* 2.4.17 SMP hangs ..
@ 2002-11-21 5:28 Manish Lachwani
2002-11-21 6:03 ` Andrew Morton
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Manish Lachwani @ 2002-11-21 5:28 UTC (permalink / raw)
To: linux-kernel; +Cc: Manish Lachwani
I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
drives in parallel. However, the hangs only occur when the I/O rate from
vmstat is high:
bash# vmstat 1
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
1 0 0 0 815800 3656 102612 0 0 163 1086 129 961 6 29
65
0 0 0 0 813096 3656 102992 0 0 363 0 265 928 44 26
31
0 0 0 0 813172 3656 102996 0 0 4 0 180 99 0 1
99
5 2 1 0 809476 3656 103044 0 0 44 0 206 274 18 11
71
0 12 0 0 802040 3656 103072 0 0 18 2 296 745 47 33
21
0 8 0 0 800696 3660 103152 0 0 34 244 349 501 10 5
85
0 8 0 0 800696 3660 103152 0 0 0 0 187 106 1 0
99
2 5 1 0 795864 3660 103200 0 0 13 45 278 717 14 29
57
1 0 0 0 795592 3660 103248 0 0 107 46 502 663 7 7
86
1 0 0 0 795592 3660 103248 0 0 0 0 184 95 0 1
99
1 0 0 0 795596 3660 103244 0 0 4 1 191 139 0 1
99
6 5 3 0 756932 3660 140192 0 0 194 36721 464 232 7 87
6
11 0 1 0 723276 3660 173784 0 0 0 33718 681 1560 0 39
61
At that point the system hangs. The system consists of a 4-port and a 8-port
3ware controllers on an Intel 21154 bridge with 12 maxtor drives. When the
IO rate is lower:
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
0 0 0 0 811888 3828 103476 0 0 0 0 172 91 0 1
99
0 0 0 0 811888 3828 103476 0 0 0 0 184 93 0 1
99
0 0 0 0 811888 3828 103476 0 0 0 0 171 153 0 1
99
0 0 0 0 811852 3828 103512 0 0 0 170 171 85 0 1
99
0 0 0 0 811852 3828 103512 0 0 0 0 174 95 1 0
99
0 0 0 0 811852 3828 103512 0 0 0 0 173 147 0 1
99
0 0 0 0 811852 3828 103512 0 0 0 0 176 98 0 1
99
0 0 0 0 811852 3828 103512 0 0 0 1 175 96 0 1
99
1 0 0 0 811852 3828 103512 0 0 0 0 173 110 1 3
96
1 0 0 0 811704 3828 103516 0 0 0 0 179 145 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 194 121 1 1
98
0 0 0 0 811688 3828 103516 0 0 0 19 186 119 1 1
98
0 0 0 0 811688 3828 103516 0 0 0 0 174 149 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 172 86 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 175 96 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 1 171 149 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 173 91 0 1
99
0 0 1 0 811688 3828 103516 0 0 0 0 173 86 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 179 154 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 1 174 91 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 179 98 1 0
99
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
1 0 0 0 811688 3828 103516 0 0 0 0 174 100 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 184 137 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 1 171 89 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 173 95 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 176 150 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 175 121 3 0
97
0 0 0 0 811688 3828 103516 0 0 0 15 171 116 3 0
97
0 0 0 0 811688 3828 103516 0 0 0 0 179 149 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 173 88 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 171 88 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 15 172 149 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 174 98 0 1
99
1 0 0 0 811688 3828 103516 0 0 0 0 178 127 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 1 179 153 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 173 88 0 1
99
0 0 0 0 811688 3828 103516 0 0 0 0 175 158 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 1 175 96 1 0
99
0 0 0 0 811688 3828 103516 0 0 0 0 171 90 0 1
99
0 0 0 0 811924 3828 103516 0 0 0 0 173 204 1 5
94
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
1 0 0 0 811924 3828 103516 0 0 0 0 174 90 1 0
