rpms/glusterfs
5
0
Fork 0
Code Issues Pull Requests Releases Activity

autobuild v3.12.2-9

Browse Source 
Resolves: bz#1546717 bz#1557551 bz#1558948 bz#1561999 bz#1563804
Resolves: bz#1565015 bz#1565119 bz#1565399 bz#1565577 bz#1567100
Resolves: bz#1567899 bz#1568374 bz#1568969 bz#1569490 bz#1570514
Resolves: bz#1570541 bz#1570582 bz#1571645 bz#1572087 bz#1572585
Resolves: bz#1575895
Signed-off-by: Milind Changire <mchangir@redhat.com>
This commit is contained in:
Milind Changire 2018-05-09 08:42:05 -04:00
parent c211c8d97e
commit 312c9bafe1
25 changed files with 2981 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

117
0237-cluster-dht-Fix-dht_rename-lock-order.patch Normal file
View File

@ -0,0 +1,117 @@
From 67c89f4fc9a80c9ed4a7d78df57c1ca76f4adc3d Mon Sep 17 00:00:00 2001
From: N Balachandran <nbalacha@redhat.com>
Date: Tue, 17 Apr 2018 15:37:05 +0530
Subject: [PATCH 237/260] cluster/dht: Fix dht_rename lock order
Fixed dht_order_rename_lock to use the same inodelk ordering
as that of the dht selfheal locks (dictionary order of
lock subvolumes).
upstream: https://review.gluster.org/#/c/19886/
> Change-Id: Ia3f8353b33ea2fd3bc1ba7e8e777dda6c1d33e0d
> fixes: bz#1568348
> Signed-off-by: N Balachandran <nbalacha@redhat.com>
Change-Id: I09022705f5b77af0f50a2bab4579d4d3cb902155
BUG: 1565119
Signed-off-by: N Balachandran <nbalacha@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136451
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/cluster/dht/src/dht-rename.c | 65 ++++++++++++++++++++++++++----------
1 file changed, 47 insertions(+), 18 deletions(-)
diff --git a/xlators/cluster/dht/src/dht-rename.c b/xlators/cluster/dht/src/dht-rename.c
index 3068499..ca6b5f4 100644
--- a/xlators/cluster/dht/src/dht-rename.c
+++ b/xlators/cluster/dht/src/dht-rename.c
@@ -490,37 +490,66 @@ err:
return 0;
}
+
+/*
+ * If the hashed subvolumes of both source and dst are the different,
+ * lock in dictionary order of hashed subvol->name. This is important
+ * in case the parent directory is the same for both src and dst to
+ * prevent inodelk deadlocks when racing with a fix-layout op on the parent.
+ *
+ * If the hashed subvols are the same, use the gfid/name to determine
+ * the order of taking locks to prevent entrylk deadlocks when the parent
+ * dirs are the same.
+ *
+ */
static void
dht_order_rename_lock (call_frame_t *frame, loc_t **loc, xlator_t **subvol)
{
- dht_local_t *local = NULL;
- char src[GF_UUID_BNAME_BUF_SIZE] = {0};
- char dst[GF_UUID_BNAME_BUF_SIZE] = {0};
+ int ret = 0;
+ dht_local_t *local = NULL;
+ char src[GF_UUID_BNAME_BUF_SIZE] = {0};
+ char dst[GF_UUID_BNAME_BUF_SIZE] = {0};
+
local = frame->local;
- if (local->loc.pargfid)
- uuid_utoa_r (local->loc.pargfid, src);
- else if (local->loc.parent)
- uuid_utoa_r (local->loc.parent->gfid, src);
+ if (local->src_hashed->name == local->dst_hashed->name) {
+ ret = 0;
+ } else {
+ ret = strcmp (local->src_hashed->name, local->dst_hashed->name);
+ }
- strcat (src, local->loc.name);
+ if (ret == 0) {
- if (local->loc2.pargfid)
- uuid_utoa_r (local->loc2.pargfid, dst);
- else if (local->loc2.parent)
- uuid_utoa_r (local->loc2.parent->gfid, dst);
+ /* hashed subvols are the same for src and dst */
+ /* Entrylks need to be ordered*/
+ if (local->loc.pargfid)
+ uuid_utoa_r (local->loc.pargfid, src);
+ else if (local->loc.parent)
+ uuid_utoa_r (local->loc.parent->gfid, src);
- strcat (dst, local->loc2.name);
+ strcat (src, local->loc.name);
- if (strcmp(src, dst) > 0) {
- local->current = &local->lock[1];
- *loc = &local->loc2;
- *subvol = local->dst_hashed;
- } else {
+ if (local->loc2.pargfid)
+ uuid_utoa_r (local->loc2.pargfid, dst);
+ else if (local->loc2.parent)
+ uuid_utoa_r (local->loc2.parent->gfid, dst);
+
+ strcat (dst, local->loc2.name);
+ ret = strcmp (src, dst);
+ }
+
+ if (ret <= 0) {
+ /*inodelk in dictionary order of hashed subvol names*/
+ /*entrylk in dictionary order of gfid/basename */
local->current = &local->lock[0];
*loc = &local->loc;
*subvol = local->src_hashed;
+
+ } else {
+ local->current = &local->lock[1];
+ *loc = &local->loc2;
+ *subvol = local->dst_hashed;
}
return;
--
1.8.3.1

39
0238-quota-Build-is-failed-due-to-access-rpc-refcount-in-.patch Normal file
View File

@ -0,0 +1,39 @@
From 0b61a3b9c948a55da0df51a64483dcb8bc3954ba Mon Sep 17 00:00:00 2001
From: moagrawa <moagrawa@redhat.com>
Date: Mon, 23 Apr 2018 15:12:19 +0530
Subject: [PATCH 238/260] quota: Build is failed due to access rpc->refcount in
wrong way
Problem: Build is failed on centos due to access rpc-refcount
in wrong way in quota.c
Solution: Update quota.c to access rpc->refcount
BUG: 1570582
Change-Id: I881617259f73a7b1430e0ea80adf11168356d41e
Signed-off-by: moagrawa <moagrawa@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136505
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
xlators/features/quota/src/quota.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/xlators/features/quota/src/quota.c b/xlators/features/quota/src/quota.c
index c4817bc..3d68ffa 100644
--- a/xlators/features/quota/src/quota.c
+++ b/xlators/features/quota/src/quota.c
@@ -5259,7 +5259,9 @@ fini (xlator_t *this)
priv->rpc_clnt = NULL;
this->private = NULL;
if (rpc) {
- cnt = GF_ATOMIC_GET (rpc->refcount);
+ pthread_mutex_lock (&rpc->lock);
+ cnt = rpc->refcount;
+ pthread_mutex_unlock (&rpc->lock);
for (i = 0; i < cnt; i++)
rpc_clnt_unref (rpc);
}
--
1.8.3.1

54
0239-geo-rep-Fix-syncing-of-symlink.patch Normal file
View File

@ -0,0 +1,54 @@
From 80087936cf49a08848ee054179e4044ef42cd29e Mon Sep 17 00:00:00 2001
From: Kotresh HR <khiremat@redhat.com>
Date: Fri, 13 Apr 2018 10:52:14 -0400
Subject: [PATCH 239/260] geo-rep: Fix syncing of symlink
Problem:
If symlink is created on master pointing
to current directory (e.g symlink -> ".") with
non root uid or gid, geo-rep worker crashes
with ENOTSUP.
Cause:
Geo-rep creates the symlink on slave and
fixes the uid and gid using chown cmd.
os.chown dereferences the symlink which is
pointing to ".gfid" which is not supported.
Note that geo-rep operates on aux-gfid-mount
(e.g. "/mnt/.gfid/<gfid-of-symlink-file>").
Solution:
The uid or gid change is acutally on symlink
file. So use os.lchown, i.e, don't deference.
Upstream Reference
> BUG: 1567209
> Change-Id: I63575fc589d71f987bef1d350c030987738c78ad
> Patch: https://review.gluster.org/19872
BUG: 1565399
Change-Id: Ib4613719ac735a4e2856bc0655351f69f2467dac
Signed-off-by: Kotresh HR <khiremat@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136820
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
geo-replication/syncdaemon/resource.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/geo-replication/syncdaemon/resource.py b/geo-replication/syncdaemon/resource.py
index 4b2a266..d6618c1 100644
--- a/geo-replication/syncdaemon/resource.py
+++ b/geo-replication/syncdaemon/resource.py
@@ -868,7 +868,7 @@ class Server(object):
# UID:0 and GID:0, and then call chown to set UID/GID
if uid != 0 or gid != 0:
path = os.path.join(pfx, gfid)
- cmd_ret = errno_wrap(os.chown, [path, uid, gid], [ENOENT],
+ cmd_ret = errno_wrap(os.lchown, [path, uid, gid], [ENOENT],
[ESTALE, EINVAL])
collect_failure(e, cmd_ret)
--
1.8.3.1

72
0240-shared-storage-Prevent-mounting-shared-storage-from-.patch Normal file
View File

