ocfs2-devel.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: herbert.van.den.bergh at oracle.com <herbert.van.den.bergh@oracle.com>
To: ocfs2-devel@oss.oracle.com
Subject: [Ocfs2-devel] NFS clients crash OCFS2 nodes (general protection fault: 0000 [#1] SMP PTI)
Date: Tue, 2 Jun 2020 08:56:51 -0700	[thread overview]
Message-ID: <3873cb24-9db7-cc20-550a-ab82568464ca@oracle.com> (raw)
In-Reply-To: <75e2aedb-de1c-b21b-0d32-6c370ba209b8@qsdf.org>

Hi Pierre,

This is tracked in an internal bug tracking system, which is not 
available to the public or to customers. We'll make sure to post an 
announcement to this list when the bug is fixed.

Thanks,
Herbert.

On 6/2/20 3:10 AM, Pierre Dinh-van wrote:
> Hi Herbert,
>
> Is there somewhere a Bugtracker for it ? Or are the bugs @oracle not
> public anymore ?
>
> Browsing/searching bugzilla don't show me any bug :
> https://bugzilla.oracle.com/bugzilla/describecomponents.cgi
>
> Cheers
>
> Pierre
>
> On 5/21/20 12:01 AM, herbert.van.den.bergh at oracle.com wrote:
>> Pierre,
>>
>> A similar if not identical bug was reported against Oracle Linux. The
>> fix is still being developed. I will update that bug to reference this
>> thread, and ask the developer to submit the fix upstream and/or to the
>> ocfs2-devel list once it's ready.
>>
>> Thanks,
>> Herbert.
>>
>> On 5/19/20 3:48 AM, Pierre Dinh-van wrote:
>> Hi,
>>
>> I'm experiencing quite reproducable crashs on my OCFS2 cluster the last
>> weeks, and I think I found out part of the Problem. But before I make a
>> bug report (I'm not sure where), I though I could post my research here
>> in case someone understands it better than I.
>>
>> When it happens, the kernel is giving me a 'general protection fault:
>> 0000 [#1] SMP PTI' or ' BUG: unable to handle page fault for address:'
>> message, local access to the OCFS2 volumes are all hanging forever, and
>> the load is getting so high that only a hard reset "solves" the problem.
>> It happened the same way on both nodes (always on the active node
>> serving NFS). It happens when both nodes are up, or when only one is
>> active.
>>
>> First, a few words about my setup :
>>
>> - storage is Fiberchannel LUNs served by a EMC VNX5200 (4x 5TB).
>>
>> - LUNs are encrypted with LUKS
>>
>> - LUKS volumes are formated with OCFS2
>>
>> - heartbeat=local
>>
>> - The OCFS2 cluster is communicating on a dedictated IP over InfinyBand
>> direct link (actually a fail-over bonding of the 2 ib interfaces).
>>
>> - nodes are 2 debian buster hardware servers (24 cores, 128GB RAM).
>>
>> - I use corosync for cluster management, but I don't think it's related
>>
>> - The cluster serves 4 NFS exports (3 are NFSv3, and one is NFSv4 with
>> kerberos) and samba shares
>>
>> - The samba shares are served in parallel by the 2 nodes managed by
>> ctdb
>>
>> - The NFS exports are served by one of the node having the dedicated IP
>> for this export.
>>
>> - When corosync is moving the IP of a NFS export to the other node, it
>> restarts nfsd to issue a grace time of 10s on the new active node.
