Description of problem: I create a replica 2 volume on one node, like this: Volume Name: test Type: Distributed-Replicate Volume ID: 7eca4759-2ffd-4970-aab8-804bb4daaca1 Status: Started Number of Bricks: 4 x 2 = 8 Transport-type: tcp Bricks: Brick1: node-1:/disk1 Brick2: node-1:/disk2 Brick3: node-1:/disk3 Brick4: node-1:/disk4 Brick5: node-1:/disk5 Brick6: node-1:/disk6 Brick7: node-1:/disk7 Brick8: node-1:/disk8 Options Reconfigured: performance.readdir-ahead: on transport.address-family: inet when I perform "gluster vol heal test statistics heal-count replica node-1:/disk[1......8]",only Show the first set of results（test-replica-0）: root@node-1:~/coding/glusterfs# gluster vol heal test statistics heal-count replica node-1:/disk5 Gathering count of entries to be healed per replica on volume test has been successful Brick node-1:/disk1 Number of entries: 0 Brick node-1:/disk2 Number of entries: 0 Brick node-1:/disk3 No gathered input for this brick Brick node-1:/disk4 No gathered input for this brick Brick node-1:/disk5 No gathered input for this brick Brick node-1:/disk6 No gathered input for this brick Brick node-1:/disk7 No gathered input for this brick Brick node-1:/disk8 No gathered input for this brick Is this a bug? Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
int _select_hxlator_with_matching_brick (xlator_t *this, glusterd_volinfo_t *volinfo, dict_t *dict, int *index) { char *hostname = NULL; char *path = NULL; glusterd_brickinfo_t *brickinfo = NULL; glusterd_conf_t *priv = NULL; int hxl_children = 0; priv = this->private; if (!dict || dict_get_str (dict, "per-replica-cmd-hostname", &hostname) || dict_get_str (dict, "per-replica-cmd-path", &path)) return -1; hxl_children = _get_hxl_children_count (volinfo); if ((*index) == 0) (*index)++; cds_list_for_each_entry (brickinfo, &volinfo->bricks, brick_list) { if (gf_uuid_is_null (brickinfo->uuid)) (void)glusterd_resolve_brick (brickinfo); if (!gf_uuid_compare (MY_UUID, brickinfo->uuid)) { _add_hxlator_to_dict (dict, volinfo, ((*index) - 1)/hxl_children, 0); return 1; } (*index)++; } return 0; } _select_hxlator_with_matching_brick() will get hostname and path, but not use. so match error ?
(In reply to jiademing.dd from comment #0) > Description of problem: > I create a replica 2 volume on one node, like this: > Volume Name: test > Type: Distributed-Replicate > Volume ID: 7eca4759-2ffd-4970-aab8-804bb4daaca1 > Status: Started > Number of Bricks: 4 x 2 = 8 > Transport-type: tcp > Bricks: > Brick1: node-1:/disk1 > Brick2: node-1:/disk2 > Brick3: node-1:/disk3 > Brick4: node-1:/disk4 > Brick5: node-1:/disk5 > Brick6: node-1:/disk6 > Brick7: node-1:/disk7 > Brick8: node-1:/disk8 > Options Reconfigured: > performance.readdir-ahead: on > transport.address-family: inet > > when I perform "gluster vol heal test statistics heal-count replica > node-1:/disk[1......8]",only Show the first set of results（test-replica-0）: > > root@node-1:~/coding/glusterfs# gluster vol heal test statistics heal-count > replica node-1:/disk5 > Gathering count of entries to be healed per replica on volume test has been > successful > > Brick node-1:/disk1 > Number of entries: 0 > > Brick node-1:/disk2 > Number of entries: 0 > > Brick node-1:/disk3 > No gathered input for this brick > > Brick node-1:/disk4 > No gathered input for this brick > > Brick node-1:/disk5 > No gathered input for this brick > > Brick node-1:/disk6 > No gathered input for this brick > > Brick node-1:/disk7 > No gathered input for this brick > > Brick node-1:/disk8 > No gathered input for this brick > > Is this a bug? > > Hi jiademing, This is how the current implementation is. See the last section of https://github.com/gluster/glusterfs-specs/blob/master/done/Features/afr-statistics.md , The command prints the values only for the replica pair given and for others, it prints "No gathered input for this brick". What we could do is print the number of entries only for the bricks of the replica given in the command and not print anything for the other bricks. Is that acceptable?
