diff options
author | Sunil Kumar Acharya <sheggodu@redhat.com> | 2017-07-05 16:41:38 +0530 |
---|---|---|
committer | Shyamsundar Ranganathan <srangana@redhat.com> | 2017-07-19 11:26:21 +0000 |
commit | 425c5acca90bd8c00b94cdcd5082ccc7c1ba078b (patch) | |
tree | 810a802769c77d030f5837837bc4b7a663bcaeba /xlators | |
parent | 59d29cfc4300251b8bc92e41e5bc29790d8f2db6 (diff) |
cluster/ec: Non-disruptive upgrade on EC volume fails
Problem:
Enabling optimistic changelog on EC volume was not
handling node down scenarios appropriately resulting
in volume data inaccessibility.
Solution:
Update dirty xattr appropriately on good bricks whenever
nodes are down. This would fix the metadata information
as part of heal and thus ensures data accessibility.
>BUG: 1468261
>Change-Id: I08b0d28df386d9b2b49c3de84b4aac1c729ac057
>Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
>Reviewed-on: https://review.gluster.org/17703
>Smoke: Gluster Build System <jenkins@build.gluster.org>
>CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
>Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
BUG: 1470938
Change-Id: I08b0d28df386d9b2b49c3de84b4aac1c729ac057
Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
Reviewed-on: https://review.gluster.org/17773
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Ashish Pandey <aspandey@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Diffstat (limited to 'xlators')
-rw-r--r-- | xlators/cluster/ec/src/ec-common.c | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/xlators/cluster/ec/src/ec-common.c b/xlators/cluster/ec/src/ec-common.c index 284c2cd5a62..d3840a9057f 100644 --- a/xlators/cluster/ec/src/ec-common.c +++ b/xlators/cluster/ec/src/ec-common.c @@ -2065,7 +2065,10 @@ ec_update_info(ec_lock_link_t *link) /* If we set the dirty flag for update fop, we have to unset it. * If fop has failed on some bricks, leave the dirty as marked. */ if (lock->unlock_now) { - if (!(ec->node_mask & ~lock->good_mask)) { + /* Ensure that nodes are up while doing final + * metadata update.*/ + if (!(ec->node_mask & ~lock->good_mask) && + !(ec->node_mask & ~ec->xl_up)) { if (ctx->dirty[0] != 0) { dirty[0] = -1; } |