@ -0,0 +1,72 @@
From 73b0dc833716484cba04b46fe0d645975dba44db Mon Sep 17 00:00:00 2001
From: Mohammed Rafi KC <rkavunga@redhat.com>
Date: Mon, 26 Mar 2018 20:27:34 +0530
Subject: [PATCH 240/260] shared storage: Prevent mounting shared storage from
non-trusted client
gluster shared storage is a volume used for internal storage for
various features including ganesha, geo-rep, snapshot.
So this volume should not be exposed to the client, as it is
a special volume for internal use.
This fix wont't generate non trusted volfile for shared storage volume.
backport of https://review.gluster.org/#/c/19920/
>Change-Id: I8ffe30ae99ec05196d75466210b84db311611a4c
>updates: bz#1570432
>Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Change-Id: Ic540b983bcc53a783fda7ca7a283a9ab48d9eeb7
BUG: 1568969
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136708
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-volgen.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/xlators/mgmt/glusterd/src/glusterd-volgen.c b/xlators/mgmt/glusterd/src/glusterd-volgen.c
index 0e287b6..1c43f24 100644
--- a/xlators/mgmt/glusterd/src/glusterd-volgen.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volgen.c
@@ -5828,6 +5828,7 @@ generate_client_volfiles (glusterd_volinfo_t *volinfo,
int i = 0;
int ret = -1;
char filepath[PATH_MAX] = {0,};
+ char *volname = NULL;
char *types[] = {NULL, NULL, NULL};
dict_t *dict = NULL;
xlator_t *this = NULL;
@@ -5835,6 +5836,26 @@ generate_client_volfiles (glusterd_volinfo_t *volinfo,
this = THIS;
+ volname = volinfo->is_snap_volume ?
+ volinfo->parent_volname : volinfo->volname;
+
+
+ if (volname && !strcmp (volname, GLUSTER_SHARED_STORAGE) &&
+ client_type != GF_CLIENT_TRUSTED) {
+ /*
+ * shared storage volume cannot be mounted from non trusted
+ * nodes. So we are not creating volfiles for non-trusted
+ * clients for shared volumes as well as snapshot of shared
+ * volumes.
+ */
+
+ ret = 0;
+ gf_msg_debug ("glusterd", 0, "Skipping the non-trusted volfile"
+ "creation for shared storage volume. Volume %s",
+ volname);
+ goto out;
+ }
+
enumerate_transport_reqs (volinfo->transport_type, types);
dict = dict_new ();
if (!dict)
--
1.8.3.1

283
0241-server-auth-add-option-for-strict-authentication.patch Normal file
View File

@ -0,0 +1,283 @@
From bceb9586b96cbf3e0f1747affa6f56cfe29ac140 Mon Sep 17 00:00:00 2001
From: Mohammed Rafi KC <rkavunga@redhat.com>
Date: Mon, 2 Apr 2018 12:20:47 +0530
Subject: [PATCH 241/260] server/auth: add option for strict authentication
When this option is enabled, we will check for a matching
username and password, if not found then the connection will
be rejected. This also does a checksum validation of volfile
The option is invalid when SSL/TLS is in use, at which point
the SSL/TLS certificate user name is used to validate and
hence authorize the right user. This expects TLS allow rules
to be setup correctly rather than the default *.
This option is not settable, as a result this cannot be enabled
for volumes using the CLI. This is used with the shared storage
volume, to restrict access to the same in non-SSL/TLS environments
to the gluster peers only.
Tested:
./tests/bugs/protocol/bug-1321578.t
./tests/features/ssl-authz.t
- Ran tests on volumes with and without strict auth
checking (as brick vol file needed to be edited to test,
or rather to enable the option)
- Ran tests on volumes to ensure existing mounts are
disconnected when we enable strict checking
backport of upstream https://review.gluster.org/#/c/19921/
>Change-Id: I2ac4f0cfa5b59cc789cc5a265358389b04556b59
>fixes: bz#1570432
>Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
>Signed-off-by: ShyamsundarR <srangana@redhat.com>
Change-Id: I7fd3194b91ba22d57b851a2a042ff16b9a616a5a
BUG: 1568969
Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136709
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-volgen.c | 16 +++++++-
xlators/protocol/auth/login/src/login.c | 51 ++++++++++++++++++++++----
xlators/protocol/server/src/authenticate.h | 4 +-
xlators/protocol/server/src/server-handshake.c | 2 +-
xlators/protocol/server/src/server.c | 18 +++++++++
xlators/protocol/server/src/server.h | 2 +
6 files changed, 81 insertions(+), 12 deletions(-)
diff --git a/xlators/mgmt/glusterd/src/glusterd-volgen.c b/xlators/mgmt/glusterd/src/glusterd-volgen.c
index 1c43f24..3926bd3 100644
--- a/xlators/mgmt/glusterd/src/glusterd-volgen.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volgen.c
@@ -2341,6 +2341,7 @@ brick_graph_add_server (volgen_graph_t *graph, glusterd_volinfo_t *volinfo,
char *password = NULL;
char key[1024] = {0};
char *ssl_user = NULL;
+ char *volname = NULL;
char *address_family_data = NULL;
if (!graph || !volinfo || !set_dict || !brickinfo)
@@ -2416,6 +2417,19 @@ brick_graph_add_server (volgen_graph_t *graph, glusterd_volinfo_t *volinfo,
if (ret)
return -1;
+ volname = volinfo->is_snap_volume ?
+ volinfo->parent_volname : volinfo->volname;
+
+
+ if (volname && !strcmp (volname, GLUSTER_SHARED_STORAGE)) {
+ memset (key, 0, sizeof (key));
+ snprintf (key, sizeof (key), "strict-auth-accept");
+
+ ret = xlator_set_option (xl, key, "true");
+ if (ret)
+ return -1;
+ }
+
if (dict_get_str (volinfo->dict, "auth.ssl-allow", &ssl_user) == 0) {
memset (key, 0, sizeof (key));
snprintf (key, sizeof (key), "auth.login.%s.ssl-allow",
@@ -5841,7 +5855,7 @@ generate_client_volfiles (glusterd_volinfo_t *volinfo,
if (volname && !strcmp (volname, GLUSTER_SHARED_STORAGE) &&
- client_type != GF_CLIENT_TRUSTED) {
+ client_type != GF_CLIENT_TRUSTED) {
/*
* shared storage volume cannot be mounted from non trusted
* nodes. So we are not creating volfiles for non-trusted
diff --git a/xlators/protocol/auth/login/src/login.c b/xlators/protocol/auth/login/src/login.c
index e799dd2..da10d0b 100644
--- a/xlators/protocol/auth/login/src/login.c
+++ b/xlators/protocol/auth/login/src/login.c
@@ -11,6 +11,16 @@
#include <fnmatch.h>
#include "authenticate.h"
+/* Note on strict_auth
+ * - Strict auth kicks in when authentication is using the username, password
+ * in the volfile to login
+ * - If enabled, auth is rejected if the username and password is not matched
+ * or is not present
+ * - When using SSL names, this is automatically strict, and allows only those
+ * names that are present in the allow list, IOW strict auth checking has no
+ * implication when using SSL names
+*/
+
auth_result_t gf_auth (dict_t *input_params, dict_t *config_params)
{
auth_result_t result = AUTH_DONT_CARE;
@@ -27,6 +37,7 @@ auth_result_t gf_auth (dict_t *input_params, dict_t *config_params)
char *tmp = NULL;
char *username_cpy = NULL;
gf_boolean_t using_ssl = _gf_false;
+ gf_boolean_t strict_auth = _gf_false;
username_data = dict_get (input_params, "ssl-name");
if (username_data) {
@@ -35,16 +46,39 @@ auth_result_t gf_auth (dict_t *input_params, dict_t *config_params)
using_ssl = _gf_true;
}
else {
+ ret = dict_get_str_boolean (config_params, "strict-auth-accept",
+ _gf_false);
+ if (ret == -1)
+ strict_auth = _gf_false;
+ else
+ strict_auth = ret;
+
username_data = dict_get (input_params, "username");
if (!username_data) {
- gf_log ("auth/login", GF_LOG_DEBUG,
- "username not found, returning DONT-CARE");
+ if (strict_auth) {
+ gf_log ("auth/login", GF_LOG_DEBUG,
+ "username not found, strict auth"
+ " configured returning REJECT");
+ result = AUTH_REJECT;
+ } else {
+ gf_log ("auth/login", GF_LOG_DEBUG,
+ "username not found, returning"
+ " DONT-CARE");
+ }
goto out;
}
password_data = dict_get (input_params, "password");
if (!password_data) {
- gf_log ("auth/login", GF_LOG_WARNING,
- "password not found, returning DONT-CARE");
+ if (strict_auth) {
+ gf_log ("auth/login", GF_LOG_DEBUG,
+ "password not found, strict auth"
+ " configured returning REJECT");
+ result = AUTH_REJECT;
+ } else {
+ gf_log ("auth/login", GF_LOG_WARNING,
+ "password not found, returning"
+ " DONT-CARE");
+ }
goto out;
}
password = data_to_str (password_data);
@@ -62,9 +96,10 @@ auth_result_t gf_auth (dict_t *input_params, dict_t *config_params)
ret = gf_asprintf (&searchstr, "auth.login.%s.%s", brick_name,
using_ssl ? "ssl-allow" : "allow");
if (-1 == ret) {
- gf_log ("auth/login", GF_LOG_WARNING,
+ gf_log ("auth/login", GF_LOG_ERROR,
"asprintf failed while setting search string, "
- "returning DONT-CARE");
+ "returning REJECT");
+ result = AUTH_REJECT;
goto out;
}
@@ -92,8 +127,10 @@ auth_result_t gf_auth (dict_t *input_params, dict_t *config_params)
* ssl-allow=* case as well) authorization is effectively
* disabled, though authentication and encryption are still
* active.
+ *
+ * Read NOTE on strict_auth above.
*/
- if (using_ssl) {
+ if (using_ssl || strict_auth) {
result = AUTH_REJECT;
}
username_cpy = gf_strdup (allow_user->data);
diff --git a/xlators/protocol/server/src/authenticate.h b/xlators/protocol/server/src/authenticate.h
index 3f80231..5f92183 100644
--- a/xlators/protocol/server/src/authenticate.h
+++ b/xlators/protocol/server/src/authenticate.h
@@ -37,10 +37,8 @@ typedef struct {
volume_opt_list_t *vol_opt;
} auth_handle_t;
-auth_result_t gf_authenticate (dict_t *input_params,
- dict_t *config_params,
- dict_t *auth_modules);
int32_t gf_auth_init (xlator_t *xl, dict_t *auth_modules);
void gf_auth_fini (dict_t *auth_modules);
+auth_result_t gf_authenticate (dict_t *, dict_t *, dict_t *);
#endif /* _AUTHENTICATE_H */
diff --git a/xlators/protocol/server/src/server-handshake.c b/xlators/protocol/server/src/server-handshake.c
index cbc93a3..fc63f2c 100644
--- a/xlators/protocol/server/src/server-handshake.c
+++ b/xlators/protocol/server/src/server-handshake.c
@@ -747,7 +747,7 @@ server_setvolume (rpcsvc_request_t *req)
ret = dict_get_str (params, "volfile-key",
&volfile_key);
if (ret)
- gf_msg_debug (this->name, 0, "failed to set "
+ gf_msg_debug (this->name, 0, "failed to get "
"'volfile-key'");
ret = _validate_volfile_checksum (this, volfile_key,
diff --git a/xlators/protocol/server/src/server.c b/xlators/protocol/server/src/server.c
index eed4295..6f20a06 100644
--- a/xlators/protocol/server/src/server.c
+++ b/xlators/protocol/server/src/server.c
@@ -943,6 +943,10 @@ do_rpc:
goto out;
}
+ GF_OPTION_RECONF ("strict-auth-accept", conf->strict_auth_enabled,
+ options, bool, out);
+
+
GF_OPTION_RECONF ("dynamic-auth", conf->dync_auth, options,
bool, out);
@@ -1183,6 +1187,14 @@ init (xlator_t *this)
"Failed to initialize group cache.");
goto out;
}
+
+ ret = dict_get_str_boolean (this->options, "strict-auth-accept",
+ _gf_false);
+ if (ret == -1)
+ conf->strict_auth_enabled = _gf_false;
+ else
+ conf->strict_auth_enabled = ret;
+
ret = dict_get_str_boolean (this->options, "dynamic-auth",
_gf_true);
if (ret == -1)
@@ -1840,5 +1852,11 @@ struct volume_options options[] = {
"transport connection immediately in response to "
"*.allow | *.reject volume set options."
},
+ { .key = {"strict-auth-accept"},
+ .type = GF_OPTION_TYPE_BOOL,
+ .default_value = "off",
+ .description = "strict-auth-accept reject connection with out"
+ "a valid username and password."
+ },
{ .key = {NULL} },
};
diff --git a/xlators/protocol/server/src/server.h b/xlators/protocol/server/src/server.h
index af987aa..691c75b 100644
--- a/xlators/protocol/server/src/server.h
+++ b/xlators/protocol/server/src/server.h
@@ -24,6 +24,7 @@
#include "client_t.h"
#include "gidcache.h"
#include "defaults.h"
+#include "authenticate.h"
#define DEFAULT_BLOCK_SIZE 4194304 /* 4MB */
#define DEFAULT_VOLUME_FILE_PATH CONFDIR "/glusterfs.vol"
@@ -109,6 +110,7 @@ struct server_conf {
* tweeked */
struct _child_status *child_status;
gf_lock_t itable_lock;
+ gf_boolean_t strict_auth_enabled;
};
typedef struct server_conf server_conf_t;
--
1.8.3.1