>>
>> - The bug appeared with default debian buster kernel 4.19.0-9-amd64 and
>> with the backported kernel 5.5.0-0.bpo.2-amd64 as well
>>
>>
>> My cluster went live a few weeks ago, and it first worked with both
>> nodes active. During some maintenance work where the active NFS node had
>> to switch, I had a a bug, that I was able to reproduce against my will
>> at least 20 times in the last nights :
>>
>> 1) around 100 NFS clients are connected with the exports and happy
>> working with it
>>
>> 2) The IPs of the NFS shares are switching from nodeA to nodeB
>>
>> 3) nodeB's kernel says "general protection fault: 0000 [#1] SMP PTI"
>>
>> 4) nodeB's load is getting high
>>
>> 5) access to the OCFS2 is hanging (clean reboot never end, system cannot
>> unmount OCFS volumes)
>>
>> 6) I power cycle nodeB
>>
>> 7) I start nodeA again
>>
>> 8) nodeA give the same symptoms, as soon as some NFS clients do some
>> special requests
>>
>>
>> Most of the time, when the IP jumps to the other node, I get some of
>> this messages :
>>
>>
>> [  501.707566] (nfsd,2133,14):ocfs2_test_inode_bit:2841 ERROR: unable to
>> get alloc inode in slot 65535
>> [  501.716660] (nfsd,2133,14):ocfs2_test_inode_bit:2867 ERROR: status
>> = -22
>> [  501.726585] (nfsd,2133,6):ocfs2_test_inode_bit:2841 ERROR: unable to
>> get alloc inode in slot 65535
>> [  501.735579] (nfsd,2133,6):ocfs2_test_inode_bit:2867 ERROR: status =
>> -22
>>
>> But it also happens when the node is not crashing, so I think it's
>> another Problem.
>>
>> Last night, I saved a few of the kernel output while the servers where
>> crashing one after the other like in a ping-pong game :
>>
>>
>> -----8< on nodeB  8<-----
>>
>> [  502.070431] nfsd: nfsv4 idmapping failing: has idmapd not been
>> started?
>> [  502.475747] general protection fault: 0000 [#1] SMP PTI
>> [  502.481027] CPU: 6 PID: 2104 Comm: nfsd Not tainted
>> 5.5.0-0.bpo.2-amd64 #1 Debian 5.5.17-1~bpo10+1
>> [  502.490016] Hardware name: IBM System x3650 M3
>> -[7945UHV]-/94Y7614     , BIOS -[D6E150CUS-1.11]- 02/08/2011
>> [  502.499821] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [  502.504206] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [  502.523018] RSP: 0018:ffffa5af4838bac0 EFLAGS: 00010246
>> [  502.528277] RAX: 0000000000000000 RBX: 0fdbbeded053c1d6 RCX:
>> 0000000000000000
>> [  502.535433] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 0fdbbeded053c25e
>> [  502.542588] RBP: 0fdbbeded053c25e R08: ffff9672ffa73068 R09:
>> 000000000002c340
>> [  502.549743] R10: 00000184d03a2115 R11: 0000000000000000 R12:
>> ffff9672eff43fd0
>> [  502.556899] R13: ffff9672fff66bc0 R14: 000000000000ffff R15:
>> ffff9672fff660c8
>> [  502.564055] FS:  0000000000000000(0000) GS:ffff968203a80000(0000)
>> knlGS:0000000000000000
>> [  502.572166] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [  502.577928] CR2: 00007f0a4bb5df50 CR3: 0000001e4169a004 CR4:
>> 00000000000206e0
>> [  502.585082] Call Trace:
>> [  502.587546]  igrab+0x19/0x50
>> [  502.590484]  ocfs2_get_system_file_inode+0x65/0x2c0 [ocfs2]
>> [  502.596094]  ? ocfs2_read_blocks_sync+0x159/0x330 [ocfs2]
>> [  502.601534]  ocfs2_test_inode_bit+0xe8/0x900 [ocfs2]
