features/bit-rot-stub: versioning of objects in write/truncate fop instead of open

* This patch brings in the changes where object versioning is done in write and
  truncate fops instead of tracking them in open and create fops. This model
  works for both regular and anonymous fds. It also removes the race associated
  with open calls, create and lookups.

  This patch follows the below method for object versioning and notifications:

  Before sending writev on the fd, increase the ongoing
  version first. This makes anonymous fd write similar to the regular
  fd write by having the ongoing version increased before doing the
  write.

  Do following steps to do versioning:
  1) For anonymous fds set the fd context (so that release is invoked) and add
     the fd context to the list maintained in the inode context.
     For regular fds the above think would have been done in open itself.
  2) Increase the on-disk ongoing version
  3) Increase the in memory ongoing version and mark inode as non-dirty
  3) Once versioning is successfully done send write operation. If
     versioning fails, then fail the write fop.
  5) In writev_cbk mark inode as modified.

> Change-Id: I7104391bbe076d8fc49b68745d2ec29a6e92476c
> BUG: 1207979
> Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
> Reviewed-on: http://review.gluster.org/10233
> Tested-by: Gluster Build System <jenkins@build.gluster.com>
> Reviewed-by: Vijay Bellur <vbellur@redhat.com>

Change-Id: I4bb86989b5fab02b9ed2950798b1a80e566f1024
BUG: 1220041
Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Reviewed-on: http://review.gluster.org/10722
Reviewed-by: Gaurav Kumar Garg <ggarg@redhat.com>
Tested-by: NetBSD Build System
Tested-by: Gluster Build System <jenkins@build.gluster.com>

1 files changed, 38 insertions, 9 deletions

diff --git a/tests/bitrot/br-stub.c b/tests/bitrot/br-stub.c
index 6cb12ed0a8e..e164170bb83 100644
--- a/tests/bitrot/br-stub.c
+++ b/tests/bitrot/br-stub.c
@@ -31,8 +31,10 @@ brstub_validate_version (char *bpath, unsigned long version)
         if (ret < 0)
                 goto err;
 
-        if (xv->ongoingversion != version)
+        if (xv->ongoingversion != version) {
                 match = -1;
+                fprintf (stderr, "ongoingversion: %lu\n", xv->ongoingversion);
+        }
         free (xv);
 
         return match;
@@ -42,11 +44,12 @@ brstub_validate_version (char *bpath, unsigned long version)
 }
 
 int
-brstub_open_validation (char *filp, char *bpath, unsigned long startversion)
+brstub_write_validation (char *filp, char *bpath, unsigned long startversion)
 {
         int fd1 = 0;
         int fd2 = 0;
         int ret = 0;
+        char *string = "string\n";
 
         /* read only check */
         fd1 = open (filp, O_RDONLY);
@@ -55,18 +58,37 @@ brstub_open_validation (char *filp, char *bpath, unsigned long startversion)
         close (fd1);
 
         ret = brstub_validate_version (bpath, startversion);
-        if (ret < 0)
+        if (ret == 0)
                 goto err;
-
         /* single open (write/) check */
         fd1 = open (filp, O_RDWR);
         if (fd1 < 0)
                 goto err;
-        close (fd1);
 
+        ret = write (fd1, string, strlen (string));
+        if (ret <= 0)
+                goto err;
+        /**
+         * Fsync is done so that the write call has properly reached the
+         * disk. For fuse mounts write-behind xlator would have held the
+         * writes with itself and for nfs, client would have held the
+         * write in its cache. So write fop would not have triggered the
+         * versioning as it would have not reached the bit-rot-stub.
+         */
+        fsync (fd1);
         startversion++;
         ret = brstub_validate_version (bpath, startversion);
+        if (ret < 0)
+                goto err;
+        ret = write (fd1, string, strlen (string));
+        if (ret <= 0)
+                goto err;
 
+        ret = brstub_validate_version (bpath, startversion);
+        if (ret < 0)
+                goto err;
+
+        close (fd1);
         /* multi open (write/) check */
         fd1 = open (filp, O_RDWR);
         if (fd1 < 0)
@@ -74,13 +96,20 @@ brstub_open_validation (char *filp, char *bpath, unsigned long startversion)
         fd2 = open (filp, O_WRONLY);
         if (fd1 < 0)
                 goto err;
+
+        ret = write (fd1, string, strlen (string));
+        if (ret <= 0)
+                goto err;
+
+        ret = write (fd1, string, strlen (string));
+        if (ret <= 0)
+                goto err;
         close (fd1);
         close (fd2);
 
         /**
-         * incremented once per open()/open().../close()/close() sequence
+         * incremented once per write()/write().../close()/close() sequence
          */
-        startversion++;
         ret = brstub_validate_version (bpath, startversion);
         if (ret < 0)
                 goto err;
@@ -106,11 +135,11 @@ brstub_new_object_validate (char *filp, char *brick)
 
         printf ("Validating initial version..\n");
         ret = brstub_validate_version (bpath, 1);
-        if (ret < 0)
+        if (ret == 0)
                 goto err;
 
         printf ("Validating version on modifications..\n");
-        ret = brstub_open_validation (filp, bpath, 1);
+        ret = brstub_write_validation (filp, bpath, 1);
         if (ret < 0)
                 goto err;