31
0242-feature-changelog-remove-unused-variable.patch Normal file
View File

@ -0,0 +1,31 @@
From 05942267fc5d33be457cf07f95d9dae2341e1050 Mon Sep 17 00:00:00 2001
From: Sunil Kumar Acharya <sheggodu@redhat.com>
Date: Wed, 25 Apr 2018 14:56:55 +0530
Subject: [PATCH 242/260] feature/changelog: remove unused variable
Fix removes unsed variable in notify().
BUG: 1571645
Change-Id: I60fb61ae4ed3e80e2ecf93935477db0fcb2cd156
Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136830
Tested-by: RHGS Build Bot <nigelb@redhat.com>
---
xlators/features/changelog/src/changelog.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/xlators/features/changelog/src/changelog.c b/xlators/features/changelog/src/changelog.c
index 121ab04..7e36172 100644
--- a/xlators/features/changelog/src/changelog.c
+++ b/xlators/features/changelog/src/changelog.c
@@ -2102,7 +2102,6 @@ notify (xlator_t *this, int event, void *data, ...)
int ret = 0;
int ret1 = 0;
struct list_head queue = {0, };
- int i = 0;
INIT_LIST_HEAD (&queue);
--
1.8.3.1

86
0243-timer-Fix-possible-race-during-cleanup.patch Normal file
View File

@ -0,0 +1,86 @@
From 4623ebe9a69a5284565b4ad253084a1236478988 Mon Sep 17 00:00:00 2001
From: Soumya Koduri <skoduri@redhat.com>
Date: Fri, 3 Nov 2017 15:41:34 +0530
Subject: [PATCH 243/260] timer: Fix possible race during cleanup
As mentioned in bug1509189, there is a possible race
between gf_timer_cancel(), gf_timer_proc() and
gf_timer_registry_destroy() leading to use_after_free.
Problem:
1) gf_timer_proc() is called, locks reg, and gets an event.
It unlocks reg, and calls the callback.
2) Meanwhile gf_timer_registry_destroy() is called, and removes
reg from ctx, and joins on gf_timer_proc().
3) gf_timer_call_cancel() is called on the event being
processed. It cannot find reg (since it's been removed from reg),
so it frees event.
4) the callback returns into gf_timer_proc(), and it tries to free
event, but it's already free, so double free.
Solution:
The fix is to bail out in gf_timer_cancel() when registry
is not found. The logic behind this is that, gf_timer_cancel()
is called only on any existing event. That means there was a valid
registry earlier while creating that event. And the only reason
we cannot find that registry now is that it must have got set to
NULL when context cleanup is started.
Since gf_timer_proc() takes care of releasing all the remaining
events active on that registry, it seems safe to bail out
in gf_timer_cancel().
>Change-Id: Ia9b088533141c3bb335eff2fe06b52d1575bb34f
>BUG: 1509189
>Reported-by: Daniel Gryniewicz <dang@redhat.com>
>Signed-off-by: Soumya Koduri <skoduri@redhat.com>
Upstream Link: https://review.gluster.org/#/c/18652/
BUG: 1568374
Change-Id: Ia9b088533141c3bb335eff2fe06b52d1575bb34f
Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/135896
Tested-by: RHGS Build Bot <nigelb@redhat.com>
---
libglusterfs/src/timer.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/libglusterfs/src/timer.c b/libglusterfs/src/timer.c
index 34dfd35..d6f008d 100644
--- a/libglusterfs/src/timer.c
+++ b/libglusterfs/src/timer.c
@@ -83,6 +83,13 @@ gf_timer_call_cancel (glusterfs_ctx_t *ctx,
return 0;
}
+ if (ctx->cleanup_started) {
+ gf_msg_callingfn ("timer", GF_LOG_INFO, 0,
+ LG_MSG_CTX_CLEANUP_STARTED,
+ "ctx cleanup started");
+ return 0;
+ }
+
LOCK (&ctx->lock);
{
reg = ctx->timer;
@@ -90,9 +97,11 @@ gf_timer_call_cancel (glusterfs_ctx_t *ctx,
UNLOCK (&ctx->lock);
if (!reg) {
- gf_msg ("timer", GF_LOG_ERROR, 0, LG_MSG_INIT_TIMER_FAILED,
- "!reg");
- GF_FREE (event);
+ /* This can happen when cleanup may have just started and
+ * gf_timer_registry_destroy() sets ctx->timer to NULL.
+ * Just bail out as success as gf_timer_proc() takes
+ * care of cleaning up the events.
+ */
return 0;
}
--
1.8.3.1

93
0244-common-ha-All-statd-related-files-need-to-be-owned-b.patch Normal file
View File

@ -0,0 +1,93 @@
From 9a458a8a0bdd88ff44f80b6b5a4c5fbaa970500f Mon Sep 17 00:00:00 2001
From: Soumya Koduri <skoduri@redhat.com>
Date: Thu, 19 Jan 2017 15:01:12 +0530
Subject: [PATCH 244/260] common-ha: All statd related files need to be owned
by rpcuser
Statd service is started as rpcuser by default. Hence the
files/directories needed by it under '/var/lib/nfs' should be
owned by the same user.
Note: This change is not in mainline as the cluster-bits
are being moved to storehaug project -
http://review.gluster.org/#/c/16349/
http://review.gluster.org/#/c/16333/
Label: BACKPORT FROM UPSTREAM 3.10
Upstream Reference :
> Change-Id: I89fd06aa9700c5ce60026ac825da7c154d9f48fd
> BUG: 1414665
> Signed-off-by: Soumya Koduri <skoduri@redhat.com>
> Reviewed-on: http://review.gluster.org/16433
> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com>
> Smoke: Gluster Build System <jenkins@build.gluster.org>
> Tested-by: Kaleb KEITHLEY <kkeithle@redhat.com>
> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Change-Id: I89fd06aa9700c5ce60026ac825da7c154d9f48fd
BUG: 1565015
Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137254
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
extras/ganesha/scripts/ganesha-ha.sh | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/extras/ganesha/scripts/ganesha-ha.sh b/extras/ganesha/scripts/ganesha-ha.sh
index 4459068..5cdafad 100644
--- a/extras/ganesha/scripts/ganesha-ha.sh
+++ b/extras/ganesha/scripts/ganesha-ha.sh
@@ -756,9 +756,11 @@ setup_state_volume()
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd
fi
if [ ! -e ${mnt}/nfs-ganesha/${dirname}/nfs/state ]; then
touch ${mnt}/nfs-ganesha/${dirname}/nfs/state
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/state
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/ganesha/v4recov ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/ganesha/v4recov
@@ -768,9 +770,11 @@ setup_state_volume()
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak
fi
if [ ! -e ${mnt}/nfs-ganesha/${dirname}/nfs/statd/state ]; then
touch ${mnt}/nfs-ganesha/${dirname}/nfs/statd/state
@@ -830,9 +834,11 @@ addnode_state_volume()
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd
fi
if [ ! -e ${mnt}/nfs-ganesha/${dirname}/nfs/state ]; then
touch ${mnt}/nfs-ganesha/${dirname}/nfs/state
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/state
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/ganesha/v4recov ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/ganesha/v4recov
@@ -842,9 +848,11 @@ addnode_state_volume()
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm
fi
if [ ! -d ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak ]; then
mkdir ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak
+ chown rpcuser:rpcuser ${mnt}/nfs-ganesha/${dirname}/nfs/statd/sm.bak
fi
if [ ! -e ${mnt}/nfs-ganesha/${dirname}/nfs/statd/state ]; then
touch ${mnt}/nfs-ganesha/${dirname}/nfs/statd/state
--
1.8.3.1

