With certain combinations of settings both on the server and the client, trying
to run an executable from a Samba share results in a crash (I pasted an example
from the journal at the end). To reproduce it, try to run any executable from a
mounted share, e.g. even an empty Bash script will do:
$ echo '#!/bin/bash' > /mnt/share/test.sh
$ chmod +x /mnt/share/test.sh
$ /mnt/share/test.sh
exec: Failed to execute process '/mnt/smbt/test.sh', unknown error number 5
Software versions:
- Client: Fedora 39, ("Linux server 6.6.9-200.fc39.x86_64"), although I already
noticed this problem about a year ago.
- Server: CentOS 7.9.2009 with Samba 4.10.16, AlmaLinux 8.10 with Samba 4.19.4
The options that seem to be related are the value of "cache" on the client and
whether "oplocks" are on or off on the server. These are the results of testing
different combinations:
- When mounting with "cache=loose", the problem never occurs, regardless of the
value of "oplocks" in the smb.conf
- With "cache=none", the crash always happens, no matter the value of "oplocks"
- But when mounting with "cache=strict" (or without specifying the option at
all), the crash occurs with "oplocks = yes", and doesn't with "oplocks = no"
For completeness, this is how the share is mounted in fstab:
//remote/share /mnt/share smb3 _netdev,noauto,multiuser,mfsymlinks,sec=krb5,user=$user,domain=$domain,x-systemd.automount 0 0
This is what the smb.conf looks like, without the (hopefully) irrelevant
domain membership and printing settings:
---------------------------------
[global]
case sensitive = yes
delete readonly = yes
map acl inherit = yes
vfs objects = acl_xattr
oplocks = no
level2 oplocks = no
min protocol = SMB2
[share]
path = /data/share
read only = no
acl_xattr:ignore system acl = yes
---------------------------------
And here is a crash record from the journal:
--------------------------------------------------------------------------------
Jun 13 10:08:13 server kernel: ------------[ cut here ]------------
Jun 13 10:08:13 server kernel: WARNING: CPU: 121 PID: 3906695 at fs/smb/client/file.c:3341 cifs_limit_bvec_su bset.constprop.0+0xf2/0x130 [cifs]
Jun 13 10:08:13 server kernel: Modules linked in: ib_core nls_utf8 cifs cifs_arc4 nls_ucs2_utils cifs_md4 dns _resolver fscache netfs nf_conntrack_netlink nfnetlink_acct wireguard curve25519_x86_64 libcurve25519_generic ip6_udp_tunnel udp _tunnel cfg80211 binfmt_misc nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_re ject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables nfnetlink qrtr vfat fat intel_rapl_msr intel_rapl_common amd64_edac edac_mce_amd dell_wmi ledtrig_audio kvm_amd sparse_keymap rfkill ipmi_ssif video kvm irqbypass dell_smbios dcdbas rapl dell_wmi_descriptor wmi_bmof acpi_cpufreq pcspkr acpi_ipmi ipmi_si ptdma ipmi_devintf i2c_piix4 k10temp ipmi_msghandler acpi_power_meter auth_rpcgss sunrpc fuse loop zram xfs crct10dif_pclmul crc32_pclmul qede crc32c_intel polyval_c lmulni polyval_generic ghash_clmulni_intel qed sha512_ssse3 sha256_ssse3 sha1_ssse3 tg3 mgag200 megaraid_sas i2c_algo_bit ccp cr c8 sp5100_tco wmi scsi_dh_rdac scsi_dh_emc
Jun 13 10:08:13 server kernel: scsi_dh_alua dm_multipath
Jun 13 10:08:13 server kernel: CPU: 121 PID: 3906695 Comm: fish Tainted: G W 6.6.9-200.fc39.x86_64 #1
Jun 13 10:08:13 server kernel: Hardware name: Dell Inc. PowerEdge R7525/0590KW, BIOS 2.12.4 07/26/2023
Jun 13 10:08:13 server kernel: RIP: 0010:cifs_limit_bvec_subset.constprop.0+0xf2/0x130 [cifs]
Jun 13 10:08:13 server kernel: Code: 41 8b 44 01 08 49 0f 46 db 48 39 d8 48 0f 47 c3 83 c7 01 83 c2 01 48 01 c1 48 29 c6 75 c4 41 89 3a 48 89 c8 5b e9 ae a8 2a c3 <0f> 0b 31 c9 48 89 c8 e9 a2 a8 2a c3 31 c9 31 ff 41 89 3a 48 89 c8
