[Gluster-users] missing files on FUSE mount

Eli V eliventer at gmail.com
Fri Oct 23 15:33:42 UTC 2020


On Tue, Oct 20, 2020 at 8:41 AM Martín Lorenzo <mlorenzo at gmail.com> wrote:
>
> Hi, I have the following problem, I have a distributed replicated cluster set up with samba and CTDB, over fuse mount points
> I am having inconsistencies across the FUSE mounts, users report that files are disappearing after being copied/moved. I take a look at the mount points on each node, and they don't display the same data
>
> #### faulty mount point####
> [root at gluster6 ARRIBA GENTE martes 20 de octubre]# ll
> ls: cannot access PANEO VUELTA A CLASES CON TAPABOCAS.mpg: No such file or directory
> ls: cannot access PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg: No such file or directory
> total 633723
> drwxr-xr-x. 5 arribagente PN      4096 Oct 19 10:52 COMERCIAL AG martes 20 de octubre
> -rw-r--r--. 1 arribagente PN 648927236 Jun  3 07:16 PANEO FACHADA PALACIO LEGISLATIVO DRONE DIA Y NOCHE.mpg
> -?????????? ? ?           ?          ?            ? PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg
> -?????????? ? ?           ?          ?            ? PANEO VUELTA A CLASES CON TAPABOCAS.mpg
>
>
> ###healthy mount point###
> [root at gluster7 ARRIBA GENTE martes 20 de octubre]# ll
> total 3435596
> drwxr-xr-x. 5 arribagente PN       4096 Oct 19 10:52 COMERCIAL AG martes 20 de octubre
> -rw-r--r--. 1 arribagente PN  648927236 Jun  3 07:16 PANEO FACHADA PALACIO LEGISLATIVO DRONE DIA Y NOCHE.mpg
> -rw-r--r--. 1 arribagente PN 2084415492 Aug 18 09:14 PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg
> -rw-r--r--. 1 arribagente PN  784701444 Sep  4 07:23 PANEO VUELTA A CLASES CON TAPABOCAS.mpg
>
>  - So far the only way to solve this is to create a directory in the healthy mount point, on the same path:
> [root at gluster7 ARRIBA GENTE martes 20 de octubre]# mkdir hola
>
> - When you refresh the other mountpoint, and the issue is resolved:
> [root at gluster6 ARRIBA GENTE martes 20 de octubre]# ll
> total 3435600
> drwxr-xr-x. 5 arribagente PN         4096 Oct 19 10:52 COMERCIAL AG martes 20 de octubre
> drwxr-xr-x. 2 root        root       4096 Oct 20 08:45 hola
> -rw-r--r--. 1 arribagente PN    648927236 Jun  3 07:16 PANEO FACHADA PALACIO LEGISLATIVO DRONE DIA Y NOCHE.mpg
> -rw-r--r--. 1 arribagente PN   2084415492 Aug 18 09:14 PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg
> -rw-r--r--. 1 arribagente PN    784701444 Sep  4 07:23 PANEO VUELTA A CLASES CON TAPABOCAS.mpg
>
> Interestingly, the error occurs on the mount point where the files were copied. They don't show up as pending heal entries. I have around 15 people using them over samba, I think I'm having this issue reported every two days.
>
> I have an older cluster with similar issues, different gluster version, but a very similar topology (4 bricks, initially two bricks then expanded)
> Please note , the bricks aren't the same size (but their replicas are), so my other suspicion is that rebalancing has something to do with it.
>
> I'm trying to reproduce it over a small virtualized cluster, so far no results.