56
0245-build-make-RHGS-version-available-for-server.patch Normal file
View File

@ -0,0 +1,56 @@
From 01b371af8dc6f95c00facd5215ab4c939a86f1db Mon Sep 17 00:00:00 2001
From: Milind Changire <mchangir@redhat.com>
Date: Mon, 23 Apr 2018 13:16:30 +0530
Subject: [PATCH 245/260] build: make RHGS version available for server
Make /usr/share/glusterfs/release available for gluserfs-server package.
This file contains the RHGS release number for the release.
Label: DOWNSTREAM ONLY
Change-Id: I7485f77cfb8ca7f0f8363a20124900ae9ae8a528
BUG: 1570514
Signed-off-by: Milind Changire <mchangir@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137139
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
---
glusterfs.spec.in | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/glusterfs.spec.in b/glusterfs.spec.in
index 64e7e29..36b465a 100644
--- a/glusterfs.spec.in
+++ b/glusterfs.spec.in
@@ -865,6 +865,10 @@ find ./tests ./run-tests.sh -type f | cpio -pd %{buildroot}%{_prefix}/share/glus
install -p -m 0744 -D extras/command-completion/gluster.bash \
%{buildroot}%{_sysconfdir}/bash_completion.d/gluster
+%if ( 0%{?_build_server} )
+echo "RHGS 3.4.0" > %{buildroot}%{_datadir}/glusterfs/release
+%endif
+
%clean
rm -rf %{buildroot}
@@ -1566,6 +1570,7 @@ exit 0
# Extra utility script
%dir %{_libexecdir}/glusterfs
+ %{_datadir}/glusterfs/release
%dir %{_datadir}/glusterfs/scripts
%{_datadir}/glusterfs/scripts/stop-all-gluster-processes.sh
%if ( 0%{?_with_systemd:1} )
@@ -2155,6 +2160,9 @@ fi
%endif
%changelog
+* Mon Apr 23 2018 Milind Changire <mchangir@redhat.com>
+- make RHGS release number available in /usr/share/glusterfs/release (#1570514)
+
* Wed Apr 18 2018 Atin Mukherjee <amukherj@redhat.com>
- Revert SELinux hooks (#1565962)
--
1.8.3.1

45
0246-glusterd-Fix-for-memory-leak-in-get-state-detail.patch Normal file
View File

@ -0,0 +1,45 @@
From 4ff96e11ea80d9a0d9dec2031c9cdd99d2afdc61 Mon Sep 17 00:00:00 2001
From: Sanju Rakonde <srakonde@redhat.com>
Date: Mon, 30 Apr 2018 11:12:34 +0530
Subject: [PATCH 246/260] glusterd: Fix for memory leak in get-state detail
>Fixes: bz#1573066
>Change-Id: I76fe3bdde7351736b32eb3d6c4cc5f8f276257ed
>Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
upstream patch: https://review.gluster.org/#/c/19948/
BUG: 1567899
Change-Id: I76fe3bdde7351736b32eb3d6c4cc5f8f276257ed
Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137469
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-handler.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/xlators/mgmt/glusterd/src/glusterd-handler.c b/xlators/mgmt/glusterd/src/glusterd-handler.c
index 30adb99..962c87e 100644
--- a/xlators/mgmt/glusterd/src/glusterd-handler.c
+++ b/xlators/mgmt/glusterd/src/glusterd-handler.c
@@ -5409,8 +5409,15 @@ out:
if (pending_node)
GF_FREE (pending_node);
- if (brick_req)
+ if (brick_req) {
+ if (brick_req->input.input_val)
+ GF_FREE (brick_req->input.input_val);
GF_FREE (brick_req);
+ }
+ if (args.dict)
+ dict_unref (args.dict);
+ if (args.errstr)
+ GF_FREE (args.errstr);
return ret;
}
--
1.8.3.1

74
0247-protocol-server-unwind-as-per-op-version.patch Normal file
View File

@ -0,0 +1,74 @@
From 3c8ff0fceb008dca51cebfc9e808f6268548661a Mon Sep 17 00:00:00 2001
From: Ashish Pandey <aspandey@redhat.com>
Date: Wed, 25 Apr 2018 16:48:56 +0530
Subject: [PATCH 247/260] protocol/server : unwind as per op version
>Change-Id: Id6717640ac14881b490e512c4682e45ffffa7f5b
upstream patch -
https://review.gluster.org/#/c/19938/
Change-Id: I43034c2974108c434bd3e4c553ff416339ece870
BUG: 1558948
Signed-off-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137625
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
libglusterfs/src/client_t.h | 1 +
xlators/features/locks/src/posix.c | 8 +++++++-
xlators/protocol/server/src/server-handshake.c | 5 +++--
3 files changed, 11 insertions(+), 3 deletions(-)
diff --git a/libglusterfs/src/client_t.h b/libglusterfs/src/client_t.h
index 530c0a3..088508e 100644
--- a/libglusterfs/src/client_t.h
+++ b/libglusterfs/src/client_t.h
@@ -45,6 +45,7 @@ typedef struct _client {
char *subdir_mount;
inode_t *subdir_inode;
uuid_t subdir_gfid;
+ int32_t opversion;
} client_t;
#define GF_CLIENTCTX_INITIAL_SIZE 8
diff --git a/xlators/features/locks/src/posix.c b/xlators/features/locks/src/posix.c
index a158227..ef4bebf 100644
--- a/xlators/features/locks/src/posix.c
+++ b/xlators/features/locks/src/posix.c
@@ -1563,7 +1563,13 @@ int
pl_flush_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
int32_t op_ret, int32_t op_errno, dict_t *xdata)
{
- PL_STACK_UNWIND (flush, xdata, frame, op_ret, op_errno, xdata);
+
+ if (frame->root->client &&
+ (frame->root->client->opversion < GD_OP_VERSION_3_12_0)) {
+ STACK_UNWIND_STRICT (flush, frame, op_ret, op_errno, xdata);
+ } else {
+ PL_STACK_UNWIND (flush, xdata, frame, op_ret, op_errno, xdata);
+ }
return 0;
}
diff --git a/xlators/protocol/server/src/server-handshake.c b/xlators/protocol/server/src/server-handshake.c
index fc63f2c..a59bcda 100644
--- a/xlators/protocol/server/src/server-handshake.c
+++ b/xlators/protocol/server/src/server-handshake.c
@@ -778,11 +778,12 @@ server_setvolume (rpcsvc_request_t *req)
}
ret = dict_get_uint32 (params, "opversion", &opversion);
- if (ret)
+ if (ret) {
gf_msg (this->name, GF_LOG_INFO, 0,
PS_MSG_CLIENT_OPVERSION_GET_FAILED,
"Failed to get client opversion");
-
+ }
+ client->opversion = opversion;
/* Assign op-version value to the client */
pthread_mutex_lock (&conf->mutex);
list_for_each_entry (xprt, &conf->xprt_list, list) {
--
1.8.3.1

1042
0248-performance-md-cache-purge-cache-on-ENOENT-ESTALE-er.patch Normal file
View File

File diff suppressed because it is too large Load Diff

46
0249-cluster-dht-unwind-if-dht_selfheal_dir_mkdir-returns.patch Normal file
View File

@ -0,0 +1,46 @@
From 0e9d29c1ae1ba6954299489012ba69466709af0f Mon Sep 17 00:00:00 2001
From: Raghavendra G <rgowdapp@redhat.com>
Date: Thu, 3 May 2018 09:34:39 +0530
Subject: [PATCH 249/260] cluster/dht: unwind if dht_selfheal_dir_mkdir returns
an error
If dht_selfheal_dir_mkdir returns an error, cbk passed to
dht_selfheal_directory is not invoked. So, Current codepath leaves an
unwound frame resulting in a hung fop forever.
> Change-Id: I422308b8a34a074301ca46b029ffe676f5e0f66c
> fixes: bz#1574305
> Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
upstream patch: https://review.gluster.org/19953/
Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
Change-Id: I3ea6510d936eca6198147704d93580bd733eaeea
BUG: 1561999
Reviewed-on: https://code.engineering.redhat.com/gerrit/137725
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Nithya Balachandran <nbalacha@redhat.com>
---
xlators/cluster/dht/src/dht-selfheal.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/xlators/cluster/dht/src/dht-selfheal.c b/xlators/cluster/dht/src/dht-selfheal.c
index 7b192d3..53c59a2 100644
--- a/xlators/cluster/dht/src/dht-selfheal.c
+++ b/xlators/cluster/dht/src/dht-selfheal.c
@@ -2336,7 +2336,11 @@ dht_selfheal_directory (call_frame_t *frame, dht_selfheal_dir_cbk_t dir_cbk,
local->heal_layout = _gf_false;
}
- dht_selfheal_dir_mkdir (frame, loc, layout, 0);
+ ret = dht_selfheal_dir_mkdir (frame, loc, layout, 0);
+ if (ret < 0) {
+ ret = 0;
+ goto sorry_no_fix;
+ }
return 0;
--
1.8.3.1