99
0 0 0 0 811924 3828 103516 0 0 0 24 182 93 1 0
99
1 0 0 0 811924 3828 103516 0 0 0 0 173 142 0 1
99
0 0 0 0 811924 3828 103516 0 0 0 0 171 93 1 0
99
0 0 0 0 811924 3828 103516 0 0 0 0 175 94 0 1
99
1 0 0 0 811924 3828 103516 0 0 0 2 173 92 1 0
99
0 0 0 0 811924 3828 103516 0 0 0 0 175 151 1 0
99
0 0 0 0 811924 3828 103516 0 0 0 0 173 87 1 0
99
0 0 1 0 811924 3828 103516 0 0 0 0 173 89 1 0
99
0 0 0 0 811924 3828 103516 0 0 0 1 171 142 0 1
99
bash# vmstat 1
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
5 0 1 0 815896 3656 102228 0 0 156 1076 129 917 6 34
60
2 0 1 0 812860 3656 102992 0 0 737 0 252 1015 36 40
25
0 0 0 0 813104 3656 102992 0 0 0 0 187 129 2 1
97
0 0 0 0 813180 3656 102996 0 0 4 0 168 93 0 1
99
6 8 1 0 802392 3656 103220 0 0 222 0 251 757 62 37
1
0 10 0 0 804588 3684 103252 0 0 133 40 214 871 50 43
8
2 8 1 0 804324 3712 103272 0 0 162 40 222 817 43 40
18
9 4 0 0 805400 3712 103288 0 0 4 0 196 276 13 8
79
9 0 1 0 804724 3812 103348 0 0 644 96 297 1255 41 59
0
14 0 1 0 804144 3816 103344 0 0 0 0 167 888 56 44
0
10 0 1 0 804448 3820 103340 0 0 0 0 171 873 57 43
0
0 1 0 0 812288 3828 103436 0 0 97 0 222 1051 53 42
5
0 0 0 0 811868 3828 103476 0 0 84 0 395 429 0 3
97
1 0 0 0 811868 3828 103476 0 0 0 0 167 91 0 1
99
1 0 0 0 811868 3828 103476 0 0 0 0 167 141 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 177 100 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 171 96 0 1
99
1 0 0 0 811868 3828 103476 0 0 0 0 171 132 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 170 100 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 167 90 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 171 93 1 0
99
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
0 0 0 0 811868 3828 103476 0 0 0 0 170 148 1 0
99
0 0 0 0 811868 3828 103476 0 0 0 0 176 83 0 1
99
0 0 0 0 811868 3828 103476 0 0 0 0 167 86 0 1
99
0 0 0 0 811868 3828 103476 0 0 0 0 169 146 0 1
99
0 0 0 0 811832 3828 103512 0 0 0 173 168 84 1 0
99
0 0 0 0 811832 3828 103512 0 0 0 0 178 104 1 0
99
0 0 0 0 811832 3828 103512 0 0 0 0 167 141 0 1
99
0 0 0 0 811832 3828 103512 0 0 0 0 173 92 1 2
97
0 0 0 0 811832 3828 103512 0 0 0 1 169 89 0 2
98
1 0 0 0 811832 3828 103512 0 0 0 0 173 138 2 3
95
1 0 0 0 811684 3828 103516 0 0 0 0 183 132 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 178 103 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 19 180 108 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 169 148 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 169 88 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 171 94 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 1 170 147 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 168 94 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 171 95 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 173 175 3 0
97
0 0 0 0 811668 3828 103516 0 0 0 15 169 89 1 0
99
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
0 0 0 0 811668 3828 103516 0 0 0 0 168 87 1 0
99
1 0 0 0 811668 3828 103516 0 0 0 0 178 156 2 1
97
0 0 0 0 811668 3828 103516 0 0 0 0 175 122 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 15 167 86 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 171 92 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 177 151 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 169 93 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 1 169 86 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 167 146 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 172 95 0 1
99
0 0 0 0 811668 3828 103516 0 0 0 0 167 84 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 1 178 189 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 171 89 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 0 171 89 0 1
99