>> [  502.606536]  ? ocfs2_free_dir_lookup_result+0x24/0x50 [ocfs2]
>> [  502.612321]  ocfs2_get_parent+0xa3/0x300 [ocfs2]
>> [  502.616960]  reconnect_path+0xa1/0x2c0
>> [  502.620742]  ? nfsd_proc_setattr+0x1b0/0x1b0 [nfsd]
>> [  502.625636]  exportfs_decode_fh+0x111/0x2d0
>> [  502.629844]  ? exp_find_key+0xcd/0x160 [nfsd]
>> [  502.634218]  ? __kmalloc+0x180/0x270
>> [  502.637810]  ? security_prepare_creds+0x6f/0xa0
>> [  502.642356]  ? tomoyo_write_self+0x1b0/0x1b0
>> [  502.646642]  ? security_prepare_creds+0x49/0xa0
>> [  502.651995]  fh_verify+0x3e5/0x5f0 [nfsd]
>> [  502.656737]  nfsd3_proc_getattr+0x6b/0x100 [nfsd]
>> [  502.662152]  nfsd_dispatch+0x9e/0x210 [nfsd]
>> [  502.667136]  svc_process_common+0x386/0x6f0 [sunrpc]
>> [  502.672791]  ? svc_sock_secure_port+0x12/0x30 [sunrpc]
>> [  502.678628]  ? svc_recv+0x2ff/0x9c0 [sunrpc]
>> [  502.683597]  ? nfsd_svc+0x2c0/0x2c0 [nfsd]
>> [  502.688395]  ? nfsd_destroy+0x50/0x50 [nfsd]
>> [  502.693377]  svc_process+0xd1/0x110 [sunrpc]
>> [  502.698361]  nfsd+0xe3/0x140 [nfsd]
>> [  502.702561]  kthread+0x112/0x130
>> [  502.706504]  ? kthread_park+0x80/0x80
>> [  502.710886]  ret_from_fork+0x35/0x40
>> [  502.715181] Modules linked in: rpcsec_gss_krb5 nfsd nfs_acl lockd
>> grace ocfs2 quota_tree sctp libcrc32c nft_counter ipt_REJECT
>> nf_reject_ipv4 xt_tcpudp nft_compat nf_tables nfnetlink binfmt_misc
>> ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm ocfs2_nodemanager
>> cpufreq_userspace ocfs2_stackglue cpufreq_powersave cpufreq_conservative
>> configfs bonding dm_round_robin dm_multipath scsi_dh_rdac scsi_dh_emc
>> scsi_dh_alua intel_powerclamp coretemp kvm_intel ipmi_ssif kvm cdc_ether
>> irqbypass usbnet mii tpm_tis intel_cstate tpm_tis_core intel_uncore
>> joydev tpm ipmi_si ioatdma sg iTCO_wdt ib_mthca iTCO_vendor_support
>> watchdog dca pcspkr ipmi_devintf rng_core ib_uverbs ipmi_msghandler
>> i5500_temp i7core_edac evdev ib_umad ib_ipoib ib_cm ib_core auth_rpcgss
>> sunrpc ip_tables x_tables autofs4 algif_skcipher af_alg ext4 crc16
>> mbcache jbd2 crc32c_generic dm_crypt dm_mod sd_mod hid_generic usbhid
>> hid sr_mod cdrom crct10dif_pclmul crc32_pclmul crc32c_intel
>> ghash_clmulni_intel mgag200 drm_vram_helper drm_ttm_helper
>> [  502.715218]  i2c_algo_bit ttm qla2xxx drm_kms_helper drm ata_generic
>> ata_piix aesni_intel libata nvme_fc ehci_pci uhci_hcd crypto_simd
>> ehci_hcd nvme_fabrics cryptd glue_helper lpc_ich nvme_core megaraid_sas
>> mfd_core usbcore scsi_transport_fc e1000e scsi_mod i2c_i801 usb_common
>> ptp pps_core bnx2 button
>> [  502.837917] ---[ end trace 6e1a8e507ac30d8d ]---
>> [  502.843379] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [  502.848584] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [  502.869295] RSP: 0018:ffffa5af4838bac0 EFLAGS: 00010246
>> [  502.875404] RAX: 0000000000000000 RBX: 0fdbbeded053c1d6 RCX:
>> 0000000000000000
>> [  502.883423] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 0fdbbeded053c25e
>> [  502.891436] RBP: 0fdbbeded053c25e R08: ffff9672ffa73068 R09:
>> 000000000002c340
>> [  502.899446] R10: 00000184d03a2115 R11: 0000000000000000 R12:
>> ffff9672eff43fd0
>> [  502.907455] R13: ffff9672fff66bc0 R14: 000000000000ffff R15:
>> ffff9672fff660c8
>> [  502.915473] FS:  0000000000000000(0000) GS:ffff968203a80000(0000)
>> knlGS:0000000000000000
>> [  502.924496] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [  502.931147] CR2: 00007f0a4bb5df50 CR3: 0000001e4169a004 CR4:
>> 00000000000206e0
>>
>> -----8< cut here 8<-----
>>
>>
>> shortly after when trying to start nodeA while nodeB was down :
>>
>> -----8< nodeA 8<-----
>>
>> [  728.927750] general protection fault: 0000 [#1] SMP PTI
>> [  728.934647] CPU: 8 PID: 12156 Comm: nfsd Not tainted
>> 5.5.0-0.bpo.2-amd64 #1 Debian 5.5.17-1~bpo10+1
>> [  728.944800] Hardware name: IBM System x3650 M3 -[7945H2G]-/69Y4438,
>> BIOS -[D6E158AUS-1.16]- 11/26/2012
>> [  728.955176] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [  728.960595] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [  728.981559] RSP: 0018:ffffa24d09187ac0 EFLAGS: 00010246
>> [  728.987885] RAX: 0000000000000000 RBX: 625f656369766564 RCX:
>> 0000000000000000
>> [  728.996135] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 625f6563697665ec
>> [  729.004384] RBP: 625f6563697665ec R08: ffff91f1272e3680 R09:
>> 000000000002c340
>> [  729.012628] R10: 000002114c00d22b R11: 0000000000000000 R12:
>> ffff91e13858bbd0
>> [  729.020877] R13: ffff91f13c03bbc0 R14: 000000000000ffff R15:
>> ffff91f13c03b0c8
>> [  729.029108] FS:  0000000000000000(0000) GS:ffff91f13fa80000(0000)
>> knlGS:0000000000000000
>> [  729.038308] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [  729.045161] CR2: 000055fe95d1d9a8 CR3: 0000001c8420a003 CR4:
>> 00000000000206e0
>> [  729.053415] Call Trace:
>> [  729.056978]  igrab+0x19/0x50
>> [  729.061023]  ocfs2_get_system_file_inode+0x65/0x2c0 [ocfs2]
>> [  729.067736]  ? ocfs2_read_blocks_sync+0x159/0x330 [ocfs2]
>> [  729.074284]  ocfs2_test_inode_bit+0xe8/0x900 [ocfs2]
>> [  729.080386]  ? ocfs2_free_dir_lookup_result+0x24/0x50 [ocfs2]
>> [  729.087268]  ocfs2_get_parent+0xa3/0x300 [ocfs2]
>> [  729.092995]  reconnect_path+0xa1/0x2c0
>> [  729.097863]  ? nfsd_proc_setattr+0x1b0/0x1b0 [nfsd]
>> [  729.103840]  exportfs_decode_fh+0x111/0x2d0
>> [  729.109133]  ? exp_find_key+0xcd/0x160 [nfsd]
>> [  729.114574]  ? dbs_update_util_handler+0x16/0x90
>> [  729.120273]  ? __kmalloc+0x180/0x270
>> [  729.124920]  ? security_prepare_creds+0x6f/0xa0
>> [  729.130521]  ? tomoyo_write_self+0x1b0/0x1b0
>> [  729.135858]  ? security_prepare_creds+0x49/0xa0
>> [  729.141464]  fh_verify+0x3e5/0x5f0 [nfsd]
>> [  729.146536]  nfsd3_proc_getattr+0x6b/0x100 [nfsd]
>> [  729.152297]  nfsd_dispatch+0x9e/0x210 [nfsd]
>> [  729.157647]  svc_process_common+0x386/0x6f0 [sunrpc]
>> [  729.163675]  ? svc_sock_secure_port+0x12/0x30 [sunrpc]
>> [  729.169873]  ? svc_recv+0x2ff/0x9c0 [sunrpc]
>> [  729.175190]  ? nfsd_svc+0x2c0/0x2c0 [nfsd]
>> [  729.180323]  ? nfsd_destroy+0x50/0x50 [nfsd]
>> [  729.185631]  svc_process+0xd1/0x110 [sunrpc]
>> [  729.190925]  nfsd+0xe3/0x140 [nfsd]
>> [  729.195421]  kthread+0x112/0x130
>> [  729.199636]  ? kthread_park+0x80/0x80
>> [  729.204279]  ret_from_fork+0x35/0x40
>> [  729.208818] Modules linked in: ocfs2 quota_tree dm_crypt crypto_simd
>> cryptd glue_helper algif_skcipher af_alg nfsd nfs_acl lockd grace sctp
