rpms/glusterfs
5
0
Fork 0
Code Issues Pull Requests Releases Activity
cd32b8a252
glusterfs/0084-cluster-afr-Fix-for-arbiter-becoming-source.patch
Milind Changire cf62f1947f autobuild v3.12.2-2 
Resolves: bz#1264911 bz#1277924 bz#1286820 bz#1360331 bz#1401969
Resolves: bz#1410719 bz#1419438 bz#1426042 bz#1444820 bz#1459101
Resolves: bz#1464150 bz#1464350 bz#1466122 bz#1466129 bz#1467903
Resolves: bz#1468972 bz#1476876 bz#1484446 bz#1492591 bz#1498391
Resolves: bz#1498730 bz#1499865 bz#1500704 bz#1501345 bz#1505570
Resolves: bz#1507361 bz#1507394 bz#1509102 bz#1509191 bz#1509810
Resolves: bz#1509833 bz#1511766 bz#1512470 bz#1512496 bz#1512963
Resolves: bz#1515051 bz#1519076 bz#1519740 bz#1534253 bz#1534530
Signed-off-by: Milind Changire <mchangir@redhat.com>
2018-01-17 02:21:37 -05:00

238 lines
8.9 KiB
Diff
Raw Blame History

From 5579f616c2c21a2a2cd2ef70b58149df85550db7 Mon Sep 17 00:00:00 2001
From: karthik-us <ksubrahm@redhat.com>
Date: Mon, 27 Nov 2017 12:51:16 +0530
Subject: [PATCH 084/128] cluster/afr: Fix for arbiter becoming source
Problem:
When eager-lock is on, and two writes happen in parallel on a FD
we were observing the following behaviour:
- First write fails on one data brick
- Since the post-op is not yet happened, the inode refresh will get
both the data bricks as readable and set it in the inode context
- In flight split brain check see both the data bricks as readable
and allows the second write
- Second write fails on the other data brick
- Now the post-op happens and marks both the data bricks as bad and
arbiter will become source for healing
Fix:
Adding one more variable called write_suvol in inode context and it
will have the in memory representation of the writable subvols. Inode
refresh will not update this value and its lifetime is pre-op through
unlock in the afr transaction. Initially the pre-op will set this
value same as read_subvol in inode context and then in the in flight
split brain check we will use this value instead of read_subvol.
After all the checks we will update the value of this and set the
read_subvol same as this to avoid having incorrect value in that.
Upstream patch: https://review.gluster.org/#/c/18049/
> Change-Id: I2ef6904524ab91af861d59690974bbc529ab1af3
> BUG: 1482064
> Signed-off-by: karthik-us <ksubrahm@redhat.com>
Change-Id: I91cd21e378a7ae3757c2209fcb91a613d73e09ee
BUG: 1401969
Signed-off-by: karthik-us <ksubrahm@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/124292
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
xlators/cluster/afr/src/afr-common.c | 76 ++++++++++++++++++++++++++++++-
xlators/cluster/afr/src/afr-lk-common.c | 18 ++++++--
xlators/cluster/afr/src/afr-transaction.c | 4 ++
xlators/cluster/afr/src/afr.h | 10 ++++
4 files changed, 102 insertions(+), 6 deletions(-)
diff --git a/xlators/cluster/afr/src/afr-common.c b/xlators/cluster/afr/src/afr-common.c
index 9c96056..a8ba5a0 100644
--- a/xlators/cluster/afr/src/afr-common.c
+++ b/xlators/cluster/afr/src/afr-common.c
@@ -149,6 +149,7 @@ __afr_inode_ctx_get (xlator_t *this, inode_t *inode, afr_inode_ctx_t **ctx)
}
tmp_ctx->spb_choice = -1;
tmp_ctx->read_subvol = 0;
+ tmp_ctx->write_subvol = 0;
} else {
tmp_ctx = (afr_inode_ctx_t *) ctx_int;
}
@@ -216,7 +217,7 @@ __afr_set_in_flight_sb_status (xlator_t *this, afr_local_t *local,
if (ret < 0)
return ret;
- val = ctx->read_subvol;
+ val = ctx->write_subvol;
metadatamap_old = metadatamap = (val & 0x000000000000ffff);
datamap_old = datamap = (val & 0x00000000ffff0000) >> 16;
@@ -276,6 +277,7 @@ __afr_set_in_flight_sb_status (xlator_t *this, afr_local_t *local,
(((uint64_t) datamap) << 16) |
(((uint64_t) event) << 32);
+ ctx->write_subvol = val;
ctx->read_subvol = val;
return ret;
@@ -6421,3 +6423,75 @@ afr_serialize_xattrs_with_delimiter (call_frame_t *frame, xlator_t *this,
out:
return ret;
}
+
+int
+afr_write_subvol_set (call_frame_t *frame, xlator_t *this)
+{
+ afr_local_t *local = NULL;
+ afr_inode_ctx_t *ctx = NULL;
+ uint64_t val = 0;
+ uint64_t val1 = 0;
+ int ret = -1;
+
+ local = frame->local;
+ LOCK(&local->inode->lock);
+ {
+ ret = __afr_inode_ctx_get (this, local->inode, &ctx);
+ if (ret < 0) {
+ gf_msg (this->name, GF_LOG_ERROR, 0,
+ AFR_MSG_DICT_GET_FAILED,
+ "ERROR GETTING INODE CTX");
+ UNLOCK(&local->inode->lock);
+ return ret;
+ }
+
+ val = ctx->write_subvol;
+ /*
+ * We need to set the value of write_subvol to read_subvol in 2
+ * cases:
+ * 1. Initially when the value is 0. i.e., it's the first lock
+ * request.