>
> Here are the cluster details
> four nodes, replica 2, plus one arbiter hosting 2 bricks
> I have 2 bricks with ~20 TB capacity and the other pair is ~48TB
> Volume Name: tapeless
> Type: Distributed-Replicate
> Volume ID: 53bfa86d-b390-496b-bbd7-c4bba625c956
> Status: Started
> Snapshot Count: 0
> Number of Bricks: 2 x (2 + 1) = 6
> Transport-type: tcp
> Bricks:
> Brick1: gluster6.glustersaeta.net:/data/glusterfs/tapeless/brick_6/brick
> Brick2: gluster7.glustersaeta.net:/data/glusterfs/tapeless/brick_7/brick
> Brick3: kitchen-store.glustersaeta.net:/data/glusterfs/tapeless/brick_1a/brick (arbiter)
> Brick4: gluster12.glustersaeta.net:/data/glusterfs/tapeless/brick_12/brick
> Brick5: gluster13.glustersaeta.net:/data/glusterfs/tapeless/brick_13/brick
> Brick6: kitchen-store.glustersaeta.net:/data/glusterfs/tapeless/brick_2a/brick (arbiter)
> Options Reconfigured:
> features.quota-deem-statfs: on
> performance.client-io-threads: on
> nfs.disable: on
> transport.address-family: inet
> features.quota: on
> features.inode-quota: on
> features.cache-invalidation: on
> features.cache-invalidation-timeout: 600
> performance.cache-samba-metadata: on
> performance.stat-prefetch: on
> performance.cache-invalidation: on
> performance.md-cache-timeout: 600
> network.inode-lru-limit: 200000
> performance.nl-cache: on
> performance.nl-cache-timeout: 600
> performance.readdir-ahead: on
> performance.parallel-readdir: on
> performance.cache-size: 1GB
> client.event-threads: 4
> server.event-threads: 4
> performance.normal-prio-threads: 16
> performance.io-thread-count: 32
> performance.write-behind-window-size: 8MB
> storage.batch-fsync-delay-usec: 0
> cluster.data-self-heal: on
> cluster.metadata-self-heal: on
> cluster.entry-self-heal: on
> cluster.self-heal-daemon: on
> performance.write-behind: on
> performance.open-behind: on
>
> Log section form faulty mount point. I think the [file exists] entries are from people trying to copy the missing files over an over
>
>
> [2020-10-20 11:31:03.034220] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:32:06.684329] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:33:02.191863] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:34:05.841608] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:35:20.736633] I [MSGID: 108026] [afr-self-heal-metadata.c:52:__afr_selfheal_metadata_do] 0-tapeless-replicate-1: performing metadata selfheal on 958dbd7a-3cd7-4b66-9038-76e5c5669644
> [2020-10-20 11:35:20.741213] I [MSGID: 108026] [afr-self-heal-common.c:1750:afr_log_selfheal] 0-tapeless-replicate-1: Completed metadata selfheal on 958dbd7a-3cd7-4b66-9038-76e5c5669644. sources=[0] 1  sinks=2
> [2020-10-20 11:35:04.278043] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> The message "I [MSGID: 108026] [afr-self-heal-metadata.c:52:__afr_selfheal_metadata_do] 0-tapeless-replicate-1: performing metadata selfheal on 958dbd7a-3cd7-4b66-9038-76e5c5669644" repeated 3 times between [2020-10-20 11:35:20.736633] and [2020-10-20 11:35:26.733298]
> The message "I [MSGID: 108026] [afr-self-heal-common.c:1750:afr_log_selfheal] 0-tapeless-replicate-1: Completed metadata selfheal on 958dbd7a-3cd7-4b66-9038-76e5c5669644. sources=[0] 1  sinks=2 " repeated 3 times between [2020-10-20 11:35:20.741213] and [2020-10-20 11:35:26.737629]
> [2020-10-20 11:36:02.548350] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:36:57.365537] I [MSGID: 108026] [afr-self-heal-metadata.c:52:__afr_selfheal_metadata_do] 0-tapeless-replicate-1: performing metadata selfheal on f4907af2-1775-4c46-89b5-e9776df6d5c7
> [2020-10-20 11:36:57.370824] I [MSGID: 108026] [afr-self-heal-common.c:1750:afr_log_selfheal] 0-tapeless-replicate-1: Completed metadata selfheal on f4907af2-1775-4c46-89b5-e9776df6d5c7. sources=[0] 1  sinks=2
> [2020-10-20 11:37:01.363925] I [MSGID: 108026] [afr-self-heal-metadata.c:52:__afr_selfheal_metadata_do] 0-tapeless-replicate-1: performing metadata selfheal on f4907af2-1775-4c46-89b5-e9776df6d5c7
> [2020-10-20 11:37:01.368069] I [MSGID: 108026] [afr-self-heal-common.c:1750:afr_log_selfheal] 0-tapeless-replicate-1: Completed metadata selfheal on f4907af2-1775-4c46-89b5-e9776df6d5c7. sources=[0] 1  sinks=2
> The message "I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0" repeated 3 times between [2020-10-20 11:36:02.548350] and [2020-10-20 11:37:36.389208]
> [2020-10-20 11:38:07.367113] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:39:01.595981] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:40:04.184899] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:41:07.833470] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:42:01.871621] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:43:04.399194] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:44:04.558647] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:44:15.953600] W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-5: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]