>> libcrc32c nft_counter xt_tcpudp nft_compat nf_tables nfnetlink
>> binfmt_misc ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm ocfs2_nodemanager
>> ocfs2_stackglue configfs bonding dm_round_robin dm_multipath
>> scsi_dh_rdac scsi_dh_emc scsi_dh_alua intel_powerclamp ipmi_ssif
>> coretemp kvm_intel kvm irqbypass cdc_ether intel_cstate usbnet ib_mthca
>> mii joydev intel_uncore ib_uverbs iTCO_wdt sg ioatdma
>> iTCO_vendor_support i7core_edac watchdog i5500_temp pcspkr ipmi_si
>> tpm_tis tpm_tis_core tpm ipmi_devintf ipmi_msghandler rng_core evdev
>> acpi_cpufreq ib_umad auth_rpcgss ib_ipoib ib_cm sunrpc ib_core ip_tables
>> x_tables autofs4 ext4 crc16 mbcache jbd2 crc32c_generic dm_mod
>> hid_generic sd_mod usbhid hid sr_mod cdrom uas usb_storage mgag200
>> drm_vram_helper drm_ttm_helper qla2xxx ata_generic i2c_algo_bit ttm
>> nvme_fc ixgbe nvme_fabrics ata_piix xfrm_algo ehci_pci uhci_hcd
>> [  729.208868]  drm_kms_helper nvme_core ehci_hcd dca libata
>> megaraid_sas scsi_transport_fc libphy ptp drm usbcore scsi_mod
>> crc32c_intel lpc_ich i2c_i801 pps_core mfd_core usb_common bnx2 mdio
>> button
>> [  729.323900] ---[ end trace 212cefe75207e17f ]---
>> [  729.329644] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [  729.335121] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [  729.356192] RSP: 0018:ffffa24d09187ac0 EFLAGS: 00010246
>> [  729.362584] RAX: 0000000000000000 RBX: 625f656369766564 RCX:
>> 0000000000000000
>> [  729.370889] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 625f6563697665ec
>> [  729.379184] RBP: 625f6563697665ec R08: ffff91f1272e3680 R09:
>> 000000000002c340
>> [  729.387475] R10: 000002114c00d22b R11: 0000000000000000 R12:
>> ffff91e13858bbd0
>> [  729.395767] R13: ffff91f13c03bbc0 R14: 000000000000ffff R15:
>> ffff91f13c03b0c8
>> [  729.404066] FS:  0000000000000000(0000) GS:ffff91f13fa80000(0000)
>> knlGS:0000000000000000
>> [  729.413336] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [  729.420262] CR2: 000055fe95d1d9a8 CR3: 0000001c8420a003 CR4:
>> 00000000000206e0
>>
>> -----8< cut here 8<-----
>>
>> Since the backtrace is speaking about nfsd, I tried to isolate the
>> Problem by droping all NFS packet entering to the active node before
>> starting the services :
>>
>> iptables -I INPUT -p tcp --dport 2049 -j DROP
>>
>> Everything fine, access over samba were doing fine.
>>
>> I started to allow NFS for part of my network with iptables, until the
>> crash occurs.
>>
>> On this one, the crash occured after allowing only a NFSv3 client.
>>
>> -----8< nodeB 8<-----
>>
>> [ 2494.893697] BUG: unable to handle page fault for address:
>> 00000000000fab7d
>> [ 2494.900626] #PF: supervisor write access in kernel mode
>> [ 2494.905873] #PF: error_code(0x0002) - not-present page
>> [ 2494.911042] PGD 0 P4D 0
>> [ 2494.913596] Oops: 0002 [#1] SMP PTI
>> [ 2494.917106] CPU: 11 PID: 21342 Comm: nfsd Not tainted
>> 5.5.0-0.bpo.2-amd64 #1 Debian 5.5.17-1~bpo10+1
>> [ 2494.926285] Hardware name: IBM System x3650 M3
>> -[7945UHV]-/94Y7614     , BIOS -[D6E150CUS-1.11]- 02/08/2011
>> [ 2494.936077] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [ 2494.940461] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [ 2494.959269] RSP: 0018:ffffb949cca7fac0 EFLAGS: 00010246