69
0250-cluster-dht-act-as-passthrough-for-renames-on-single.patch Normal file
View File

@ -0,0 +1,69 @@
From 4eef56dc01a9ab11d56f30dd29e2eb65c82cc01d Mon Sep 17 00:00:00 2001
From: Raghavendra G <rgowdapp@redhat.com>
Date: Wed, 4 Apr 2018 10:03:18 +0530
Subject: [PATCH 250/260] cluster/dht: act as passthrough for renames on single
child DHT
Various synchronization present in dht_rename while handling
directories and files is necessary only if we have more than only one
child.
> Change-Id: Ie21ad419125504ca2f391b1ae2e5c1d166fee247
> fixes: bz#1563511
> Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
upstream patch: https://review.gluster.org/19817/
Change-Id: Ie21ad419125504ca2f391b1ae2e5c1d166fee247
BUG: 1572087
fixes: bz#1572087
Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137257
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Nithya Balachandran <nbalacha@redhat.com>
---
xlators/cluster/dht/src/dht-rename.c | 22 +++++++++++++++-------
1 file changed, 15 insertions(+), 7 deletions(-)
diff --git a/xlators/cluster/dht/src/dht-rename.c b/xlators/cluster/dht/src/dht-rename.c
index ca6b5f4..5a72c1f 100644
--- a/xlators/cluster/dht/src/dht-rename.c
+++ b/xlators/cluster/dht/src/dht-rename.c
@@ -1690,13 +1690,14 @@ int
dht_rename (call_frame_t *frame, xlator_t *this,
loc_t *oldloc, loc_t *newloc, dict_t *xdata)
{
- xlator_t *src_cached = NULL;
- xlator_t *src_hashed = NULL;
- xlator_t *dst_cached = NULL;
- xlator_t *dst_hashed = NULL;
- int op_errno = -1;
- int ret = -1;
- dht_local_t *local = NULL;
+ xlator_t *src_cached = NULL;
+ xlator_t *src_hashed = NULL;
+ xlator_t *dst_cached = NULL;
+ xlator_t *dst_hashed = NULL;
+ int op_errno = -1;
+ int ret = -1;
+ dht_local_t *local = NULL;
+ dht_conf_t *conf = NULL;
char gfid[GF_UUID_BUF_SIZE] = {0};
VALIDATE_OR_GOTO (frame, err);
@@ -1704,6 +1705,13 @@ dht_rename (call_frame_t *frame, xlator_t *this,
VALIDATE_OR_GOTO (oldloc, err);
VALIDATE_OR_GOTO (newloc, err);
+ conf = this->private;
+
+ if (conf->subvolume_cnt == 1) {
+ default_rename (frame, this, oldloc, newloc, xdata);
+ return 0;
+ }
+
gf_uuid_unparse(oldloc->inode->gfid, gfid);
src_hashed = dht_subvol_get_hashed (this, oldloc);
--
1.8.3.1

52
0251-dht-gf_defrag_settle_hash-should-ignore-ENOENT-and-E.patch Normal file
View File

@ -0,0 +1,52 @@
From c5401605f80c2630e39b283f10b7da92ff777863 Mon Sep 17 00:00:00 2001
From: Susant Palai <spalai@redhat.com>
Date: Fri, 27 Apr 2018 16:40:02 +0530
Subject: [PATCH 251/260] dht: gf_defrag_settle_hash should ignore ENOENT and
ESTALE error
Problem: A directory deletion can happen just before gf_defrag_settle_hash
which internally does a setxattr operation on a directory.
Solution: Ignore ENOENT and ESTALE errors
> Fixes: bz#1572581
> Change-Id: I2f91809f3b5e02976c4c3a5a596406a8b2f8f6f2
> Signed-off-by: Susant Palai <spalai@redhat.com>
(cherry picked from commit e2fda098112803bf651c4795952376cb8c1ad204)
upstream patch: https://review.gluster.org/#/c/19945/
BUG: 1572585
Change-Id: I2f91809f3b5e02976c4c3a5a596406a8b2f8f6f2
Signed-off-by: Susant Palai <spalai@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137527
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Nithya Balachandran <nbalacha@redhat.com>
---
xlators/cluster/dht/src/dht-rebalance.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/xlators/cluster/dht/src/dht-rebalance.c b/xlators/cluster/dht/src/dht-rebalance.c
index eee00b8..51af11c 100644
--- a/xlators/cluster/dht/src/dht-rebalance.c
+++ b/xlators/cluster/dht/src/dht-rebalance.c
@@ -3493,8 +3493,15 @@ gf_defrag_settle_hash (xlator_t *this, gf_defrag_info_t *defrag,
ret = syncop_setxattr (this, loc, fix_layout, 0, NULL, NULL);
if (ret) {
- gf_log (this->name, GF_LOG_ERROR,
+ gf_msg (this->name, GF_LOG_ERROR, -ret,
+ DHT_MSG_LAYOUT_FIX_FAILED,
"fix layout on %s failed", loc->path);
+
+ if (-ret == ENOENT || -ret == ESTALE) {
+ /* Dir most likely is deleted */
+ return 0;
+ }
+
return -1;
}
--
1.8.3.1

60
0252-glusterd-ganesha-Skip-non-ganesha-nodes-properly-for.patch Normal file
View File

@ -0,0 +1,60 @@
From 90d143d0011c15a5a334946cb628b5415825c786 Mon Sep 17 00:00:00 2001
From: Jiffin Tony Thottan <jthottan@redhat.com>
Date: Mon, 30 Apr 2018 12:35:01 +0530
Subject: [PATCH 252/260] glusterd/ganesha : Skip non-ganesha nodes properly
for ganesha HA set up
Label: BACKPORT FROM UPSTREAM 3.10
Upstream reference:
>Patch unlink https://review.gluster.org/#/c/19949/
>Change-Id: Iff7bc3ead43e97847219c5a5cc8b967bf0967903
>BUG: 1573078
>Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
Change-Id: Iff7bc3ead43e97847219c5a5cc8b967bf0967903
BUG: 1570541
Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137768
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Kaleb Keithley <kkeithle@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-ganesha.c | 21 ++++++++++++---------
1 file changed, 12 insertions(+), 9 deletions(-)
diff --git a/xlators/mgmt/glusterd/src/glusterd-ganesha.c b/xlators/mgmt/glusterd/src/glusterd-ganesha.c
index b130d5e..ae7485f 100644
--- a/xlators/mgmt/glusterd/src/glusterd-ganesha.c
+++ b/xlators/mgmt/glusterd/src/glusterd-ganesha.c
@@ -834,15 +834,18 @@ pre_setup (gf_boolean_t run_setup, char **op_errstr)
{
int ret = 0;
- if (check_host_list()) {
- ret = setup_cluster(run_setup);
- if (ret == -1)
- gf_asprintf (op_errstr, "Failed to set up HA "
- "config for NFS-Ganesha. "
- "Please check the log file for details");
- } else
- ret = -1;
-
+ if (run_setup) {
+ if (!check_host_list()) {
+ gf_asprintf (op_errstr, "Running nfs-ganesha setup command "
+ "from node which is not part of ganesha cluster");
+ return -1;
+ }
+ }
+ ret = setup_cluster(run_setup);
+ if (ret == -1)
+ gf_asprintf (op_errstr, "Failed to set up HA "
+ "config for NFS-Ganesha. "
+ "Please check the log file for details");
return ret;
}
--
1.8.3.1

59
0253-geo-rep-Fix-upgrade-issue.patch Normal file
View File

@ -0,0 +1,59 @@
From 2d1c394b7ea9913c69445d865fb8762ecb4917ce Mon Sep 17 00:00:00 2001
From: Kotresh HR <khiremat@redhat.com>
Date: Mon, 7 May 2018 01:23:46 -0400
Subject: [PATCH 253/260] geo-rep: Fix upgrade issue
Cause and Analysis:
The last synced changelog for entry operations is
marked in current version to avoid re-processing
of already processed entry operations in a batch
during crash/restart of geo-rep. This was not
present in previous versoins.
The marker is maintained in the dictionary with the
key 'last_synced_entry' and dictionary is persisted
into status file. So upgrading to current version in
which the marker is present was failing with KeyError.
Solution:
Load the dictionary with default keys first which
contains all the keys including latest ones and then
load the values from status file instead of doing
otherwise.
Upstream reference:
> fixes: bz#1575490
> Patch: https://review.gluster.org/19969
BUG: 1569490
Change-Id: Ic654e6f9a3c97f616761f1362f890352a2186fb4
Signed-off-by: Kotresh HR <khiremat@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137909
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
geo-replication/syncdaemon/gsyncdstatus.py | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/geo-replication/syncdaemon/gsyncdstatus.py b/geo-replication/syncdaemon/gsyncdstatus.py
index 38ca92c..909c669 100644
--- a/geo-replication/syncdaemon/gsyncdstatus.py
+++ b/geo-replication/syncdaemon/gsyncdstatus.py
@@ -152,11 +152,12 @@ class GeorepStatus(object):
**kwargs)
def _update(self, mergerfunc):
+ data = self.default_values
with LockedOpen(self.filename, 'r+') as f:
try:
- data = json.load(f)
+ data.update(json.load(f))
except ValueError:
- data = self.default_values
+ pass
data = mergerfunc(data)
# If Data is not changed by merger func
--
1.8.3.1