> [2020-10-20 11:44:15.953819] W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-2: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]
> [2020-10-20 11:44:15.954072] W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-3: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]
> [2020-10-20 11:44:15.954680] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043294: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> [2020-10-20 11:44:15.963175] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043306: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> [2020-10-20 11:44:15.971839] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043318: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> [2020-10-20 11:44:16.010242] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043403: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> [2020-10-20 11:44:16.020291] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043415: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> [2020-10-20 11:44:16.028857] W [fuse-bridge.c:2606:fuse_create_cbk] 0-glusterfs-fuse: 31043427: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg => -1 (File exists)
> The message "W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-5: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]" repeated 5 times between [2020-10-20 11:44:15.953600] and [2020-10-20 11:44:16.027785]
> The message "W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-2: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]" repeated 5 times between [2020-10-20 11:44:15.953819] and [2020-10-20 11:44:16.028331]
> The message "W [MSGID: 114031] [client-rpc-fops_v2.c:2114:client4_0_create_cbk] 0-tapeless-client-3: remote operation failed. Path: /PN/arribagente/PLAYER 2020/ARRIBA GENTE martes 20 de octubre/PANEO NIÑOS ESCUELAS CON TAPABOCAS.mpg [File exists]" repeated 5 times between [2020-10-20 11:44:15.954072] and [2020-10-20 11:44:16.028355]
> [2020-10-20 11:45:03.572106] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:45:40.080010] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> The message "I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0" repeated 2 times between [2020-10-20 11:45:40.080010] and [2020-10-20 11:47:10.871801]
> [2020-10-20 11:48:03.913129] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:49:05.082165] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:50:06.725722] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:51:04.254685] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:52:07.903617] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:53:01.420513] I [MSGID: 108026] [afr-self-heal-metadata.c:52:__afr_selfheal_metadata_do] 0-tapeless-replicate-0: performing metadata selfheal on 3c316533-5f47-4267-ac19-58b3be305b94
> [2020-10-20 11:53:01.428657] I [MSGID: 108026] [afr-self-heal-common.c:1750:afr_log_selfheal] 0-tapeless-replicate-0: Completed metadata selfheal on 3c316533-5f47-4267-ac19-58b3be305b94. sources=[0]  sinks=1 2
> The message "I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0" repeated 3 times between [2020-10-20 11:52:07.903617] and [2020-10-20 11:53:12.037835]
> [2020-10-20 11:54:02.208354] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:55:04.360284] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:56:09.508092] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:57:02.580970] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
> [2020-10-20 11:58:06.230698] I [MSGID: 108031] [afr-common.c:2581:afr_local_discovery_cbk] 0-tapeless-replicate-0: selecting local read_child tapeless-client-0
>
>
> Let me know if you need something else. Thank you for you suppoort!
> Best Regards,
> Martin Lorenzo
>
>
> ________
>
>
>
> Community Meeting Calendar:
>
> Schedule -
> Every 2nd and 4th Tuesday at 14:30 IST / 09:00 UTC
> Bridge: https://bluejeans.com/441850968
>
> Gluster-users mailing list
> Gluster-users at gluster.org
> https://lists.gluster.org/mailman/listinfo/gluster-users

I've seen very similar issues with changes, specifically in my case
deletion and then recreation of a large directory not appearing for a
host using the fuse client. My gluster has all bricks of identical
size, so I don't think that's the issue in and of itself. It seems
like some sort of fuse client cache invalidation bug to me, since a
umount && mount on the client with the problem always fixes it.


More information about the Gluster-users mailing list