Description of problem: Version-Release number of selected component (if applicable): Test system 01 { Fedora 30 Gluster 6.3 Samba 4.10.5-1 } Test system 02 { Gluster 4.1 Samba 4.9.10 } How reproducible: Everytime Control test: 1. Connect to known good share which doesn't use gluster VFS module 2. With an application like pathfinder, copy some files to the share 3. Files copy as expected Steps to Reproduce: 1. Connect to share 2. With an application like pathfinder, copy some files to the share 3. 'Attributes error' shown, 0 byte files created on share Actual results: 'Attributes error' shown, 0 byte files created on share Expected results: File's copy as expected Additional info: Tested with both a Gluster share and a standard XFS share. This was repeated on test system 01 and test system 02 (see above for versions). Test results: Test 01 = Copy files with pathfinder to Gluster VFS share = Fail Test 02 = Copy files with pathfinder to normal XFS share = Pass Test 03 = Copy files with Finder to Gluster VFS share = Pass Test 04 = Copy files with Finder to normal XFS share = Pass I've collected trace gluster client logs, and debug 10 samba logs. Please find logs here: https://drive.google.com/drive/folders/1zyp3vei8JXKQYDOI3kiK17Ha1jJBRP6h?usp=sharing Samba config: [global] security = ADS workgroup = MAGENTA realm = MAGENTA.LOCAL netbios name = MNAS-FEDORA30 max protocol = SMB3 min protocol = SMB2 ea support = yes clustering = yes server signing = no max log size = 10000 glusterfs:loglevel = 9 log file = /var/log/samba/log-%M.smbd logging = file log level = 10 template shell = /sbin/nologin winbind offline logon = false winbind refresh tickets = yes winbind enum users = Yes winbind enum groups = Yes allow trusted domains = yes passdb backend = tdbsam idmap cache time = 604800 idmap negative cache time = 300 winbind cache time = 604800 idmap config magenta:backend = rid idmap config magenta:range = 10000-999999 idmap config * : backend = tdb idmap config * : range = 3000-7999 guest account = nobody map to guest = bad user force directory mode = 0777 force create mode = 0777 create mask = 0777 directory mask = 0777 hide unreadable = no store dos attributes = no unix extensions = no load printers = no printing = bsd printcap name = /dev/null disable spoolss = yes glusterfs:volfile_server = localhost kernel share modes = No strict locking = auto oplocks = yes durable handles = yes kernel oplocks = no posix locking = no level2 oplocks = no readdir_attr:aapl_rsize = yes readdir_attr:aapl_finder_info = no readdir_attr:aapl_max_access = no fruit:aapl = yes fruit:resource = xattr fruit:metadata = netatalk fruit:delete_empty_adfiles = yes [QC] guest ok = no read only = no vfs objects = catia fruit streams_xattr glusterfs glusterfs:volume = mcv01 path = "/data/qc_only" valid users = @"QC_ops" recycle:repository = .recycle recycle:keeptree = yes recycle:versions = yes recycle:directory_mode = 0770 recycle:subdir_mode = 0777 glusterfs:logfile = /var/log/samba/glusterfs-mcv01.%M.log [QC-XFS] guest ok = no read only = no vfs objects = catia fruit streams_xattr path = /tmp/share-test valid users = @"QC_ops" recycle:repository = .recycle recycle:keeptree = yes recycle:versions = yes recycle:directory_mode = 0770 recycle:subdir_mode = 0777
@Ryan, Few queries: * How do we run pathfinder and what kind of application is it? * Can you give us the output of `gluster volume info mcv01`?
Hi Anoop, Apologies, should have included that. Pathfinder is just a third party file browser for OS X, see here https://cocoatech.com/. It doesn't do anything particularly special, just has a few more options than Finder. It's similar to Total Commander on windows. Here's the output you requested: Volume Name: mcv01 Type: Distributed-Replicate Volume ID: 1580ab45-0a14-4f2f-8958-b55b435cdc47 Status: Started Snapshot Count: 0 Number of Bricks: 2 x 2 = 4 Transport-type: tcp Bricks: Brick1: mcn01:/mnt/h1a/mcv01_data Brick2: mcn02:/mnt/h1b/mcv01_data Brick3: mcn01:/mnt/h2a/mcv01_data Brick4: mcn02:/mnt/h2b/mcv01_data Options Reconfigured: features.quota-deem-statfs: on network.ping-timeout: 5 storage.batch-fsync-delay-usec: 0 performance.cache-size: 1000MB performance.stat-prefetch: on features.cache-invalidation: on features.cache-invalidation-timeout: 600 performance.cache-invalidation: on performance.cache-samba-metadata: on performance.md-cache-timeout: 600 performance.io-thread-count: 32 performance.parallel-readdir: on performance.nl-cache: on performance.nl-cache-timeout: 600 cluster.lookup-optimize: on performance.write-behind-window-size: 1MB performance.client-io-threads: on client.event-threads: 4 server.event-threads: 4 auth.allow: 172.30.30.* transport.address-family: inet features.inode-quota: on features.quota: on nfs.disable: on config.transport: tcp cluster.brick-multiplex: off cluster.server-quorum-ratio: 50% Best, Ryan
@Ryan, Can you please check whether disabling "performance.cache-samba-metadata" option on the volume helps in resolving the issue? # gluster volume set mcv01 performance.cache-samba-metadata off
Hi Anoop, Thanks for your fast reply. Disabling that option did indeed fix the issue, many thanks. Please let me know if I can assist in anyway. Best, Ryan
(In reply to ryan from comment #4) > Disabling that option did indeed fix the issue, many thanks. Thanks for the confirmation. > Please let me know if I can assist in anyway. Based on the Samba and GlusterFS client log files provided, I can confirm that the root cause for this behaviour from Mac OS X(with pathfinder application) is same as detailed in bug #1726205 description. Therefore closing out this bug report as duplicate. Further updates will be provided in the other bugzilla report. *** This bug has been marked as a duplicate of bug 1726205 ***
Great, thanks for the help! Best, Ryan