当前位置：首页 → 问答吧 → 急，内核崩溃了！！BUG: soft lockup - CPU#0 stuck for 11s!

急，内核崩溃了！！BUG: soft lockup - CPU#0 stuck for 11s!

时间：2008-01-02

来源：互联网

E文不太好，在kernel的maillist里也打到了相同的错误，是不是2.6.23-2.6.24的bug，怎样修复它呢！！
____________________________
以下是错误信息，出现以下时内核崩溃，只能硬重启！！

15:05:22 web kernel: BUG: soft lockup - CPU#0 stuck for 11s! [httpd:26440]
15:05:22 web kernel:
15:05:22 web kernel: Pid: 26440, comm:             httpd
15:05:22 web kernel: EIP: 0060:[<c0152d5c>] CPU: 0
15:05:22 web kernel: EIP is at find_get_pages_contig+0x63/0x73
15:05:22 web kernel:  EFLAGS: 00200216 Not tainted  (2.6.23-1-686 #1)
15:05:22 web kernel: EAX: 00000000 EBX: 00000003 ECX: 0000000a EDX: c2156aa0
15:05:22 web kernel: ESI: e3dc84dc EDI: 0000000b EBP: 00000003 DS: 007b ES: 007b FS: 00d8
15:05:22 web kernel: CR0: 8005003b CR2: b7413204 CR3: 0a9e0000 CR4: 000006d0
15:05:22 web kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
15:05:22 web kernel: DR6: ffff0ff0 DR7: 00000400
15:05:22 web kernel:  [<c0186674>] generic_file_splice_read+0x121/0x45e
15:05:22 web kernel:  [<c0152cbf>] find_get_pages+0x28/0x62
15:05:22 web kernel:  [<c0156e8f>] pagevec_lookup+0x1c/0x22
15:05:22 web kernel:  [<c015713e>] invalidate_inode_pages2_range+0x20e/0x226
15:05:22 web kernel:  [<f92da3ac>] nfs_sync_mapping_wait+0x1f3/0x1fd [nfs]
15:05:22 web kernel:  [<f92d3107>] nfs_invalidate_mapping_nolock+0x2b/0xc0 [nfs]
15:05:22 web kernel:  [<f92d3eb5>] nfs_revalidate_mapping+0x81/0x88 [nfs]
15:05:22 web kernel:  [<f92d1889>] nfs_file_splice_read+0x65/0x91 [nfs]
15:05:22 web kernel:  [<c027fab0>] __tcp_push_pending_frames+0x4b6/0x7d0
15:05:22 web kernel:  [<c0185965>] do_splice_to+0x74/0x7b
15:05:22 web kernel:  [<c0185bc5>] splice_direct_to_actor+0xb5/0x143
15:05:22 web kernel:  [<c01859e6>] direct_splice_actor+0x0/0x24
15:05:22 web kernel:  [<c0185c8b>] do_splice_direct+0x38/0x4d
15:05:22 web kernel:  [<c016d237>] do_sendfile+0x1cc/0x233
15:05:22 web kernel:  [<c016d2e8>] sys_sendfile64+0x4a/0x95
15:05:22 web kernel:  [<c0103e16>] sysenter_past_esp+0x6b/0xa1
15:05:22 web kernel:  [<c02a0000>] __unix_insert_socket+0x41/0x51
15:05:22 web kernel:  =======================
15:05:33 web kernel: BUG: soft lockup - CPU#0 stuck for 11s! [httpd:26440]
15:05:33 web kernel:
15:05:33 web kernel: Pid: 26440, comm:             httpd
15:05:33 web kernel: EIP: 0060:[<c0152d5c>] CPU: 0
15:05:33 web kernel: EIP is at find_get_pages_contig+0x63/0x73
15:05:33 web kernel:  EFLAGS: 00200216 Not tainted  (2.6.23-1-686 #1)
15:05:33 web kernel: EAX: 00000000 EBX: 00000003 ECX: 0000000a EDX: c2156aa0
