Re: [regression] nfsstat/nfsd crash system "general protection fault, probably for non-canonical address ..." after 6.8.9->6.8.10 update

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, May 22, 2024 at 04:36:57AM -0400, Jaroslav Pulchart wrote:
> Hello,
> 
> I would like to report some issue causing a "general protection fault"
> crash (constantly) after we updated the kernel from 6.8.9 to 6.8.10.
> This is triggered when monitoring is using nfsstat on a server where
> nfsd is running.
> 
> [ 3049.260633] general protection fault, probably for non-canonical
> address 0x66fb103e19e9cc89: 0000 [#1] PREEMPT SMP NOPTI
> [ 3049.261628] CPU: 22 PID: 74991 Comm: nfsstat Tainted: G
> E      6.8.10-1.gdc.el9.x86_64 #1
> [ 3049.262336] Hardware name: RDO OpenStack Compute/RHEL, BIOS
> edk2-20240214-2.el9 02/14/2024
> [ 3049.263003] RIP: 0010:_raw_spin_lock_irqsave+0x19/0x40
> [ 3049.263487] Code: cc 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
> 90 0f 1f 44 00 00 41 54 9c 41 5c fa 65 ff 05 a6 92 f5 42 31 c0 ba 01
> 00 00 00 <f0> 0f b1 17 75 0a 4c 89 e0 41 5c c3 cc cc cc cc 89 c6 e8 d0
> 07 00
> [ 3049.264882] RSP: 0018:ffffb1bca6b9bd00 EFLAGS: 00010046
> [ 3049.265365] RAX: 0000000000000000 RBX: 66fb103e19e9c989 RCX: 0000000000000001
> [ 3049.265953] RDX: 0000000000000001 RSI: 0000000000000001 RDI: 66fb103e19e9cc89
> [ 3049.266542] RBP: ffffffffc15df280 R08: 0000000000000001 R09: ffffa049a1785cb8
> [ 3049.267112] R10: ffffb1bca6b9bd70 R11: ffffa04964e49000 R12: 0000000000000246
> [ 3049.267702] R13: 66fb103e19e9cc89 R14: ffffa048445590a0 R15: 0000000000000001
> [ 3049.268278] FS:  00007fa3ddf03740(0000) GS:ffffa05703d00000(0000)
> knlGS:0000000000000000
> [ 3049.268928] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 3049.269443] CR2: 00007fa3dddfca50 CR3: 0000000342d1e004 CR4: 0000000000770ef0
> [ 3049.270025] PKRU: 55555554
> [ 3049.270371] Call Trace:
> [ 3049.270723]  <TASK>
> [ 3049.271035]  ? die_addr+0x33/0x90
> [ 3049.271423]  ? exc_general_protection+0x1ea/0x450
> [ 3049.271879]  ? asm_exc_general_protection+0x22/0x30
> [ 3049.272344]  ? _raw_spin_lock_irqsave+0x19/0x40
> [ 3049.272803]  __percpu_counter_sum+0xd/0x70
> [ 3049.273219]  nfsd_show+0x4f/0x1d0 [nfsd]
> [ 3049.273666]  seq_read_iter+0x11d/0x4d0
> [ 3049.274073]  ? avc_has_perm+0x42/0xc0
> [ 3049.274489]  seq_read+0xfe/0x140
> [ 3049.274866]  proc_reg_read+0x56/0xa0
> [ 3049.275257]  vfs_read+0xa7/0x340
> [ 3049.275647]  ? __do_sys_newfstat+0x57/0x60
> [ 3049.276059]  ksys_read+0x5f/0xe0
> [ 3049.276439]  do_syscall_64+0x5e/0x170
> [ 3049.276836]  entry_SYSCALL_64_after_hwframe+0x78/0x80
> [ 3049.277296] RIP: 0033:0x7fa3ddcfd9b2
> [ 3049.277719] Code: c0 e9 b2 fe ff ff 50 48 8d 3d ea 1d 0c 00 e8 c5
> fd 01 00 0f 1f 44 00 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75
> 10 0f 05 <48> 3d 00 f0 ff ff 77 56 c3 0f 1f 44 00 00 48 83 ec 28 48 89
> 54 24
> [ 3049.279139] RSP: 002b:00007ffd930672e8 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000000
> [ 3049.279788] RAX: ffffffffffffffda RBX: 0000555ded47c2a0 RCX: 00007fa3ddcfd9b2
> [ 3049.280402] RDX: 0000000000000400 RSI: 0000555ded47c480 RDI: 0000000000000003
> [ 3049.281046] RBP: 00007fa3dddf75e0 R08: 0000000000000003 R09: 0000000000000077
> [ 3049.281673] R10: 000000000000005d R11: 0000000000000246 R12: 0000555ded47c2a0
> [ 3049.282307] R13: 0000000000000d68 R14: 00007fa3dddf69e0 R15: 0000000000000d68
> [ 3049.282928]  </TASK>
> [ 3049.283310] Modules linked in: mptcp_diag(E) xsk_diag(E)
> raw_diag(E) unix_diag(E) af_packet_diag(E) netlink_diag(E) udp_diag(E)
> tcp_diag(E) inet_diag(E) tun(E) br_netfilter(E) bridge(E) stp(E)
> llc(E) nfsd(E) auth_rpcgss(E) nfs_acl(E) lockd(E) grace(E) sunrpc(E)
> nf_conntrack(E) nf_defrag_ipv6(E) nf_defrag_ipv4(E) binfmt_misc(E)
> zram(E) tls(E) isofs(E) vfat(E) fat(E) intel_rapl_msr(E)
> intel_rapl_common(E) kvm_amd(E) ccp(E) kvm(E) irqbypass(E)
> virtio_net(E) i2c_i801(E) virtio_gpu(E) i2c_smbus(E) net_failover(E)
> virtio_balloon(E) failover(E) virtio_dma_buf(E) fuse(E) ext4(E)
> mbcache(E) jbd2(E) sr_mod(E) cdrom(E) sg(E) ahci(E) libahci(E)
> crct10dif_pclmul(E) crc32_pclmul(Ea) polyval_clmulni(E)
> polyval_generic(E) libata(E) ghash_clmulni_intel(E) sha512_ssse3(E)
> virtio_blk(E) serio_raw(E) btrfs(E) xor(E) zstd_compress(E)
> raid6_pq(E) libcrc32c(E) crc32c_intel(E) dm_mirror(E)
> dm_region_hash(E) dm_log(E) dm_mod(E)
> [ 3049.283345] Unloaded tainted modules: edac_mce_amd(E):1 padlock_aes(E)
> 
> Any suggestion on how to fix it is appreciated.

Bisect between v6.8.9 and v6.8.10 would give us the exact point
where the failures were introduced.

I see that GregKH pulled in:

26a0ddb04230 ("nfsd: rename NFSD_NET_* to NFSD_STATS_*")
b7b05f98f3f0 ("nfsd: expose /proc/net/sunrpc/nfsd in net namespaces")
abf5fb593c90 ("nfsd: make all of the nfsd stats per-network namespace")

for v6.8.10 as a Stable-Dep-of: 18180a4550d0 ("NFSD: Fix nfsd4_encode_fattr4() crasher")

Which is a little baffling, I don't see how those two change sets
are mechanically related to each other. But I suspect the culprit is
one of those three stat-related patches.


-- 
Chuck Lever




[Index of Archives]     [Linux Filesystem Development]     [Linux USB Development]     [Linux Media Development]     [Video for Linux]     [Linux NILFS]     [Linux Audio Users]     [Yosemite Info]     [Linux SCSI]

  Powered by Linux