Bug 1733425

Summary: Setting volume option when one of the glusterd is stopped in the cluster, post glusterd restart seeing couldn't find vol info in glusterd logs and shd, brick process offline
Product: [Community] GlusterFS Reporter: Atin Mukherjee <amukherj>
Component: glusterdAssignee: Sanju <srakonde>
Status: CLOSED NEXTRELEASE QA Contact:
Severity: high Docs Contact:
Priority: high    
Version: mainlineCC: amukherj, atumball, bmekala, bugs, nchilaka, rhs-bugs, sankarshan, storage-qa-internal, vbellur, vdas
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1727269
: 1812849 (view as bug list) Environment:
Last Closed: 2019-08-26 06:27:50 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1727269, 1812849    

Comment 1 Atin Mukherjee 2019-07-26 05:09:14 UTC
Description of problem:
When glusterd is stopped on one node and performed volume set operation on only two volumes(one is upgraded and another volume is newly created.). Restarted glusterd on the node where it stopped on that node seeing "Couldn't find volinfo" errors and shd and brick process on the two volume where volume option is set are offline.


Version-Release number of selected component (if applicable):
mainline

How reproducible:
1/1

Steps to Reproduce:
1. Create 1000 volumes and start them
2. Performed in-service upgrade on the cluster. 
3. Create 1000 more volumes (1X3 replicate) and start them
4. Stop glusterd on node N1.
5. Perform volume set operation on two volumes from n2 'performance.readdir-ahead on'. One which is being used for upgrade and another newly created volume. 
6. Start glusterd on node N1.

Actual results:
glusterfsd and shd on these two volumes are offline.
seeing "couldn't find vol info" in glusterd logs. 

Expected results:
glusterfsd and shd should be online and should not see any errors with "Couldn't find volinfo" in glusterd log.

Additional info:

glusterd log snippet
#########################################
[2019-07-05 06:49:29.078858] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:29.078892] E [MSGID: 106619] [glusterd-svc-helper.c:791:__glusterd_send_svc_configure_req] 0-management: Unable to stat  (No such file or directory)
[2019-07-05 06:49:30.079325] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:31.079862] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:32.080355] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:33.080899] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:34.081374] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
The message "E [MSGID: 106619] [glusterd-svc-helper.c:791:__glusterd_send_svc_configure_req] 0-management: Unable to stat  (No such file or directory)" repeated 5 times between [2019-07-05 06:49:29.078892] and [2019-07-05 06:49:34.081406]
[2019-07-05 06:49:35.081508] W [MSGID: 106619] [glusterd-svc-helper.c:931:glusterd_attach_svc] 0-glusterd: attach failed for glustershd(volume=ronnie_-999)
[2019-07-05 06:49:35.081583] E [MSGID: 106048] [glusterd-shd-svc.c:477:glusterd_shdsvc_start] 0-glusterd: Failed to attach shd svc(volume=ronnie_-999) to pid=-1
[2019-07-05 06:49:44.928310] E [MSGID: 106558] [glusterd-volgen.c:3289:volgen_graph_build_clients] 0-glusterd: volume inconsistency: brick count is 0
[2019-07-05 06:49:44.928395] E [MSGID: 106068] [glusterd-shd-svc.c:191:glusterd_shdsvc_create_volfile] 0-management: Failed to create volfile
[2019-07-05 06:49:44.941696] I [glusterd-utils.c:6560:glusterd_brick_start] 0-management: discovered already-running brick /bricks/brick9/testvol_-999
[2019-07-05 06:49:44.941768] I [MSGID: 106142] [glusterd-pmap.c:290:pmap_registry_bind] 0-pmap: adding brick /bricks/brick9/testvol_-999 on port 49155
[2019-07-05 06:49:44.941824] I [rpc-clnt.c:1014:rpc_clnt_connection_init] 0-management: setting frame-timeout to 600
[2019-07-05 06:49:44.945117] I [MSGID: 106620] [glusterd-svc-helper.c:900:glusterd_attach_svc] 0-glusterd: adding svc glustershd (volume=testvol_-999) to existing process with pid 590
[2019-07-05 06:49:44.945285] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:44.945316] E [MSGID: 106619] [glusterd-svc-helper.c:791:__glusterd_send_svc_configure_req] 0-management: Unable to stat  (No such file or directory)
[2019-07-05 06:49:45.018727] I [socket.c:3837:socket_submit_outgoing_msg] 0-socket.management: not connected (priv->connected = -1)
[2019-07-05 06:49:45.018748] E [rpcsvc.c:1583:rpcsvc_submit_generic] 0-rpc-service: failed to submit message (XID: 0x2, Program: GlusterD svc cli, ProgVers: 2, Proc: 27) to rpc-transport (socket.management)
[2019-07-05 06:49:45.018761] E [MSGID: 106430] [glusterd-utils.c:600:glusterd_submit_reply] 0-glusterd: Reply submission failed
[2019-07-05 06:49:45.023858] W [MSGID: 106056] [glusterd-snapshot-utils.c:376:glusterd_snap_volinfo_find_by_volume_id] 0-management: Snap volume not found
[2019-07-05 06:49:45.945716] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:45.945750] E [MSGID: 106619] [glusterd-svc-helper.c:791:__glusterd_send_svc_configure_req] 0-management: Unable to stat  (No such file or directory)
[2019-07-05 06:49:46.447404] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:46.946292] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:47.946837] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:48.947321] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:49.947840] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:50.948339] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:51.948895] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:52.949406] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:53.950006] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:54.950527] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:55.951051] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:56.951573] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:57.952095] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
[2019-07-05 06:49:58.952597] E [glusterd-handshake.c:343:build_volfile_path] 0-management: Couldn't find volinfo
The message "E [MSGID: 106619] [glusterd-svc-helper.c:791:__glusterd_send_svc_configure_req] 0-management: Unable to stat  (No such file or directory)" repeated 13 times between [2019-07-05 06:49:45.945750] and [2019-07-05 06:49:58.952676]
[2019-07-05 06:49:59.952756] W [MSGID: 106619] [glusterd-svc-helper.c:931:glusterd_attach_svc] 0-glusterd: attach failed for glustershd(volume=testvol_-999)
[2019-07-05 06:49:59.953137] E [MSGID: 106048] [glusterd-shd-svc.c:477:glusterd_shdsvc_start] 0-glusterd: Failed to attach shd svc(volume=testvol_-999) to pid=590

Comment 2 Worker Ant 2019-07-26 05:14:31 UTC
REVIEW: https://review.gluster.org/23042 (glusterd: stop stale bricks during handshaking in brick mux mode) posted (#3) for review on master by Atin Mukherjee

Comment 3 Worker Ant 2019-08-26 06:27:50 UTC
REVIEW: https://review.gluster.org/23042 (glusterd: stop stale bricks during handshaking in brick mux mode) merged (#6) on master by Atin Mukherjee