15:05:33 web kernel: ESI: e3dc84dc EDI: 0000000b EBP: 00000003 DS: 007b ES: 007b FS: 00d8
15:05:33 web kernel: CR0: 8005003b CR2: b7413204 CR3: 0a9e0000 CR4: 000006d0
15:05:33 web kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
15:05:33 web kernel: DR6: ffff0ff0 DR7: 00000400
15:05:33 web kernel:  [<c0186674>] generic_file_splice_read+0x121/0x45e
15:05:33 web kernel:  [<c0152cbf>] find_get_pages+0x28/0x62
15:05:33 web kernel:  [<c0156e8f>] pagevec_lookup+0x1c/0x22
15:05:33 web kernel:  [<c015713e>] invalidate_inode_pages2_range+0x20e/0x226
15:05:33 web kernel:  [<f92da3ac>] nfs_sync_mapping_wait+0x1f3/0x1fd [nfs]
15:05:33 web kernel:  [<f92d3107>] nfs_invalidate_mapping_nolock+0x2b/0xc0 [nfs]
15:05:33 web kernel:  [<f92d3eb5>] nfs_revalidate_mapping+0x81/0x88 [nfs]
15:05:33 web kernel:  [<f92d1889>] nfs_file_splice_read+0x65/0x91 [nfs]
15:05:33 web kernel:  [<c027fab0>] __tcp_push_pending_frames+0x4b6/0x7d0
15:05:33 web kernel:  [<c0185965>] do_splice_to+0x74/0x7b
15:05:33 web kernel:  [<c0185bc5>] splice_direct_to_actor+0xb5/0x143
15:05:33 web kernel:  [<c01859e6>] direct_splice_actor+0x0/0x24
15:05:33 web kernel:  [<c0185c8b>] do_splice_direct+0x38/0x4d
15:05:33 web kernel:  [<c016d237>] do_sendfile+0x1cc/0x233
15:05:33 web kernel:  [<c016d2e8>] sys_sendfile64+0x4a/0x95
15:05:33 web kernel:  [<c0103e16>] sysenter_past_esp+0x6b/0xa1
15:05:33 web kernel:  [<c02a0000>] __unix_insert_socket+0x41/0x51
15:05:33 web kernel:  =======================
15:05:45 web kernel: BUG: soft lockup - CPU#0 stuck for 11s! [httpd:26440]
15:05:45 web kernel:
15:05:45 web kernel: Pid: 26440, comm:             httpd
15:05:45 web kernel: EIP: 0060:[<c0156c06>] CPU: 0
15:05:45 web kernel: EIP is at put_page+0x1a/0x9b
15:05:45 web kernel:  EFLAGS: 00200202 Not tainted  (2.6.23-1-686 #1)
15:05:45 web kernel: EAX: 80000000 EBX: c2156aa0 ECX: 000085f0 EDX: 00000000
15:05:45 web kernel: ESI: 00000003 EDI: 00000863 EBP: e3dc84dc DS: 007b ES: 007b FS: 00d8
15:05:45 web kernel: CR0: 8005003b CR2: b7413204 CR3: 0a9e0000 CR4: 000006d0
15:05:45 web kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
15:05:45 web kernel: DR6: ffff0ff0 DR7: 00000400
15:05:45 web kernel:  [<c018691d>] generic_file_splice_read+0x3ca/0x45e
15:05:45 web kernel:  [<c0152cbf>] find_get_pages+0x28/0x62
15:05:45 web kernel:  [<c0156e8f>] pagevec_lookup+0x1c/0x22
15:05:45 web kernel:  [<c015713e>] invalidate_inode_pages2_range+0x20e/0x226
15:05:45 web kernel:  [<f92da3ac>] nfs_sync_mapping_wait+0x1f3/0x1fd [nfs]