99
0254-posix-Avoid-changelog-retries-for-geo-rep.patch Normal file
View File

@ -0,0 +1,99 @@
From 2de9ac218241648f1ec86b73aac728d713862a30 Mon Sep 17 00:00:00 2001
From: moagrawa <moagrawa@redhat.com>
Date: Fri, 4 May 2018 16:17:42 +0530
Subject: [PATCH 254/260] posix: Avoid changelog retries for geo-rep
Problem: georep is slowdown to migrate directory
from master volume to slave volume due to lot
of changelog retries
Solution: Update the condition in posix_getxattr to
ignore MDS_INTERNAL_XATTR as it(posix) ignored
other internal xattrs
> BUG: 1571069
> Change-Id: I4d91ec73e5b1ca1cb3ecf0825ab9f49e261da70e
> fixes: bz#1571069
> Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
> (cherry picked from commit f46047ea0f36084f4456887673552a01b926382d)
> (Reviewed on: https://review.gluster.org/#/c/19930/)
BUG: 1565577
Change-Id: I85b2e8ed4adcbe34ee5a03acd068693978e9cfe4
Signed-off-by: moagrawa <moagrawa@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/137779
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Kotresh Hiremath Ravishankar <khiremat@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/storage/posix/src/posix.c | 33 +++++++++++++++++++++++++++++++++
1 file changed, 33 insertions(+)
diff --git a/xlators/storage/posix/src/posix.c b/xlators/storage/posix/src/posix.c
index c680e3f..416d9e4 100644
--- a/xlators/storage/posix/src/posix.c
+++ b/xlators/storage/posix/src/posix.c
@@ -27,6 +27,7 @@
#include <sys/uio.h>
#include <unistd.h>
#include <ftw.h>
+#include <regex.h>
#ifndef GF_BSD_HOST_OS
#include <alloca.h>
@@ -4625,6 +4626,27 @@ out:
return ret;
}
+gf_boolean_t
+posix_is_mds_xattr (const char *name)
+{
+ regex_t regcmpl;
+ char *key = {"trusted.glusterfs.*.mds$"};
+ regmatch_t result[1] = {{0} };
+ gf_boolean_t status = _gf_false;
+
+ if (regcomp (&regcmpl, key, REG_EXTENDED)) {
+ goto out;
+ }
+ if (!regexec (&regcmpl, name, 1, result, 0)) {
+ status = _gf_true;
+ goto out;
+ }
+out:
+ regfree(&regcmpl);
+ return status;
+}
+
+
/**
* posix_getxattr - this function returns a dictionary with all the
* key:value pair present as xattr. used for
@@ -4680,6 +4702,13 @@ posix_getxattr (call_frame_t *frame, xlator_t *this,
goto out;
}
+ if (name && posix_is_mds_xattr (name)) {
+ op_ret = -1;
+ op_errno = ENOATTR;
+ goto out;
+ }
+
+
if (loc->inode && IA_ISDIR(loc->inode->ia_type) && name &&
ZR_FILE_CONTENT_REQUEST(name)) {
ret = posix_get_file_contents (this, loc->gfid, &name[15],
@@ -5049,6 +5078,10 @@ posix_getxattr (call_frame_t *frame, xlator_t *this,
goto ignore;
}
+ if (posix_is_mds_xattr (keybuffer)) {
+ goto ignore;
+ }
+
memset (value_buf, '\0', sizeof(value_buf));
have_val = _gf_false;
size = sys_lgetxattr (real_path, keybuffer, value_buf,
--
1.8.3.1

46
0255-glusterd-update-listen-backlog-value-to-1024.patch Normal file
View File

@ -0,0 +1,46 @@
From 8b2af4ee879261ea6e690c3c52d9acc5d0a92cfd Mon Sep 17 00:00:00 2001
From: Milind Changire <mchangir@redhat.com>
Date: Thu, 19 Apr 2018 11:18:58 +0530
Subject: [PATCH 255/260] glusterd: update listen-backlog value to 1024
Update default value of listen-backlog to 1024 to reflect the changes in
socket.c
This keeps the actual implementation in socket.c and the help text in
glusterd-volume-set.c consistent
mainline:
> Reviewed-on: https://review.gluster.org/19874
> fixes: bz#1564600
> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
(cherry picked from commit d9cf6d25a4719c4f6fb4d88325f08e49fca18e6b)
Change-Id: If04c9e0bb5afb55edcc7ca57bbc10922b85b7075
BUG: 1563804
Signed-off-by: Milind Changire <mchangir@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/136215
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Raghavendra Gowdappa <rgowdapp@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-volume-set.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/xlators/mgmt/glusterd/src/glusterd-volume-set.c b/xlators/mgmt/glusterd/src/glusterd-volume-set.c
index 9bc0933..b9da961 100644
--- a/xlators/mgmt/glusterd/src/glusterd-volume-set.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volume-set.c
@@ -2131,7 +2131,7 @@ struct volopt_map_entry glusterd_volopt_map[] = {
.description = "This option uses the value of backlog argument that "
"defines the maximum length to which the queue of "
"pending connections for socket fd may grow.",
- .value = "10",
+ .value = "1024",
},
/* Generic transport options */
--
1.8.3.1

57
0256-rpc-set-listen-backlog-to-high-value.patch Normal file
View File

@ -0,0 +1,57 @@
From 4448795af501355f3893aed05551ee62551e8438 Mon Sep 17 00:00:00 2001
From: Milind Changire <mchangir@redhat.com>
Date: Fri, 13 Apr 2018 14:51:32 +0530
Subject: [PATCH 256/260] rpc: set listen-backlog to high value
Problem:
On node reboot, when glusterd starts volumes rapidly, there's a flood of
connections from the bricks to glusterd and from the self-heal daemons
to the bricks. This causes SYN Flooding and dropped connections when the
listen-backlog is not enough to hold the pending connections to
compensate for the rate at which connections are accepted by the RPC
layer.
Solution:
Increase the listen-backlog value to 1024. This is a partial solution.
Part of the solution is to rearm the listener socket early for quicker
accept() of connections.
See commit 6964640a977cb10c0c95a94e03c229918fa6eca8 (change 19833)
mainline:
> Reviewed-on: https://review.gluster.org/19836
> Change-Id: I56781149919dd6e18cc43884234c6f60f2c93f88
> fixes: bz#1564600
> Signed-off-by: Milind Changire <mchangir@redhat.com>
Change-Id: I1d7ac940eb5410cdbf7508a24bfdfc052b35170b
BUG: 1563804
Signed-off-by: Milind Changire <mchangir@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/135557
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Raghavendra Gowdappa <rgowdapp@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
rpc/rpc-transport/socket/src/socket.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/rpc/rpc-transport/socket/src/socket.c b/rpc/rpc-transport/socket/src/socket.c
index 157b5b7..a6be911 100644
--- a/rpc/rpc-transport/socket/src/socket.c
+++ b/rpc/rpc-transport/socket/src/socket.c
@@ -4613,6 +4613,13 @@ struct volume_options options[] = {
.min = GF_MIN_SOCKET_WINDOW_SIZE,
.max = GF_MAX_SOCKET_WINDOW_SIZE
},
+ { .key = {"transport.listen-backlog"},
+ .type = GF_OPTION_TYPE_SIZET,
+ .description = "This option uses the value of backlog argument that "
+ "defines the maximum length to which the queue of "
+ "pending connections for socket fd may grow.",
+ .default_value = "1024",
+ },
{ .key = {"transport.tcp-user-timeout"},
.type = GF_OPTION_TYPE_INT,
.default_value = "0"
--
1.8.3.1

57
0257-rpc-rearm-listener-socket-early.patch Normal file
View File

@ -0,0 +1,57 @@
From 8879d72884396de4ae2ff41649de5f2d916d272b Mon Sep 17 00:00:00 2001
From: Milind Changire <mchangir@redhat.com>
Date: Fri, 13 Apr 2018 10:53:46 +0530
Subject: [PATCH 257/260] rpc: rearm listener socket early
Problem:
On node reboot, when glusterd starts volumes, a setup with a large
number of bricks might cause SYN Flooding and connections to be dropped
if the connections are not accepted quickly enough.
Solution:
accept() the connection and rearm the listener socket early to receive
more connection requests as soon as possible.
mainline:
> Reviewed-on: https://review.gluster.org/19833
> Change-Id: Ibed421e50284c3f7a8fcdb4de7ac86cf53d4b74e
> fixes: bz#1564600
> Signed-off-by: Milind Changire <mchangir@redhat.com>
Change-Id: I4bf45b8a12f738ed9b398246e22dfd29e6b1010c
BUG: 1563804
Signed-off-by: Milind Changire <mchangir@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/135513
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Raghavendra Gowdappa <rgowdapp@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
rpc/rpc-transport/socket/src/socket.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/rpc/rpc-transport/socket/src/socket.c b/rpc/rpc-transport/socket/src/socket.c
index a6be911..b98efdc 100644
--- a/rpc/rpc-transport/socket/src/socket.c
+++ b/rpc/rpc-transport/socket/src/socket.c
@@ -2746,6 +2746,9 @@ socket_server_event_handler (int fd, int idx, int gen, void *data,
if (poll_in) {
new_sock = accept (priv->sock, SA (&new_sockaddr), &addrlen);
+ if (ctx)
+ event_handled (ctx->event_pool, fd, idx, gen);
+
if (new_sock == -1) {
gf_log (this->name, GF_LOG_WARNING,
"accept on %d failed (%s)",
@@ -2968,8 +2971,6 @@ socket_server_event_handler (int fd, int idx, int gen, void *data,
}
}
out:
- event_handled (ctx->event_pool, fd, idx, gen);
-
if (cname && (cname != this->ssl_name)) {
GF_FREE(cname);
}
--
1.8.3.1

286
0258-cluster-dht-log-error-only-if-layout-healing-is-requ.patch Normal file
View File