1 0 0 0 811668 3828 103516 0 0 0 0 171 124 1 0
99
0 0 0 0 811668 3828 103516 0 0 0 1 183 127 0 1
99
bash# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/ram1 125011 85304 33154 73% /
coreserver:/var/cores
74858752 475144 70580928 1%
/var/.automount/cores
/dev/sdj1 7827172 24 7429540 1% /disks/disk10.1
/dev/sdi1 7827172 24 7429540 1% /disks/disk9.1
/dev/sdl1 7827172 24 7429540 1% /disks/disk12.1
/dev/sdk1 7827172 24 7429540 1% /disks/disk11.1
/dev/sdh1 7827172 24 7429540 1% /disks/disk8.1
/dev/sdf1 7827172 24 7429540 1% /disks/disk6.1
/dev/sde1 7827172 24 7429540 1% /disks/disk5.1
/dev/sdb1 7827172 24 7429540 1% /disks/disk2.1
/dev/sdg1 7827172 24 7429540 1% /disks/disk7.1
/dev/sda1 7827172 24 7429540 1% /disks/disk1.1
/dev/sdd1 7827172 24 7429540 1% /disks/disk4.1
/dev/sdc1 7827172 24 7429540 1% /disks/disk3.1
bash#
bash# vmstat 1
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
0 0 0 0 755440 3924 105192 0 0 83 8108 353 655 5 25
69
0 0 1 0 755440 3924 105192 0 0 0 0 167 70 0 0
100
0 0 0 0 755440 3924 105192 0 0 0 24 159 82 0 1
99
0 0 0 0 755052 3924 105192 0 0 0 0 162 133 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 0 156 68 0 1
99
0 0 0 0 755052 3924 105192 0 0 0 0 155 59 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 1 157 124 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 0 174 82 0 1
99
0 0 0 0 755052 3924 105192 0 0 0 0 161 73 0 0
100
1 0 0 0 755052 3924 105192 0 0 0 0 159 101 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 1 155 92 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 0 155 57 0 0
100
0 0 0 0 755052 3924 105192 0 0 0 0 155 67 1 0
99
0 0 0 0 755052 3924 105192 0 0 0 0 155 112 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 6 157 67 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 0 155 62 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 0 157 128 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 0 160 66 0 1
99
0 0 0 0 754440 3924 105192 0 0 0 1 157 72 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 0 155 117 0 0
100
0 0 0 0 754440 3924 105192 0 0 0 0 155 71 0 0
100
procs memory swap io system
cpu
r b w swpd free buff cache si so bi bo in cs us sy
id
0 0 0 0 754440 3924 105192 0 0 0 0 157 66 0 0
100
1 0 0 0 754440 3924 105192 0 0 0 1 166 114 0 1
99
0 0 0 0 754440 3924 105192 0 0 0 0 158 93 0 0
100
there are no hangs. On startup, I am doing parallel mje2fs accross all the
drives. 3ware 4-port controller shows that LEDs are ON. I have tried
replacing the controllers but that also does not help ...
Thanks
Manish
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: 2.4.17 SMP hangs ..
2002-11-21 5:28 2.4.17 SMP hangs Manish Lachwani
@ 2002-11-21 6:03 ` Andrew Morton
2002-11-26 15:22 ` Theodore Ts'o
2002-11-21 6:33 ` Andre Hedrick
2002-11-22 12:00 ` William Lee Irwin III
2 siblings, 1 reply; 7+ messages in thread
From: Andrew Morton @ 2002-11-21 6:03 UTC (permalink / raw)
To: Manish Lachwani; +Cc: linux-kernel
Manish Lachwani wrote:
>
> I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
> drives in parallel. However, the hangs only occur when the I/O rate from
> vmstat is high:
>
Quite possibly it has not hung. You just need to wait half an
hour or so.
The algorithm isn't very good.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: 2.4.17 SMP hangs ..
2002-11-21 5:28 2.4.17 SMP hangs Manish Lachwani
2002-11-21 6:03 ` Andrew Morton
@ 2002-11-21 6:33 ` Andre Hedrick
2002-11-22 12:00 ` William Lee Irwin III
2 siblings, 0 replies; 7+ messages in thread
From: Andre Hedrick @ 2002-11-21 6:33 UTC (permalink / raw)
To: Manish Lachwani; +Cc: linux-kernel
The problem may be associated w/ the interrupt mapping over the bridge
riser. I have seen this before and can reproduce it regardless of the
system.