15:05:45 web kernel:  [<f92d3107>] nfs_invalidate_mapping_nolock+0x2b/0xc0 [nfs]
15:05:45 web kernel:  [<f92d3eb5>] nfs_revalidate_mapping+0x81/0x88 [nfs]
15:05:45 web kernel:  [<f92d1889>] nfs_file_splice_read+0x65/0x91 [nfs]
15:05:45 web kernel:  [<c027fab0>] __tcp_push_pending_frames+0x4b6/0x7d0
15:05:45 web kernel:  [<c0185965>] do_splice_to+0x74/0x7b
15:05:45 web kernel:  [<c0185bc5>] splice_direct_to_actor+0xb5/0x143
15:05:45 web kernel:  [<c01859e6>] direct_splice_actor+0x0/0x24
15:05:45 web kernel:  [<c0185c8b>] do_splice_direct+0x38/0x4d
15:05:45 web kernel:  [<c016d237>] do_sendfile+0x1cc/0x233
15:05:45 web kernel:  [<c016d2e8>] sys_sendfile64+0x4a/0x95
15:05:45 web kernel:  [<c0103e16>] sysenter_past_esp+0x6b/0xa1
15:05:45 web kernel:  [<c02a0000>] __unix_insert_socket+0x41/0x51
15:05:45 web kernel:  =======================
15:05:57 web kernel: BUG: soft lockup - CPU#0 stuck for 11s! [httpd:26440]
15:05:57 web kernel:
15:05:57 web kernel: Pid: 26440, comm:             httpd
15:05:57 web kernel: EIP: 0060:[<c0156c06>] CPU: 0
15:05:57 web kernel: EIP is at put_page+0x1a/0x9b
15:05:57 web kernel:  EFLAGS: 00200202 Not tainted  (2.6.23-1-686 #1)
15:05:57 web kernel: EAX: 80000020 EBX: c1e7b900 ECX: 000085f0 EDX: 00000000
15:05:57 web kernel: ESI: 00000003 EDI: 00000863 EBP: e3dc84dc DS: 007b ES: 007b FS: 00d8
15:05:57 web kernel: CR0: 8005003b CR2: b7413204 CR3: 0a9e0000 CR4: 000006d0
15:05:57 web kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
15:05:57 web kernel: DR6: ffff0ff0 DR7: 00000400
15:05:57 web kernel:  [<c018691d>] generic_file_splice_read+0x3ca/0x45e
15:05:57 web kernel:  [<c0152cbf>] find_get_pages+0x28/0x62
15:05:57 web kernel:  [<c0156e8f>] pagevec_lookup+0x1c/0x22
15:05:57 web kernel:  [<c015713e>] invalidate_inode_pages2_range+0x20e/0x226
15:05:57 web kernel:  [<f92da3ac>] nfs_sync_mapping_wait+0x1f3/0x1fd [nfs]
15:05:57 web kernel:  [<f92d3107>] nfs_invalidate_mapping_nolock+0x2b/0xc0 [nfs]
15:05:57 web kernel:  [<f92d3eb5>] nfs_revalidate_mapping+0x81/0x88 [nfs]
15:05:57 web kernel:  [<f92d1889>] nfs_file_splice_read+0x65/0x91 [nfs]
15:05:57 web kernel:  [<c027fab0>] __tcp_push_pending_frames+0x4b6/0x7d0
15:05:57 web kernel:  [<c0185965>] do_splice_to+0x74/0x7b
15:05:57 web kernel:  [<c0185bc5>] splice_direct_to_actor+0xb5/0x143
15:05:57 web kernel:  [<c01859e6>] direct_splice_actor+0x0/0x24
15:05:57 web kernel:  [<c0185c8b>] do_splice_direct+0x38/0x4d
15:05:57 web kernel:  [<c016d237>] do_sendfile+0x1cc/0x233
15:05:57 web kernel:  [<c016d2e8>] sys_sendfile64+0x4a/0x95