+ * 2. If it's a metadata transaction. If metadata transactions
+ * comes in between data transactions and we have a brick
+ * disconnect, the next metadata transaction won't get the
+ * latest value of readables, since we do resetting of
+ * write_subvol in unlock code path only if it's a data
+ * transaction. To handle those scenarios we need to set the
+ * value of write_subvol to read_subvol in case of metadata
+ * transactions.
+ */
+ if (val == 0 ||
+ local->transaction.type == AFR_METADATA_TRANSACTION) {
+ val1 = ctx->read_subvol;
+ ctx->write_subvol = val1;
+ }
+ }
+ UNLOCK (&local->inode->lock);
+
+ return 0;
+}
+
+int
+afr_write_subvol_reset (call_frame_t *frame, xlator_t *this)
+{
+ afr_local_t *local = NULL;
+ afr_inode_ctx_t *ctx = NULL;
+ int ret = -1;
+
+ local = frame->local;
+ LOCK(&local->inode->lock);
+ {
+ ret = __afr_inode_ctx_get (this, local->inode, &ctx);
+ if (ret < 0) {
+ gf_msg (this->name, GF_LOG_ERROR, 0,
+ AFR_MSG_DICT_GET_FAILED,
+ "ERROR GETTING INODE CTX");
+ UNLOCK(&local->inode->lock);
+ return ret;
+ }
+ ctx->write_subvol = 0;
+ }
+ UNLOCK(&local->inode->lock);
+
+ return 0;
+}
diff --git a/xlators/cluster/afr/src/afr-lk-common.c b/xlators/cluster/afr/src/afr-lk-common.c
index 1f2a117..c17f60f 100644
--- a/xlators/cluster/afr/src/afr-lk-common.c
+++ b/xlators/cluster/afr/src/afr-lk-common.c
@@ -613,12 +613,16 @@ static int32_t
afr_unlock_common_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
int32_t op_ret, int32_t op_errno, dict_t *xdata)
{
- afr_local_t *local = NULL;
- afr_internal_lock_t *int_lock = NULL;
- int call_count = 0;
+ afr_local_t *local = NULL;
+ afr_internal_lock_t *int_lock = NULL;
+ afr_fd_ctx_t *fd_ctx = NULL;
+ afr_private_t *priv = NULL;
+ int call_count = 0;
+ int ret = 0;
local = frame->local;
int_lock = &local->internal_lock;
+ priv = this->private;
LOCK (&frame->lock);
{
@@ -629,11 +633,15 @@ afr_unlock_common_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
if (call_count == 0) {
gf_msg_trace (this->name, 0,
"All internal locks unlocked");
-
+ if (local->fd) {
+ fd_ctx = afr_fd_ctx_get (local->fd, this);
+ if (0 == AFR_COUNT (fd_ctx->lock_acquired, priv->child_count))
+ ret = afr_write_subvol_reset (frame, this);
+ }
int_lock->lock_cbk (frame, this);
}
- return 0;
+ return ret;
}
void
diff --git a/xlators/cluster/afr/src/afr-transaction.c b/xlators/cluster/afr/src/afr-transaction.c
index 35621d9..91c4f78 100644
--- a/xlators/cluster/afr/src/afr-transaction.c
+++ b/xlators/cluster/afr/src/afr-transaction.c
@@ -1791,6 +1791,10 @@ afr_changelog_pre_op (call_frame_t *frame, xlator_t *this)
if (pre_nop)
goto next;
+ ret = afr_write_subvol_set (frame, this);
+ if (ret)
+ goto err;
+
if (!local->pre_op_compat) {
dict_copy (xdata_req, local->xdata_req);
goto next;
diff --git a/xlators/cluster/afr/src/afr.h b/xlators/cluster/afr/src/afr.h
index c4ceb66..672d053 100644
--- a/xlators/cluster/afr/src/afr.h
+++ b/xlators/cluster/afr/src/afr.h
@@ -837,6 +837,7 @@ typedef struct _afr_local {
typedef struct _afr_inode_ctx {
uint64_t read_subvol;
+ uint64_t write_subvol;
int spb_choice;
gf_timer_t *timer;
gf_boolean_t need_refresh;
@@ -1262,4 +1263,13 @@ int
afr_serialize_xattrs_with_delimiter (call_frame_t *frame, xlator_t *this,
char *buf, const char *default_str,
int32_t *serz_len, char delimiter);
+
+int
+__afr_inode_ctx_get (xlator_t *this, inode_t *inode, afr_inode_ctx_t **ctx);
+
+int
+afr_write_subvol_set (call_frame_t *frame, xlator_t *this);
+
+int
+afr_write_subvol_reset (call_frame_t *frame, xlator_t *this);
#endif /* __AFR_H__ */
--
1.8.3.1
Reference in New Issue View Git Blame Copy Permalink