(In reply to Ravishankar N from comment #2) > (In reply to jiademing.dd from comment #0) > > Description of problem: > > I create a replica 2 volume on one node, like this: > > Volume Name: test > > Type: Distributed-Replicate > > Volume ID: 7eca4759-2ffd-4970-aab8-804bb4daaca1 > > Status: Started > > Number of Bricks: 4 x 2 = 8 > > Transport-type: tcp > > Bricks: > > Brick1: node-1:/disk1 > > Brick2: node-1:/disk2 > > Brick3: node-1:/disk3 > > Brick4: node-1:/disk4 > > Brick5: node-1:/disk5 > > Brick6: node-1:/disk6 > > Brick7: node-1:/disk7 > > Brick8: node-1:/disk8 > > Options Reconfigured: > > performance.readdir-ahead: on > > transport.address-family: inet > > > > when I perform "gluster vol heal test statistics heal-count replica > > node-1:/disk[1......8]",only Show the first set of results（test-replica-0）: > > > > root@node-1:~/coding/glusterfs# gluster vol heal test statistics heal-count > > replica node-1:/disk5 > > Gathering count of entries to be healed per replica on volume test has been > > successful > > > > Brick node-1:/disk1 > > Number of entries: 0 > > > > Brick node-1:/disk2 > > Number of entries: 0 > > > > Brick node-1:/disk3 > > No gathered input for this brick > > > > Brick node-1:/disk4 > > No gathered input for this brick > > > > Brick node-1:/disk5 > > No gathered input for this brick > > > > Brick node-1:/disk6 > > No gathered input for this brick > > > > Brick node-1:/disk7 > > No gathered input for this brick > > > > Brick node-1:/disk8 > > No gathered input for this brick > > > > Is this a bug? > > > > > > Hi jiademing, > This is how the current implementation is. > See the last section of > https://github.com/gluster/glusterfs-specs/blob/master/done/Features/afr- > statistics.md , The command prints the values only for the replica pair > given and for others, it prints "No gathered input for this brick". > > What we could do is print the number of entries only for the bricks of the > replica given in the command and not print anything for the other bricks. Is > that acceptable? Thank you for your reply, I know, but the problem is: I give other bricks not first replica pair,that command print the first replica pair (test-replica-0)'s info every time. Like this: gluster vol heal test statistics heal-count replica node-1:/disk5 Brick node-1:/disk1 Number of entries: 0 Brick node-1:/disk2 Number of entries: 0 Brick node-1:/disk3 No gathered input for this brick Brick node-1:/disk4 No gathered input for this brick Brick node-1:/disk5 No gathered input for this brick Brick node-1:/disk6 No gathered input for this brick Brick node-1:/disk7 No gathered input for this brick Brick node-1:/disk8 No gathered input for this brick it should print disk5 and disk6's entries,but print disk1 and disk2 every time.
Thanks for the clarification Jiademing. You're right about _select_hxlator_with_matching_brick(), this seems to be a regression introduced by http://review.gluster.org/#/c/9793/4/xlators/mgmt/glusterd/src/glusterd-op-sm.c. Feel free to send a patch if you like, re-adding the ALL_REPLICA and PER_REPLICA checks.
(In reply to Ravishankar N from comment #4) > Thanks for the clarification Jiademing. You're right about > _select_hxlator_with_matching_brick(), this seems to be a regression > introduced by > http://review.gluster.org/#/c/9793/4/xlators/mgmt/glusterd/src/glusterd-op- > sm.c. > > Feel free to send a patch if you like, re-adding the ALL_REPLICA and > PER_REPLICA checks. Ha, I haven't submit any patch before,for not familiar with the process.I will try to, Or you submitted and then told me.Thank you.
REVIEW: http://review.gluster.org/15494 (glusterd: "gluster v heal test statistics heal-count replica" output is not correct) posted (#1) for review on master by MOHIT AGRAWAL (moagrawa)
REVIEW: http://review.gluster.org/15494 (glusterd: "gluster v heal test statistics heal-count replica" output is not correct) posted (#2) for review on master by MOHIT AGRAWAL (moagrawa)
REVIEW: http://review.gluster.org/15494 (glusterd: "gluster v heal test statistics heal-count replica" output is not correct) posted (#3) for review on master by MOHIT AGRAWAL (moagrawa)
REVIEW: http://review.gluster.org/15494 (glusterd: "gluster v heal test statistics heal-count replica" output is not correct) posted (#4) for review on master by MOHIT AGRAWAL (moagrawa)
REVIEW: http://review.gluster.org/15494 (glusterd: "gluster v heal test statistics heal-count replica" output is not correct) posted (#5) for review on master by MOHIT AGRAWAL (moagrawa)
COMMIT: http://review.gluster.org/15494 committed in master by Atin Mukherjee (amukherj) ------ commit 7a80b6128ad91c1174a79b4fa6a0340dfd0b6d6b Author: Mohit Agrawal <moagrawa> Date: Tue Sep 13 18:27:45 2016 +0530 glusterd: "gluster v heal test statistics heal-count replica" output is not correct Problem : "gluster v heal test statistcs heal-count replica" does not show correct output. Solution: After update condition (match brick name) in _select_hxlator_with_matching_brick, it shows correct output. BUG: 1325792 Change-Id: I60cc7c68ea70bce267a747570f91dcddbc1d9016 Signed-off-by: Mohit Agrawal <moagrawa> Reviewed-on: http://review.gluster.org/15494 NetBSD-regression: NetBSD Build System <jenkins.org> Smoke: Gluster Build System <jenkins.org> CentOS-regression: Gluster Build System <jenkins.org> Reviewed-by: Ravishankar N <ravishankar> Reviewed-by: Atin Mukherjee <amukherj>
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.10.0, please open a new bug report. glusterfs-3.10.0 has been announced on the Gluster mailinglists [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution. [1] http://lists.gluster.org/pipermail/gluster-users/2017-February/030119.html [2] https://www.gluster.org/pipermail/gluster-users/