>> [ 2494.964511] RAX: 0000000000000000 RBX: 00000000000faaf5 RCX:
>> 0000000000000000
>> [ 2494.971664] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 00000000000fab7d
>> [ 2494.978815] RBP: 00000000000fab7d R08: ffff99c8f12737b8 R09:
>> 000000000002c340
>> [ 2494.985968] R10: 000005dc16e333d8 R11: 0000000000000000 R12:
>> ffff99c903066bd0
>> [ 2494.993119] R13: ffff99c9014c7bc0 R14: 000000000000ffff R15:
>> ffff99c9014c70c8
>> [ 2495.000271] FS:  0000000000000000(0000) GS:ffff99ba03bc0000(0000)
>> knlGS:0000000000000000
>> [ 2495.008382] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [ 2495.014141] CR2: 00000000000fab7d CR3: 0000000f38df2002 CR4:
>> 00000000000206e0
>> [ 2495.021291] Call Trace:
>> [ 2495.023753]  igrab+0x19/0x50
>> [ 2495.026693]  ocfs2_get_system_file_inode+0x65/0x2c0 [ocfs2]
>> [ 2495.032301]  ? ocfs2_read_blocks_sync+0x159/0x330 [ocfs2]
>> [ 2495.037741]  ocfs2_test_inode_bit+0xe8/0x900 [ocfs2]
>> [ 2495.042742]  ? ocfs2_free_dir_lookup_result+0x24/0x50 [ocfs2]
>> [ 2495.048525]  ocfs2_get_parent+0xa3/0x300 [ocfs2]
>> [ 2495.053163]  reconnect_path+0xa1/0x2c0
>> [ 2495.056943]  ? nfsd_proc_setattr+0x1b0/0x1b0 [nfsd]
>> [ 2495.061838]  exportfs_decode_fh+0x111/0x2d0
>> [ 2495.066042]  ? exp_find_key+0xcd/0x160 [nfsd]
>> [ 2495.070417]  ? __kmalloc+0x180/0x270
>> [ 2495.074007]  ? security_prepare_creds+0x6f/0xa0
>> [ 2495.078551]  ? tomoyo_write_self+0x1b0/0x1b0
>> [ 2495.082834]  ? security_prepare_creds+0x49/0xa0
>> [ 2495.087387]  fh_verify+0x3e5/0x5f0 [nfsd]
>> [ 2495.092125]  nfsd3_proc_getattr+0x6b/0x100 [nfsd]
>> [ 2495.097517]  nfsd_dispatch+0x9e/0x210 [nfsd]
>> [ 2495.102489]  svc_process_common+0x386/0x6f0 [sunrpc]
>> [ 2495.108134]  ? svc_sock_secure_port+0x12/0x30 [sunrpc]
>> [ 2495.113946]  ? svc_recv+0x2ff/0x9c0 [sunrpc]
>> [ 2495.118883]  ? nfsd_svc+0x2c0/0x2c0 [nfsd]
>> [ 2495.123651]  ? nfsd_destroy+0x50/0x50 [nfsd]
>> [ 2495.128604]  svc_process+0xd1/0x110 [sunrpc]
>> [ 2495.133563]  nfsd+0xe3/0x140 [nfsd]
>> [ 2495.137733]  kthread+0x112/0x130
>> [ 2495.141643]  ? kthread_park+0x80/0x80
>> [ 2495.145993]  ret_from_fork+0x35/0x40
>> [ 2495.150262] Modules linked in: ocfs2 quota_tree nfsd nfs_acl lockd
>> grace sctp libcrc32c nft_counter xt_tcpudp nft_compat nf_tables
>> nfnetlink binfmt_misc ocfs2_dlmfs ocfs2_stack_o2cb ocfs2_dlm
>> ocfs2_nodemanager ocfs2_stackglue cpufreq_userspace cpufreq_powersave
>> cpufreq_conservative configfs bonding dm_round_robin dm_multipath
>> scsi_dh_rdac scsi_dh_emc scsi_dh_alua intel_powerclamp coretemp
>> kvm_intel kvm ipmi_ssif irqbypass cdc_ether usbnet intel_cstate mii
>> ib_mthca intel_uncore evdev iTCO_wdt joydev ipmi_si tpm_tis sg
>> tpm_tis_core tpm ipmi_devintf iTCO_vendor_support ib_uverbs pcspkr
>> watchdog ioatdma rng_core i7core_edac dca i5500_temp ipmi_msghandler
>> ib_umad ib_ipoib auth_rpcgss ib_cm sunrpc ib_core ip_tables x_tables
>> autofs4 algif_skcipher af_alg ext4 crc16 mbcache jbd2 crc32c_generic
>> dm_crypt dm_mod sr_mod cdrom sd_mod hid_generic usbhid hid
>> crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel mgag200
>> drm_vram_helper drm_ttm_helper i2c_algo_bit ttm qla2xxx drm_kms_helper
