force,
+ _gf_false);
if (ret)
goto out;
}
diff --git a/xlators/mgmt/glusterd/src/glusterd-reset-brick.c b/xlators/mgmt/glusterd/src/glusterd-reset-brick.c
index 2e547c7e09f..c127d64d119 100644
--- a/xlators/mgmt/glusterd/src/glusterd-reset-brick.c
+++ b/xlators/mgmt/glusterd/src/glusterd-reset-brick.c
@@ -47,6 +47,7 @@ glusterd_reset_brick_prevalidate (dict_t *dict, char **op_errstr,
char pidfile[PATH_MAX] = {0};
xlator_t *this = NULL;
gf_boolean_t is_force = _gf_false;
+ int32_t ignore_partition = 0;
pid_t pid = -1;
uuid_t volume_id = {0,};
char *dup_dstbrick = NULL;
@@ -153,11 +154,14 @@ glusterd_reset_brick_prevalidate (dict_t *dict, char **op_errstr,
volinfo->rep_brick.src_brick = src_brickinfo;
volinfo->rep_brick.dst_brick = dst_brickinfo;
+ ret = dict_get_int32 (dict, "ignore-partition", &ignore_partition);
+ ret = 0;
if (gf_is_local_addr (host)) {
ret = glusterd_validate_and_create_brickpath
(dst_brickinfo,
volinfo->volume_id,
- op_errstr, is_force);
+ op_errstr, is_force,
+ ignore_partition);
if (ret)
goto out;
} else {
@@ -234,17 +238,16 @@ out:
int
glusterd_op_reset_brick (dict_t *dict, dict_t *rsp_dict)
{
- int ret = 0;
- char *op = NULL;
- glusterd_volinfo_t *volinfo = NULL;
- char *volname = NULL;
- xlator_t *this = NULL;
- glusterd_conf_t *priv = NULL;
- char *src_brick = NULL;
- char *dst_brick = NULL;
- glusterd_brickinfo_t *src_brickinfo = NULL;
- glusterd_brickinfo_t *dst_brickinfo = NULL;
- char pidfile[PATH_MAX] = {0,};
+ int ret = 0;
+ char *op = NULL;
+ glusterd_volinfo_t *volinfo = NULL;
+ char *volname = NULL;
+ xlator_t *this = NULL;
+ glusterd_conf_t *priv = NULL;
+ char *src_brick = NULL;
+ char *dst_brick = NULL;
+ glusterd_brickinfo_t *src_brickinfo = NULL;
+ glusterd_brickinfo_t *dst_brickinfo = NULL;
this = THIS;
GF_ASSERT (this);
@@ -289,26 +292,18 @@ glusterd_op_reset_brick (dict_t *dict, dict_t *rsp_dict)
}
if (!strcmp (op, "GF_RESET_OP_START")) {
- (void) glusterd_brick_disconnect (src_brickinfo);
- GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo,
- src_brickinfo, priv);
- ret = glusterd_service_stop ("brick", pidfile,
- SIGTERM, _gf_false);
- if (ret == 0) {
- glusterd_set_brick_status (src_brickinfo,
- GF_BRICK_STOPPED);
- (void) glusterd_brick_unlink_socket_file
- (volinfo, src_brickinfo);
- gf_msg (this->name, GF_LOG_INFO, 0,
- GD_MSG_BRICK_CLEANUP_SUCCESS,
- "Brick cleanup successful.");
- } else {
+ ret = glusterd_volume_stop_glusterfs (volinfo,
+ src_brickinfo,
+ _gf_false);
+ if (ret) {
gf_msg (this->name, GF_LOG_CRITICAL, 0,
- GD_MSG_BRK_CLEANUP_FAIL,
- "Unable to cleanup src brick");
- goto out;
+ GD_MSG_BRICK_STOP_FAIL, "Unable to stop"
+ " brick: %s:%s", src_brickinfo->hostname,
+ src_brickinfo->path);
}
+
goto out;
+
} else if (!strcmp (op, "GF_RESET_OP_COMMIT") ||
!strcmp (op, "GF_RESET_OP_COMMIT_FORCE")) {
ret = dict_get_str (dict, "dst-brick", &dst_brick);
@@ -344,25 +339,17 @@ glusterd_op_reset_brick (dict_t *dict, dict_t *rsp_dict)
if (ret)
goto out;
- if (gf_is_local_addr (dst_brickinfo->hostname)) {
+ if (gf_uuid_compare (dst_brickinfo->uuid, MY_UUID)) {
gf_msg_debug (this->name, 0, "I AM THE DESTINATION HOST");
- (void) glusterd_brick_disconnect (src_brickinfo);
- GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo,
- src_brickinfo, priv);