Cheers,
Andre Hedrick
LAD Storage Consulting Group
On Wed, 20 Nov 2002, Manish Lachwani wrote:
> I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
> drives in parallel. However, the hangs only occur when the I/O rate from
> vmstat is high:
>
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 1 0 0 0 815800 3656 102612 0 0 163 1086 129 961 6 29
> 65
> 0 0 0 0 813096 3656 102992 0 0 363 0 265 928 44 26
> 31
> 0 0 0 0 813172 3656 102996 0 0 4 0 180 99 0 1
> 99
> 5 2 1 0 809476 3656 103044 0 0 44 0 206 274 18 11
> 71
> 0 12 0 0 802040 3656 103072 0 0 18 2 296 745 47 33
> 21
> 0 8 0 0 800696 3660 103152 0 0 34 244 349 501 10 5
> 85
> 0 8 0 0 800696 3660 103152 0 0 0 0 187 106 1 0
> 99
> 2 5 1 0 795864 3660 103200 0 0 13 45 278 717 14 29
> 57
> 1 0 0 0 795592 3660 103248 0 0 107 46 502 663 7 7
> 86
> 1 0 0 0 795592 3660 103248 0 0 0 0 184 95 0 1
> 99
> 1 0 0 0 795596 3660 103244 0 0 4 1 191 139 0 1
> 99
> 6 5 3 0 756932 3660 140192 0 0 194 36721 464 232 7 87
> 6
> 11 0 1 0 723276 3660 173784 0 0 0 33718 681 1560 0 39
> 61
>
> At that point the system hangs. The system consists of a 4-port and a 8-port
> 3ware controllers on an Intel 21154 bridge with 12 maxtor drives. When the
> IO rate is lower:
>
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 0 0 0 0 811888 3828 103476 0 0 0 0 172 91 0 1
> 99
> 0 0 0 0 811888 3828 103476 0 0 0 0 184 93 0 1
> 99
> 0 0 0 0 811888 3828 103476 0 0 0 0 171 153 0 1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 170 171 85 0 1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 174 95 1 0
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 173 147 0 1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 176 98 0 1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 1 175 96 0 1
> 99
> 1 0 0 0 811852 3828 103512 0 0 0 0 173 110 1 3
> 96
> 1 0 0 0 811704 3828 103516 0 0 0 0 179 145 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 194 121 1 1
> 98
> 0 0 0 0 811688 3828 103516 0 0 0 19 186 119 1 1
> 98
> 0 0 0 0 811688 3828 103516 0 0 0 0 174 149 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 172 86 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 96 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 171 149 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 91 0 1
> 99
> 0 0 1 0 811688 3828 103516 0 0 0 0 173 86 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 154 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 174 91 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 98 1 0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 1 0 0 0 811688 3828 103516 0 0 0 0 174 100 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 184 137 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 171 89 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 95 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 176 150 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 121 3 0
> 97
> 0 0 0 0 811688 3828 103516 0 0 0 15 171 116 3 0
> 97
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 149 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 88 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 15 172 149 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 174 98 0 1
> 99
> 1 0 0 0 811688 3828 103516 0 0 0 0 178 127 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 179 153 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 88 0 1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 158 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 175 96 1 0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 90 0 1
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 173 204 1 5
> 94
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 1 0 0 0 811924 3828 103516 0 0 0 0 174 90 1 0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 24 182 93 1 0
> 99
> 1 0 0 0 811924 3828 103516 0 0 0 0 173 142 0 1
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 171 93 1 0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 175 94 0 1
> 99
> 1 0 0 0 811924 3828 103516 0 0 0 2 173 92 1 0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 175 151 1 0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 173 87 1 0
> 99
> 0 0 1 0 811924 3828 103516 0 0 0 0 173 89 1 0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 1 171 142 0 1
> 99
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 5 0 1 0 815896 3656 102228 0 0 156 1076 129 917 6 34
> 60
> 2 0 1 0 812860 3656 102992 0 0 737 0 252 1015 36 40
> 25
> 0 0 0 0 813104 3656 102992 0 0 0 0 187 129 2 1
> 97
> 0 0 0 0 813180 3656 102996 0 0 4 0 168 93 0 1
> 99
> 6 8 1 0 802392 3656 103220 0 0 222 0 251 757 62 37
> 1
> 0 10 0 0 804588 3684 103252 0 0 133 40 214 871 50 43
> 8
> 2 8 1 0 804324 3712 103272 0 0 162 40 222 817 43 40
> 18
> 9 4 0 0 805400 3712 103288 0 0 4 0 196 276 13 8
> 79
> 9 0 1 0 804724 3812 103348 0 0 644 96 297 1255 41 59
> 0
> 14 0 1 0 804144 3816 103344 0 0 0 0 167 888 56 44
> 0
> 10 0 1 0 804448 3820 103340 0 0 0 0 171 873 57 43
> 0
> 0 1 0 0 812288 3828 103436 0 0 97 0 222 1051 53 42
> 5
> 0 0 0 0 811868 3828 103476 0 0 84 0 395 429 0 3
> 97
> 1 0 0 0 811868 3828 103476 0 0 0 0 167 91 0 1
> 99