@ -0,0 +1,286 @@
From 9ac423d71b9d9d8875ce2af29e2bcf5770b61d5a Mon Sep 17 00:00:00 2001
From: Raghavendra G <rgowdapp@redhat.com>
Date: Fri, 16 Mar 2018 12:16:43 +0530
Subject: [PATCH 258/260] cluster/dht: log error only if layout healing is
required
selfhealing of directory is invoked on two conditions:
1. no layout on disk or layout has some anomalies (holes/overlaps)
2. mds xattr is not set on the directory
When dht_selfheal_directory is called with a correct layout just to
set mds xattr, we see error msgs complaining about "not able to form
layout on directory", which is misleading as the layout is
correct. So, log this msg only if layout has anomalies.
>Change-Id: I4af25246fc3a2450c2426e9902d1a5b372eab125
>updates: bz#1543279
>BUG: 1543279
>Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
upstream patch: https://review.gluster.org/19727
BUG: 1567100
Change-Id: I169823ec3d4455c43f575e6c4ec0735aceb54e6b
Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/138149
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/cluster/dht/src/dht-common.h | 1 +
xlators/cluster/dht/src/dht-selfheal.c | 199 ++-------------------------------
2 files changed, 12 insertions(+), 188 deletions(-)
diff --git a/xlators/cluster/dht/src/dht-common.h b/xlators/cluster/dht/src/dht-common.h
index a785876..9d429af 100644
--- a/xlators/cluster/dht/src/dht-common.h
+++ b/xlators/cluster/dht/src/dht-common.h
@@ -312,6 +312,7 @@ struct dht_local {
uint32_t overlaps_cnt;
uint32_t down;
uint32_t misc;
+ uint32_t missing_cnt;
dht_selfheal_dir_cbk_t dir_cbk;
dht_selfheal_layout_t healer;
dht_need_heal_t should_heal;
diff --git a/xlators/cluster/dht/src/dht-selfheal.c b/xlators/cluster/dht/src/dht-selfheal.c
index 53c59a2..5812f12 100644
--- a/xlators/cluster/dht/src/dht-selfheal.c
+++ b/xlators/cluster/dht/src/dht-selfheal.c
@@ -359,7 +359,8 @@ dht_should_heal_layout (call_frame_t *frame, dht_layout_t **heal,
ret = dht_layout_anomalies (frame->this, &local->loc, *ondisk,
&local->selfheal.hole_cnt,
&local->selfheal.overlaps_cnt,
- NULL, &local->selfheal.down,
+ &local->selfheal.missing_cnt,
+ &local->selfheal.down,
&local->selfheal.misc, NULL);
if (ret < 0)
@@ -1015,115 +1016,6 @@ dht_layout_index_from_conf (dht_layout_t *layout, xlator_t *xlator)
return i;
}
-
-static int
-dht_selfheal_dir_xattr_for_nameless_lookup (call_frame_t *frame, loc_t *loc,
- dht_layout_t *layout)
-{
- dht_local_t *local = NULL;
- int missing_xattr = 0;
- int i = 0;
- xlator_t *this = NULL;
- dht_conf_t *conf = NULL;
- dht_layout_t *dummy = NULL;
- int j = 0;
-
- local = frame->local;
- this = frame->this;
- conf = this->private;
-
- for (i = 0; i < layout->cnt; i++) {
- if (layout->list[i].err != -1 || !layout->list[i].stop) {
- /* err != -1 would mean xattr present on the directory
- or the directory is non existent.
- !layout->list[i].stop would mean layout absent
- */
-
- continue;
- }
- missing_xattr++;
- }
-
- /* Also account for subvolumes with no-layout. Used for zero'ing out
- the layouts and for setting quota key's if present */
-
- /* Send where either the subvol is not part of layout,
- * or it is part of the layout but error is non-zero but error
- * is not equal to -1 or ENOENT.
- */
-
- for (i = 0; i < conf->subvolume_cnt; i++) {
- if (dht_is_subvol_part_of_layout (layout, conf->subvolumes[i])
- == _gf_false) {
- missing_xattr++;
- continue;
- }
-
- j = dht_layout_index_from_conf (layout, conf->subvolumes[i]);
-
- if ((j != -1) && (layout->list[j].err != -1) &&
- (layout->list[j].err != 0) &&
- (layout->list[j].err != ENOENT)) {
- missing_xattr++;
- }
-
- }
-
-
- gf_msg_trace (this->name, 0,
- "%d subvolumes missing xattr for %s",
- missing_xattr, loc->path);
-
- if (missing_xattr == 0) {
- dht_selfheal_dir_finish (frame, this, 0, 1);
- return 0;
- }
-
- local->call_cnt = missing_xattr;
-
- if (gf_log_get_loglevel () >= GF_LOG_DEBUG)
- dht_log_new_layout_for_dir_selfheal (this, loc, layout);
-
- for (i = 0; i < layout->cnt; i++) {
- if (layout->list[i].err != -1 || !layout->list[i].stop)
- continue;
-
- dht_selfheal_dir_xattr_persubvol (frame, loc, layout, i, NULL);
-
- if (--missing_xattr == 0)
- break;
- }
-
- dummy = dht_layout_new (this, 1);
- if (!dummy)
- goto out;
-
- for (i = 0; i < conf->subvolume_cnt && missing_xattr; i++) {
- if (dht_is_subvol_part_of_layout (layout, conf->subvolumes[i])
- == _gf_false) {
- dht_selfheal_dir_xattr_persubvol (frame, loc, dummy, 0,
- conf->subvolumes[i]);
- missing_xattr--;
- continue;
- }
-
- j = dht_layout_index_from_conf (layout, conf->subvolumes[i]);
-
- if ((j != -1) && (layout->list[j].err != -1) &&
- (layout->list[j].err != ENOENT) &&
- (layout->list[j].err != 0)) {
- dht_selfheal_dir_xattr_persubvol (frame, loc, dummy, 0,
- conf->subvolumes[i]);
- missing_xattr--;
- }
- }
-
- dht_layout_unref (this, dummy);
-out:
- return 0;
-
-}
-
int
dht_selfheal_dir_setattr_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
int op_ret, int op_errno, struct iatt *statpre,
@@ -2296,7 +2188,8 @@ dht_selfheal_directory (call_frame_t *frame, dht_selfheal_dir_cbk_t dir_cbk,
dht_layout_anomalies (this, loc, layout,
&local->selfheal.hole_cnt,
&local->selfheal.overlaps_cnt,
- NULL, &local->selfheal.down,
+ &local->selfheal.missing_cnt,
+ &local->selfheal.down,
&local->selfheal.misc, NULL);
down = local->selfheal.down;
@@ -2325,14 +2218,14 @@ dht_selfheal_directory (call_frame_t *frame, dht_selfheal_dir_cbk_t dir_cbk,
dht_layout_sort_volname (layout);
local->heal_layout = _gf_true;
- ret = dht_selfheal_dir_getafix (frame, loc, layout);
- if (ret == -1) {
- gf_msg (this->name, GF_LOG_INFO, 0,
- DHT_MSG_DIR_SELFHEAL_FAILED,
- "Directory selfheal failed: "
- "Unable to form layout for directory %s",
- loc->path);
+ /* Ignore return value as it can be inferred from result of
+ * dht_layout_anomalies
+ */
+ dht_selfheal_dir_getafix (frame, loc, layout);
+
+ if (!(local->selfheal.hole_cnt || local->selfheal.overlaps_cnt ||
+ local->selfheal.missing_cnt)) {
local->heal_layout = _gf_false;
}
@@ -2352,76 +2245,6 @@ sorry_no_fix:
}
int
-dht_selfheal_directory_for_nameless_lookup (call_frame_t *frame,
- dht_selfheal_dir_cbk_t dir_cbk,
- loc_t *loc, dht_layout_t *layout)
-{
- dht_local_t *local = NULL;
- uint32_t down = 0;
- uint32_t misc = 0;
- int ret = 0;
- xlator_t *this = NULL;
-
- local = frame->local;
- this = frame->this;
- dht_layout_anomalies (this, loc, layout,
- &local->selfheal.hole_cnt,
- &local->selfheal.overlaps_cnt,
- NULL, &local->selfheal.down,
- &local->selfheal.misc, NULL);
-
- down = local->selfheal.down;
- misc = local->selfheal.misc;
-
- local->selfheal.dir_cbk = dir_cbk;
- local->selfheal.layout = dht_layout_ref (this, layout);
-
- if (down) {
- gf_msg (this->name, GF_LOG_WARNING, 0,
- DHT_MSG_SUBVOL_DOWN_ERROR,
- "%d subvolumes down -- not fixing", down);
- ret = 0;
- goto sorry_no_fix;
- }
-
- if (misc) {
- gf_msg (this->name, GF_LOG_WARNING, 0,
- DHT_MSG_SUBVOL_ERROR,
- "%d subvolumes have unrecoverable errors", misc);
- ret = 0;
- goto sorry_no_fix;
- }
-
- dht_layout_sort_volname (layout);
- ret = dht_selfheal_dir_getafix (frame, loc, layout);
-
- if (ret == -1) {
- gf_msg (this->name, GF_LOG_WARNING, 0,
- DHT_MSG_LAYOUT_FORM_FAILED,
- "not able to form layout for the directory");
- goto sorry_no_fix;
- }
-
- ret = dht_selfheal_layout_lock (frame, layout, _gf_false,
- dht_selfheal_dir_xattr_for_nameless_lookup,
- dht_should_heal_layout);
-
- if (ret < 0) {
- goto sorry_no_fix;
- }
-
- return 0;
-
-sorry_no_fix:
- /* TODO: need to put appropriate local->op_errno */
- dht_selfheal_dir_finish (frame, this, ret, 1);
-
- return 0;
-
-
-}
-
-int
dht_selfheal_restore (call_frame_t *frame, dht_selfheal_dir_cbk_t dir_cbk,
loc_t *loc, dht_layout_t *layout)
{
--
1.8.3.1

78
0259-Quota-Turn-on-ssl-for-crawler-clients-if-needed.patch Normal file
View File