15:05:57 web kernel:  [<c0103e16>] sysenter_past_esp+0x6b/0xa1
15:05:57 web kernel:  [<c02a0000>] __unix_insert_socket+0x41/0x51
15:05:57 web kernel:  =======================
15:06:09 web kernel: BUG: soft lockup - CPU#0 stuck for 11s! [httpd:26440]
15:06:09 web kernel:
15:06:09 web kernel: Pid: 26440, comm:             httpd
15:06:09 web kernel: EIP: 0060:[<c0152d5c>] CPU: 0
15:06:09 web kernel: EIP is at find_get_pages_contig+0x63/0x73
15:06:09 web kernel:  EFLAGS: 00200216 Not tainted  (2.6.23-1-686 #1)
15:06:09 web kernel: EAX: 00000000 EBX: 00000003 ECX: 0000000a EDX: c2156aa0
15:06:09 web kernel: ESI: e3dc84dc EDI: 0000000b EBP: 00000003 DS: 007b ES: 007b FS: 00d8
15:06:09 web kernel: CR0: 8005003b CR2: b7413204 CR3: 0a9e0000 CR4: 000006d0
15:06:09 web kernel: DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
15:06:09 web kernel: DR6: ffff0ff0 DR7: 00000400
15:06:09 web kernel:  [<c0186674>] generic_file_splice_read+0x121/0x45e
15:06:09 web kernel:  [<c0152cbf>] find_get_pages+0x28/0x62
15:06:09 web kernel:  [<c0156e8f>] pagevec_lookup+0x1c/0x22
15:06:09 web kernel:  [<c015713e>] invalidate_inode_pages2_range+0x20e/0x226
15:06:09 web kernel:  [<f92da3ac>] nfs_sync_mapping_wait+0x1f3/0x1fd [nfs]
15:06:09 web kernel:  [<f92d3107>] nfs_invalidate_mapping_nolock+0x2b/0xc0 [nfs]
15:06:09 web kernel:  [<f92d3eb5>] nfs_revalidate_mapping+0x81/0x88 [nfs]
15:06:09 web kernel:  [<f92d1889>] nfs_file_splice_read+0x65/0x91 [nfs]
15:06:09 web kernel:  [<c027fab0>] __tcp_push_pending_frames+0x4b6/0x7d0
15:06:09 web kernel:  [<c0185965>] do_splice_to+0x74/0x7b
15:06:09 web kernel:  [<c0185bc5>] splice_direct_to_actor+0xb5/0x143
15:06:09 web kernel:  [<c01859e6>] direct_splice_actor+0x0/0x24
15:06:09 web kernel:  [<c0185c8b>] do_splice_direct+0x38/0x4d
15:06:09 web kernel:  [<c016d237>] do_sendfile+0x1cc/0x233
15:06:09 web kernel:  [<c016d2e8>] sys_sendfile64+0x4a/0x95
15:06:09 web kernel:  [<c0103e16>] sysenter_past_esp+0x6b/0xa1
15:06:09 web kernel:  [<c02a0000>] __unix_insert_socket+0x41/0x51
15:06:09 web kernel:  =======================

作者: mb 发布时间: 2008-01-02

不知道你使用的时2.6.23的那个版本,应该是2.6.23.3中有关于soft lockup有相关描述SCHED_FIFO:99 and looping
for more than 10 seconds.

看了看,有说nfs-utils的问题,有说内核的问题,2.6.23.8~9有人测试过,暂时能解决此问题.
不过看了一下kernel上提交的nfs的patch，觉得nfsr的bug的可能性大。

[ 本帖最后由 wysilly 于 2008-1-2 20:57 编辑 ]

作者: wysilly 发布时间: 2008-01-02

谢谢wysilly
，只为了sky2支持性，没办法，只好升级2.6.24-rc6试试

作者: mb 发布时间: 2008-01-03

看似发送死锁了, 被nmi的死锁检测查出来了

作者: fengmushu 发布时间: 2008-01-05

该问题解决了吗？

作者: dreamice 发布时间: 2011-02-14