>> [ 2495.150298]  ata_generic drm ata_piix aesni_intel libata nvme_fc
>> uhci_hcd crypto_simd nvme_fabrics ehci_pci ehci_hcd megaraid_sas cryptd
>> nvme_core glue_helper usbcore scsi_transport_fc lpc_ich mfd_core e1000e
>> scsi_mod i2c_i801 usb_common bnx2 ptp pps_core button
>> [ 2495.269073] CR2: 00000000000fab7d
>> [ 2495.273205] ---[ end trace 44f000505f987296 ]---
>> [ 2495.278647] RIP: 0010:_raw_spin_lock+0xc/0x20
>> [ 2495.283825] Code: 66 90 66 66 90 31 c0 ba ff 00 00 00 f0 0f b1 17 75
>> 05 48 89 d8 5b c3 e8 12 5f 91 ff eb f4 66 66 66 66 90 31 c0 ba 01 00 00
>> 00 <f0> 0f b1 17 75 01 c3 89 c6 e8 76 46 91 ff 66 90 c3 0f 1f 00 66 66
>> [ 2495.304287] RSP: 0018:ffffb949cca7fac0 EFLAGS: 00010246
>> [ 2495.310368] RAX: 0000000000000000 RBX: 00000000000faaf5 RCX:
>> 0000000000000000
>> [ 2495.318358] RDX: 0000000000000001 RSI: 0000000000000009 RDI:
>> 00000000000fab7d
>> [ 2495.326339] RBP: 00000000000fab7d R08: ffff99c8f12737b8 R09:
>> 000000000002c340
>> [ 2495.334314] R10: 000005dc16e333d8 R11: 0000000000000000 R12:
>> ffff99c903066bd0
>> [ 2495.342297] R13: ffff99c9014c7bc0 R14: 000000000000ffff R15:
>> ffff99c9014c70c8
>> [ 2495.350279] FS:  0000000000000000(0000) GS:ffff99ba03bc0000(0000)
>> knlGS:0000000000000000
>> [ 2495.359224] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> [ 2495.365825] CR2: 00000000000fab7d CR3: 0000000f38df2002 CR4:
>> 00000000000206e0
>>
>> -----8< cut here 8<-----
>>
>>
>> I also made some record of NFS traffic with tcpdump when the crashs are
>> occuring, but I didn't had time to analyse it. I can provide them if it
>> helps (but I might have to clean them up first to avoid some data
>> leaks :)
>>
>> After this happening a few times, I had one case where one of the few
>> NFSv4 client was involved, and I took as aworkaround to migrate the
>> NFSv4 share to XFS since I though it could be NFSv4 related. Now that I
>> looked at the kernel dumps, I think it's also NFSv3 related.
>>
>> Is it a kernel bug or am I missing something ?
>>
>> For now, I will switch back to XFS with a failover setup of my cluster,
>> cause I cannot afford this kind of crashes of a central file server.
>>
>>
>>
>> Cheers
>>
>>
>> Pierre
>>
>>
>> PS : I'll try to update with what I found in the PCAP files in some days
>>
>>
>>
>>>
>>> _______________________________________________
>>> Ocfs2-devel mailing list
>>> Ocfs2-devel at oss.oracle.com
>>> https://oss.oracle.com/mailman/listinfo/ocfs2-devel

  reply	other threads:[~2020-06-02 15:56 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-19 10:48 [Ocfs2-devel] NFS clients crash OCFS2 nodes (general protection fault: 0000 [#1] SMP PTI) Pierre Dinh-van
2020-05-20  2:52 ` Changwei Ge
2020-05-21 17:02   ` Pierre Dinh-van
2020-05-22  1:35     ` Changwei Ge
2020-05-20 22:01 ` herbert.van.den.bergh at oracle.com
2020-06-02 10:10   ` Pierre Dinh-van
2020-06-02 15:56     ` herbert.van.den.bergh at oracle.com [this message]
2020-06-12  1:36       ` Junxiao Bi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3873cb24-9db7-cc20-550a-ab82568464ca@oracle.com \
    --to=herbert.van.den.bergh@oracle.com \
    --cc=ocfs2-devel@oss.oracle.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).