> 1 0 0 0 811868 3828 103476 0 0 0 0 167 141 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 177 100 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 171 96 0 1
> 99
> 1 0 0 0 811868 3828 103476 0 0 0 0 171 132 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 170 100 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 167 90 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 171 93 1 0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 0 0 0 0 811868 3828 103476 0 0 0 0 170 148 1 0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 176 83 0 1
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 167 86 0 1
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 169 146 0 1
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 173 168 84 1 0
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 178 104 1 0
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 167 141 0 1
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 173 92 1 2
> 97
> 0 0 0 0 811832 3828 103512 0 0 0 1 169 89 0 2
> 98
> 1 0 0 0 811832 3828 103512 0 0 0 0 173 138 2 3
> 95
> 1 0 0 0 811684 3828 103516 0 0 0 0 183 132 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 178 103 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 19 180 108 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 148 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 88 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 94 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 170 147 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 168 94 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 95 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 173 175 3 0
> 97
> 0 0 0 0 811668 3828 103516 0 0 0 15 169 89 1 0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 0 0 0 0 811668 3828 103516 0 0 0 0 168 87 1 0
> 99
> 1 0 0 0 811668 3828 103516 0 0 0 0 178 156 2 1
> 97
> 0 0 0 0 811668 3828 103516 0 0 0 0 175 122 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 15 167 86 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 92 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 177 151 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 93 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 169 86 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 167 146 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 172 95 0 1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 167 84 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 178 189 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 89 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 89 0 1
> 99
> 1 0 0 0 811668 3828 103516 0 0 0 0 171 124 1 0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 183 127 0 1
> 99
>
> bash# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/ram1 125011 85304 33154 73% /
> coreserver:/var/cores
> 74858752 475144 70580928 1%
> /var/.automount/cores
> /dev/sdj1 7827172 24 7429540 1% /disks/disk10.1
> /dev/sdi1 7827172 24 7429540 1% /disks/disk9.1
> /dev/sdl1 7827172 24 7429540 1% /disks/disk12.1
> /dev/sdk1 7827172 24 7429540 1% /disks/disk11.1
> /dev/sdh1 7827172 24 7429540 1% /disks/disk8.1
> /dev/sdf1 7827172 24 7429540 1% /disks/disk6.1
> /dev/sde1 7827172 24 7429540 1% /disks/disk5.1
> /dev/sdb1 7827172 24 7429540 1% /disks/disk2.1
> /dev/sdg1 7827172 24 7429540 1% /disks/disk7.1
> /dev/sda1 7827172 24 7429540 1% /disks/disk1.1
> /dev/sdd1 7827172 24 7429540 1% /disks/disk4.1
> /dev/sdc1 7827172 24 7429540 1% /disks/disk3.1
> bash#
>
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 0 0 0 0 755440 3924 105192 0 0 83 8108 353 655 5 25
> 69
> 0 0 1 0 755440 3924 105192 0 0 0 0 167 70 0 0
> 100
> 0 0 0 0 755440 3924 105192 0 0 0 24 159 82 0 1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 162 133 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 156 68 0 1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 59 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 1 157 124 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 174 82 0 1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 161 73 0 0
> 100
> 1 0 0 0 755052 3924 105192 0 0 0 0 159 101 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 1 155 92 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 57 0 0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 67 1 0
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 112 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 6 157 67 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 62 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 157 128 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 160 66 0 1
> 99
> 0 0 0 0 754440 3924 105192 0 0 0 1 157 72 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 117 0 0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 71 0 0
> 100
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us sy
> id
> 0 0 0 0 754440 3924 105192 0 0 0 0 157 66 0 0
> 100
> 1 0 0 0 754440 3924 105192 0 0 0 1 166 114 0 1
> 99
> 0 0 0 0 754440 3924 105192 0 0 0 0 158 93 0 0
> 100
>
> there are no hangs. On startup, I am doing parallel mje2fs accross all the
> drives. 3ware 4-port controller shows that LEDs are ON. I have tried
> replacing the controllers but that also does not help ...