- ret = glusterd_service_stop ("brick", pidfile,
- SIGTERM, _gf_false);
- if (ret == 0) {
- glusterd_set_brick_status
- (src_brickinfo, GF_BRICK_STOPPED);
- (void) glusterd_brick_unlink_socket_file
- (volinfo, src_brickinfo);
- gf_msg (this->name, GF_LOG_INFO, 0,
- GD_MSG_BRICK_CLEANUP_SUCCESS,
- "Brick cleanup successful.");
- } else {
+ ret = glusterd_volume_stop_glusterfs (volinfo,
+ src_brickinfo,
+ _gf_true);
+ if (ret) {
gf_msg (this->name, GF_LOG_CRITICAL, 0,
- GD_MSG_BRK_CLEANUP_FAIL,
- "Unable to cleanup src brick");
+ GD_MSG_BRICK_STOP_FAIL,
+ "Unable to stop brick: %s:%s",
+ src_brickinfo->hostname,
+ src_brickinfo->path);
goto out;
}
}
diff --git a/xlators/mgmt/glusterd/src/glusterd-snapshot.c b/xlators/mgmt/glusterd/src/glusterd-snapshot.c
index 78a4652cf41..31f4d95f63d 100644
--- a/xlators/mgmt/glusterd/src/glusterd-snapshot.c
+++ b/xlators/mgmt/glusterd/src/glusterd-snapshot.c
@@ -2753,8 +2753,6 @@ glusterd_do_lvm_snapshot_remove (glusterd_volinfo_t *snap_vol,
GLUSTERD_GET_BRICK_PIDFILE (pidfile, snap_vol, brickinfo, priv);
if (gf_is_service_running (pidfile, &pid)) {
- int send_attach_req (xlator_t *this, struct rpc_clnt *rpc,
- char *path, int op);
(void) send_attach_req (this, brickinfo->rpc,
brickinfo->path,
GLUSTERD_BRICK_TERMINATE);
diff --git a/xlators/mgmt/glusterd/src/glusterd-utils.c b/xlators/mgmt/glusterd/src/glusterd-utils.c
index 0a4aea24d85..a66e04934db 100644
--- a/xlators/mgmt/glusterd/src/glusterd-utils.c
+++ b/xlators/mgmt/glusterd/src/glusterd-utils.c
@@ -96,8 +96,8 @@
int
send_attach_req (xlator_t *this, struct rpc_clnt *rpc, char *path, int op);
-static gf_boolean_t
-is_brick_mx_enabled ()
+gf_boolean_t
+is_brick_mx_enabled (void)
{
char *value = NULL;
int ret = 0;
@@ -1329,7 +1329,8 @@ out:
int
glusterd_validate_and_create_brickpath (glusterd_brickinfo_t *brickinfo,
uuid_t volume_id, char **op_errstr,
- gf_boolean_t is_force)
+ gf_boolean_t is_force,
+ gf_boolean_t ignore_partition)
{
int ret = -1;
char parentdir[PATH_MAX] = {0,};
@@ -1403,8 +1404,14 @@ glusterd_validate_and_create_brickpath (glusterd_brickinfo_t *brickinfo,
" Or use 'force' at the end of the command if"
" you want to override this behavior.",
brickinfo->hostname, brickinfo->path);
- ret = -1;
- goto out;
+
+ /* If --wignore-partition flag is used, ignore warnings
+ * related to bricks being on root partition when 'force'
+ * is not used */
+ if (!ignore_partition) {
+ ret = -1;
+ goto out;
+ }
}
}
@@ -2074,13 +2081,15 @@ glusterd_brick_disconnect (glusterd_brickinfo_t *brickinfo)
}
int32_t
-glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
- glusterd_brickinfo_t *brickinfo,
+glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
+ glusterd_brickinfo_t *brickinfo,
gf_boolean_t del_brick)
{
xlator_t *this = NULL;
- int ret = 0;
+ glusterd_conf_t *conf = NULL;
+ int ret = -1;
char *op_errstr = NULL;
+ char pidfile[PATH_MAX] = {0,};
GF_ASSERT (volinfo);
GF_ASSERT (brickinfo);
@@ -2088,6 +2097,11 @@ glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
this = THIS;
GF_ASSERT (this);
+ conf = this->private;
+ GF_VALIDATE_OR_GOTO (this->name, conf, out);
+
+ ret = 0;
+
if (del_brick)
cds_list_del_init (&brickinfo->brick_list);
@@ -2102,10 +2116,17 @@ glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
* an actual signal instead.
