glusterd: add stripe_count in volume info (#3070)

[Sheetalpamecha]

glusterd: add stripe_count in volume info
Add stipe-count only if version is less than 10

Change-Id: Ibb589df2bb4c00c71850d787c998aff746321a17
Signed-off-by: Sheetal Pamecha spamecha@redhat.com

[Sheetalpamecha]

/run regression

[gluster-ant]

CLANG-FORMAT FAILURE:
Before merging the patch, this diff needs to be considered for passing clang-format

index c4d75b6aa..52cbc248f 100644
--- a/xlators/mgmt/glusterd/src/glusterd-store.c
+++ b/xlators/mgmt/glusterd/src/glusterd-store.c
@@ -2809,7 +2809,8 @@ glusterd_store_retrieve_bricks(glusterd_volinfo_t *volinfo)
             cds_list_add_tail(&ta_brickinfo->brick_list, &volinfo->ta_bricks);
             ta_brick_count++;
             if (gf_store_iter_destroy(&iter)) {
-                gf_msg(this->name, GF_LOG_ERROR, 0, GD_MSG_STORE_ITER_DESTROY_FAIL,
+                gf_msg(this->name, GF_LOG_ERROR, 0,
+                       GD_MSG_STORE_ITER_DESTROY_FAIL,
                        "Failed to destroy store iter");
                 ret = -1;
                 goto out;

[gluster-ant]

1 test(s) failed
./tests/bugs/glusterd/brick-order-check-add-brick.t

0 test(s) generated core

6 test(s) needed retry
./tests/000-flaky/basic_mount-nfs-auth.t
./tests/000-flaky/bugs_nfs_bug-1116503.t
./tests/000-flaky/glusterd-restart-shd-mux.t
./tests/basic/afr/gfid-mismatch-resolution-with-cli.t
./tests/basic/nl-cache.t
./tests/bugs/glusterd/brick-order-check-add-brick.t

3 flaky test(s) marked as success even though they failed
./tests/000-flaky/basic_mount-nfs-auth.t
./tests/000-flaky/bugs_nfs_bug-1116503.t
./tests/000-flaky/glusterd-restart-shd-mux.t
https://build.gluster.org/job/gh_centos7-regression/2035/

[Sheetalpamecha]

/run regression

[gluster-ant]

1 test(s) failed
./tests/bugs/glusterfs/bug-866459.t

0 test(s) generated core

4 test(s) needed retry
./tests/000-flaky/basic_afr_split-brain-favorite-child-policy.t
./tests/000-flaky/glusterd-restart-shd-mux.t
./tests/basic/glusterd/heald.t
./tests/bugs/glusterfs/bug-866459.t
https://build.gluster.org/job/gh_centos7-regression/2037/

[Shwetha-Acharya]

/run regression

[Sheetalpamecha]

/run regression

[nirs]

This broke oVirt, assuming that stripeCount is always available.
https://github.com/oVirt/vdsm/blob/8fcb01807518dcf445dafa19f0cd31d97e695485/lib/vdsm/gluster/cli.py#L429

[mykaul]

This broke oVirt, assuming that stripeCount is always available. https://github.com/oVirt/vdsm/blob/8fcb01807518dcf445dafa19f0cd31d97e695485/lib/vdsm/gluster/cli.py#L429

Probably the easiest would be to revert, or ptovide a fake value.
There is still an open question of why oVirt even needs this value in the 1st place.

[nirs]

The code parsing stripeCount was added in 8b0ac63819c1123ff6d7f725d624d2689e9a81df.

stripeCount is used in 106 lines in current ovirt-engine:
https://github.com/oVirt/ovirt-engine/search?q=stripeCount

So it seems that vdsm need to parse this value and report it to engine, and gluster should continue to report the value for backward compatibility with older clients.

[mykaul]

The code parsing stripeCount was added in 8b0ac63819c1123ff6d7f725d624d2689e9a81df.

stripeCount is used in 106 lines in current ovirt-engine: https://github.com/oVirt/ovirt-engine/search?q=stripeCount

So it seems that vdsm need to parse this value and report it to engine, and gluster should continue to report the value for backward compatibility with older clients.

I think at this point, if it's possible to patch VDSM to just return 1 as the stripe_count - I don't see (but perhaps I'm missing context) why would oVirt need stripe size value at all.

Of course, that doesn't solve the issue with previous oVirt versions and 10.x

[Shwetha-Acharya]

@rchikatw^^

[rchikatw]

Fixed in ovirt 4.5.1, see:
oVirt/vdsm#172

Actually, ovirt does not use the stripe count but I see the code with stripe count still exists in ovirt, As of now I just handled this in VDSM will clean up this code as soon as I can