@ -0,0 +1,78 @@
From 0652899ce703542fc7fd676f3fd80748395d7094 Mon Sep 17 00:00:00 2001
From: Sanoj Unnikrishnan <sunnikri@redhat.com>
Date: Wed, 9 May 2018 13:43:48 +0530
Subject: [PATCH 259/260] Quota: Turn on ssl for crawler clients if needed
Problem: Quota uses per brick client generated by
glusterd_generate_client_per_brick_volfile to crawl the
individual bricks. These clients were not being configured
with ssl if volume has client.ssl turned on.
Solution: turn on client.ssl if the volume has client.ssl
option set to on.
Upstream Reference:
> Change-Id: Id3a13d5110c4376d734480c42da1ce6844cc8240
> fixes: bz#1575858
> Reviewed on: https://review.gluster.org/#/c/19978/
BUG: 1557551
Change-Id: Id3a13d5110c4376d734480c42da1ce6844cc8240
Signed-off-by: Sanoj Unnikrishnan <sunnikri@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/138158
Reviewed-by: Hari Gowtham Gopal <hgowtham@redhat.com>
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/mgmt/glusterd/src/glusterd-volgen.c | 21 ++++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/xlators/mgmt/glusterd/src/glusterd-volgen.c b/xlators/mgmt/glusterd/src/glusterd-volgen.c
index 3926bd3..3d8ad77 100644
--- a/xlators/mgmt/glusterd/src/glusterd-volgen.c
+++ b/xlators/mgmt/glusterd/src/glusterd-volgen.c
@@ -5726,6 +5726,8 @@ glusterd_generate_client_per_brick_volfile (glusterd_volinfo_t *volinfo)
dict_t *dict = NULL;
xlator_t *xl = NULL;
int ret = -1;
+ char *ssl_str = NULL;
+ gf_boolean_t ssl_bool = _gf_false;
dict = dict_new ();
if (!dict)
@@ -5733,7 +5735,22 @@ glusterd_generate_client_per_brick_volfile (glusterd_volinfo_t *volinfo)
ret = dict_set_uint32 (dict, "trusted-client", GF_CLIENT_TRUSTED);
if (ret)
- goto out;
+ goto free_dict;
+
+ if (dict_get_str(volinfo->dict, "client.ssl", &ssl_str) == 0) {
+ if (gf_string2boolean(ssl_str, &ssl_bool) == 0) {
+ if (ssl_bool) {
+ if (dict_set_dynstr_with_alloc(dict,
+ "client.ssl", "on") != 0) {
+ ret = -1;
+ goto free_dict;
+ }
+ }
+ } else {
+ ret = -1;
+ goto free_dict;
+ }
+ }
cds_list_for_each_entry (brick, &volinfo->bricks, brick_list) {
xl = volgen_graph_build_client (&graph, volinfo,
@@ -5760,6 +5777,8 @@ out:
if (ret)
volgen_graph_free (&graph);
+free_dict:
+
if (dict)
dict_unref (dict);
--
1.8.3.1

44
0260-dht-Avoid-dict-log-flooding-for-internal-MDS-xattr.patch Normal file
View File

@ -0,0 +1,44 @@
From 1eb5f1a725cc9afdd7b1a7b29cf446b9c60974c6 Mon Sep 17 00:00:00 2001
From: Mohit Agrawal <moagrawa@redhat.com>
Date: Tue, 8 May 2018 14:52:04 +0530
Subject: [PATCH 260/260] dht: Avoid dict log flooding for internal MDS xattr
Problem: Before populate MDS internal xattr first dht checks if MDS is
present in xattr or not.If xattr dictionary is NULL dict_get
log the message either dict or key is NULL
Solution: Before call dict_get check xattr, if it is NULL then no
need to call dict_get.
> BUG: 1575910
> Change-Id: I81604ec5945b85eba14b42f4583d06ec713028f4
> fixes: bz#1575910
> (cherry picked from commit ed5a09e4aea7f64f9a43698955e24285a936f0c0)
> (Upstream link https://review.gluster.org/#/c/19981/)
BUG: 1575895
Change-Id: I728a8068786f94d3793a91cdd5c64f813d7545ca
Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
Reviewed-on: https://code.engineering.redhat.com/gerrit/138205
Tested-by: RHGS Build Bot <nigelb@redhat.com>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
---
xlators/cluster/dht/src/dht-common.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/xlators/cluster/dht/src/dht-common.c b/xlators/cluster/dht/src/dht-common.c
index 3b8ba6d..a985481 100644
--- a/xlators/cluster/dht/src/dht-common.c
+++ b/xlators/cluster/dht/src/dht-common.c
@@ -855,7 +855,7 @@ dht_common_mark_mdsxattr (call_frame_t *frame, int *errst, int mark_during_fresh
and wind a setxattr call on hashed subvol to update
internal xattr
*/
- if (!dict_get (local->xattr, conf->mds_xattr_key)) {
+ if (!local->xattr || !dict_get (local->xattr, conf->mds_xattr_key)) {
/* It means no internal MDS xattr has been set yet
*/
/* Check the status of all subvol are up while call
--
1.8.3.1

37
glusterfs.spec
View File

@ -192,7 +192,7 @@ Release: 0.1%{?prereltag:.%{prereltag}}%{?dist}
%else
Name: glusterfs
Version: 3.12.2
Release: 8%{?dist}
Release: 9%{?dist}
%endif
License: GPLv2 or LGPLv3+
Group: System Environment/Base
@ -501,6 +501,30 @@ Patch0233: 0233-libglusterfs-fix-comparison-of-a-NULL-dict-with-a-no.patch
Patch0234: 0234-ec-Use-tiebreaker_inodelk-where-necessary.patch
Patch0235: 0235-cluster-syncop-Implement-tiebreaker-inodelk-entrylk.patch
Patch0236: 0236-cluster-syncop-Address-comments-in-3ad68df725ac32f83.patch
Patch0237: 0237-cluster-dht-Fix-dht_rename-lock-order.patch
Patch0238: 0238-quota-Build-is-failed-due-to-access-rpc-refcount-in-.patch
Patch0239: 0239-geo-rep-Fix-syncing-of-symlink.patch
Patch0240: 0240-shared-storage-Prevent-mounting-shared-storage-from-.patch
Patch0241: 0241-server-auth-add-option-for-strict-authentication.patch
Patch0242: 0242-feature-changelog-remove-unused-variable.patch
Patch0243: 0243-timer-Fix-possible-race-during-cleanup.patch
Patch0244: 0244-common-ha-All-statd-related-files-need-to-be-owned-b.patch
Patch0245: 0245-build-make-RHGS-version-available-for-server.patch
Patch0246: 0246-glusterd-Fix-for-memory-leak-in-get-state-detail.patch
Patch0247: 0247-protocol-server-unwind-as-per-op-version.patch
Patch0248: 0248-performance-md-cache-purge-cache-on-ENOENT-ESTALE-er.patch
Patch0249: 0249-cluster-dht-unwind-if-dht_selfheal_dir_mkdir-returns.patch
Patch0250: 0250-cluster-dht-act-as-passthrough-for-renames-on-single.patch
Patch0251: 0251-dht-gf_defrag_settle_hash-should-ignore-ENOENT-and-E.patch
Patch0252: 0252-glusterd-ganesha-Skip-non-ganesha-nodes-properly-for.patch
Patch0253: 0253-geo-rep-Fix-upgrade-issue.patch
Patch0254: 0254-posix-Avoid-changelog-retries-for-geo-rep.patch
Patch0255: 0255-glusterd-update-listen-backlog-value-to-1024.patch
Patch0256: 0256-rpc-set-listen-backlog-to-high-value.patch
Patch0257: 0257-rpc-rearm-listener-socket-early.patch
Patch0258: 0258-cluster-dht-log-error-only-if-layout-healing-is-requ.patch
Patch0259: 0259-Quota-Turn-on-ssl-for-crawler-clients-if-needed.patch
Patch0260: 0260-dht-Avoid-dict-log-flooding-for-internal-MDS-xattr.patch
%description
GlusterFS is a distributed file-system capable of scaling to several
@ -1152,6 +1176,10 @@ find ./tests ./run-tests.sh -type f | cpio -pd %{buildroot}%{_prefix}/share/glus
install -p -m 0744 -D extras/command-completion/gluster.bash \
%{buildroot}%{_sysconfdir}/bash_completion.d/gluster
%if ( 0%{?_build_server} )
echo "RHGS 3.4.0" > %{buildroot}%{_datadir}/glusterfs/release
%endif
%clean
rm -rf %{buildroot}
@ -1853,6 +1881,7 @@ exit 0
# Extra utility script
%dir %{_libexecdir}/glusterfs
%{_datadir}/glusterfs/release
%dir %{_datadir}/glusterfs/scripts
%{_datadir}/glusterfs/scripts/stop-all-gluster-processes.sh
%if ( 0%{?_with_systemd:1} )
@ -2442,6 +2471,12 @@ fi
%endif
%changelog
* Wed May 09 2018 Milind Changire <mchangir@redhat.com> - 3.12.2-9
- fixes bugs bz#1546717 bz#1557551 bz#1558948 bz#1561999 bz#1563804
bz#1565015 bz#1565119 bz#1565399 bz#1565577 bz#1567100 bz#1567899 bz#1568374
bz#1568969 bz#1569490 bz#1570514 bz#1570541 bz#1570582 bz#1571645 bz#1572087
bz#1572585 bz#1575895
* Fri Apr 20 2018 Milind Changire <mchangir@redhat.com> - 3.12.2-8
- fixes bugs bz#1466129 bz#1475779 bz#1523216 bz#1535281 bz#1546941
bz#1550315 bz#1550991 bz#1553677 bz#1554291 bz#1559452 bz#1560955 bz#1562744