*/
if (is_brick_mx_enabled ()) {
+ gf_msg_debug (this->name, 0, "About to send detach "
+ "request for brick %s:%s",
+ brickinfo->hostname, brickinfo->path);
+
(void) send_attach_req (this, brickinfo->rpc,
brickinfo->path,
GLUSTERD_BRICK_TERMINATE);
} else {
+ gf_msg_debug (this->name, 0, "About to stop glusterfsd"
+ " for brick %s:%s", brickinfo->hostname,
+ brickinfo->path);
(void) glusterd_brick_terminate (volinfo, brickinfo,
NULL, 0, &op_errstr);
if (op_errstr) {
@@ -2119,6 +2140,10 @@ glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
if (del_brick)
glusterd_delete_brick (volinfo, brickinfo);
+ GLUSTERD_GET_BRICK_PIDFILE (pidfile, volinfo, brickinfo, conf);
+ gf_msg_debug (this->name, 0, "Unlinking pidfile %s", pidfile);
+ (void) sys_unlink (pidfile);
+out:
return ret;
}
@@ -6563,9 +6588,6 @@ glusterd_brick_stop (glusterd_volinfo_t *volinfo,
goto out;
}
- gf_msg_debug (this->name, 0, "About to stop glusterfs"
- " for brick %s:%s", brickinfo->hostname,
- brickinfo->path);
ret = glusterd_volume_stop_glusterfs (volinfo, brickinfo, del_brick);
if (ret) {
gf_msg (this->name, GF_LOG_CRITICAL, 0,
diff --git a/xlators/mgmt/glusterd/src/glusterd-utils.h b/xlators/mgmt/glusterd/src/glusterd-utils.h
index d002ee45b6c..d694a312a41 100644
--- a/xlators/mgmt/glusterd/src/glusterd-utils.h
+++ b/xlators/mgmt/glusterd/src/glusterd-utils.h
@@ -80,6 +80,8 @@ typedef struct glusterd_dict_ctx_ {
char *prefix;
} glusterd_dict_ctx_t;
+gf_boolean_t is_brick_mx_enabled (void);
+
int
glusterd_compare_lines (const void *a, const void *b);
@@ -179,10 +181,13 @@ glusterd_volume_start_glusterfs (glusterd_volinfo_t *volinfo,
gf_boolean_t wait);
int32_t
-glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
- glusterd_brickinfo_t *brickinfo,
+glusterd_volume_stop_glusterfs (glusterd_volinfo_t *volinfo,
+ glusterd_brickinfo_t *brickinfo,
gf_boolean_t del_brick);
+int
+send_attach_req (xlator_t *this, struct rpc_clnt *rpc, char *path, int op);
+
glusterd_volinfo_t *
glusterd_volinfo_ref (glusterd_volinfo_t *volinfo);
@@ -310,7 +315,8 @@ glusterd_check_and_set_brick_xattr (char *host, char *path, uuid_t uuid,
int
glusterd_validate_and_create_brickpath (glusterd_brickinfo_t *brickinfo,
uuid_t volume_id, char **op_errstr,
- gf_boolean_t is_force);
+ gf_boolean_t is_force,
+ gf_boolean_t ignore_partition);
int
glusterd_sm_tr_log_transition_add (glusterd_sm_tr_log_t *log,
int old_state, int new_state,
diff --git a/xlators/mgmt/glusterd/src/glusterd-volume-ops.c b/xlators/mgmt/glusterd/src/glusterd-volume-ops.c
index 4738dc589d5..95056d501a3 100644
--- a/xlators/mgmt/glusterd/src/glusterd-volume-ops.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volume-ops.c
@@ -1316,7 +1316,7 @@ glusterd_op_stage_create_volume (dict_t *dict, char **op_errstr,
#endif
ret = glusterd_validate_and_create_brickpath (brick_info,
volume_uuid, op_errstr,
- is_force);
+ is_force, _gf_false);
if (ret)
goto out;
generated by cgit (git 2.34.1) at 2026-05-03 21:12:00 +0000