Jun 13 10:08:13 server kernel: RSP: 0018:ffffc90056d0bcc0 EFLAGS: 00010297
Jun 13 10:08:13 server kernel: RAX: ffff88c08373ab98 RBX: ffff88c08373ab40 RCX: 0000000000000100
Jun 13 10:08:13 server kernel: RDX: ffffc90056d0bd0c RSI: 0000000000000100 RDI: ffff88c08373ab98
Jun 13 10:08:13 server kernel: RBP: ffff88ab5ead6800 R08: 0000000000000001 R09: 0000000000000040
Jun 13 10:08:13 server kernel: R10: ffff88c08373ab40 R11: 0000000000000001 R12: ffff88e42e0c2400
Jun 13 10:08:13 server kernel: R13: 0000000000000100 R14: ffff88c0d3eade00 R15: 0000000000000000
Jun 13 10:08:13 server kernel: FS: 00007f475f6e4740(0000) GS:ffff88caffd40000(0000) knlGS:0000000000000000
Jun 13 10:08:13 server kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jun 13 10:08:13 server kernel: CR2: 00005572688872a0 CR3: 00000015eb622000 CR4: 0000000000350ee0
Jun 13 10:08:13 server kernel: Call Trace:
Jun 13 10:08:13 server kernel: <TASK>
Jun 13 10:08:13 server kernel: ? cifs_limit_bvec_subset.constprop.0+0xf2/0x130 [cifs]
Jun 13 10:08:13 server kernel: ? __warn+0x81/0x130
Jun 13 10:08:13 server kernel: ? cifs_limit_bvec_subset.constprop.0+0xf2/0x130 [cifs]
Jun 13 10:08:13 server kernel: ? report_bug+0x171/0x1a0
Jun 13 10:08:13 server kernel: ? handle_bug+0x3c/0x80
Jun 13 10:08:13 server kernel: ? exc_invalid_op+0x17/0x70
Jun 13 10:08:13 server kernel: ? asm_exc_invalid_op+0x1a/0x20
Jun 13 10:08:13 server kernel: ? cifs_limit_bvec_subset.constprop.0+0xf2/0x130 [cifs]
Jun 13 10:08:13 server kernel: cifs_send_async_read+0x12a/0x530 [cifs]
Jun 13 10:08:13 server kernel: __cifs_readv+0x244/0x440 [cifs]
Jun 13 10:08:13 server kernel: __kernel_read+0x137/0x2e0
Jun 13 10:08:13 server kernel: bprm_execve+0x225/0x650
Jun 13 10:08:13 server kernel: do_execveat_common.isra.0+0x1ad/0x250
Jun 13 10:08:13 server kernel: __x64_sys_execve+0x36/0x40
Jun 13 10:08:13 server kernel: do_syscall_64+0x60/0x90
Jun 13 10:08:13 server kernel: ? srso_return_thunk+0x5/0x10
Jun 13 10:08:13 server kernel: ? exc_page_fault+0x7f/0x180
Jun 13 10:08:13 server kernel: entry_SYSCALL_64_after_hwframe+0x6e/0xd8
Jun 13 10:08:13 server kernel: RIP: 0033:0x7f475f21b52b
Jun 13 10:08:13 server kernel: Code: 0f 1e fa 48 8b 05 ad 9a 0f 00 48 8b 10 e9 0d 00 00 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa b8 3b 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d d5 98 0f 00 f7 d8 64 89 01 48
Jun 13 10:08:13 server kernel: RSP: 002b:00007ffea9484b68 EFLAGS: 00000246 ORIG_RAX: 000000000000003b
Jun 13 10:08:13 server kernel: RAX: ffffffffffffffda RBX: 00005572691924c0 RCX: 00007f475f21b52b
Jun 13 10:08:13 server kernel: RDX: 000055726917d720 RSI: 00005572691924c0 RDI: 000055726918c490
Jun 13 10:08:13 server kernel: RBP: 00007ffea9484fd0 R08: 0000000000000000 R09: 00007f475f6e4740
Jun 13 10:08:13 server kernel: R10: 0000000000000008 R11: 0000000000000246 R12: 000055726917d720
Jun 13 10:08:13 server kernel: R13: 0000000000000000 R14: 0000000000000000 R15: 000055726918c490
Jun 13 10:08:13 server kernel: </TASK>
Jun 13 10:08:13 server kernel: ---[ end trace 0000000000000000 ]---
Jun 13 10:08:14 server abrt-dump-journal-oops[3580]: abrt-dump-journal-oops: Found oopses: 1
Jun 13 10:08:14 server abrt-dump-journal-oops[3580]: abrt-dump-journal-oops: Creating problem directories
Jun 13 10:08:14 server abrt-server[3906703]: Deleting problem directory oops-2024-06-13-10:08:14-3580-0 (dup of oops-2023-09-26-08:23:20-3349-0)
Jun 13 10:08:14 server abrt-notification[3906710]: System encountered a non-fatal error in cifs_limit_bvec_subset()
Jun 13 10:08:15 server abrt-dump-journal-oops[3580]: Reported 1 kernel oopses to Abrt