>
> Thanks
> Manish
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: 2.4.17 SMP hangs ..
2002-11-21 5:28 2.4.17 SMP hangs Manish Lachwani
2002-11-21 6:03 ` Andrew Morton
2002-11-21 6:33 ` Andre Hedrick
@ 2002-11-22 12:00 ` William Lee Irwin III
2 siblings, 0 replies; 7+ messages in thread
From: William Lee Irwin III @ 2002-11-22 12:00 UTC (permalink / raw)
To: Manish Lachwani; +Cc: linux-kernel
On Wed, Nov 20, 2002 at 09:28:06PM -0800, Manish Lachwani wrote:
> I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
> drives in parallel. However, the hangs only occur when the I/O rate from
> vmstat is high:
This is the BKL/irqlock/io_request_lock contention issue. 2.5.x is
the only answer here: it has per-queue locking and has removed the
irqlock from the BKL.
Bill
^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: 2.4.17 SMP hangs ..
@ 2002-11-22 22:30 Manish Lachwani
0 siblings, 0 replies; 7+ messages in thread
From: Manish Lachwani @ 2002-11-22 22:30 UTC (permalink / raw)
To: 'Andre Hedrick', Manish Lachwani; +Cc: linux-kernel
How can I conveniently produce this hang? Can you give me an example?
Thanks
-Manish
-----Original Message-----
From: Andre Hedrick [mailto:andre@linux-ide.org]
Sent: Wednesday, November 20, 2002 10:33 PM
To: Manish Lachwani
Cc: linux-kernel@vger.kernel.org
Subject: Re: 2.4.17 SMP hangs ..
The problem may be associated w/ the interrupt mapping over the bridge
riser. I have seen this before and can reproduce it regardless of the
system.
Cheers,
Andre Hedrick
LAD Storage Consulting Group
On Wed, 20 Nov 2002, Manish Lachwani wrote:
> I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros
12
> drives in parallel. However, the hangs only occur when the I/O rate from
> vmstat is high:
>
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 1 0 0 0 815800 3656 102612 0 0 163 1086 129 961 6
29
> 65
> 0 0 0 0 813096 3656 102992 0 0 363 0 265 928 44
26
> 31
> 0 0 0 0 813172 3656 102996 0 0 4 0 180 99 0
1
> 99
> 5 2 1 0 809476 3656 103044 0 0 44 0 206 274 18
11
> 71
> 0 12 0 0 802040 3656 103072 0 0 18 2 296 745 47
33
> 21
> 0 8 0 0 800696 3660 103152 0 0 34 244 349 501 10
5
> 85
> 0 8 0 0 800696 3660 103152 0 0 0 0 187 106 1
0
> 99
> 2 5 1 0 795864 3660 103200 0 0 13 45 278 717 14
29
> 57
> 1 0 0 0 795592 3660 103248 0 0 107 46 502 663 7
7
> 86
> 1 0 0 0 795592 3660 103248 0 0 0 0 184 95 0
1
> 99
> 1 0 0 0 795596 3660 103244 0 0 4 1 191 139 0
1
> 99
> 6 5 3 0 756932 3660 140192 0 0 194 36721 464 232 7
87
> 6
> 11 0 1 0 723276 3660 173784 0 0 0 33718 681 1560 0
39
> 61
>
> At that point the system hangs. The system consists of a 4-port and a
8-port
> 3ware controllers on an Intel 21154 bridge with 12 maxtor drives. When the
> IO rate is lower:
>
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 0 0 0 0 811888 3828 103476 0 0 0 0 172 91 0
1
> 99
> 0 0 0 0 811888 3828 103476 0 0 0 0 184 93 0
1
> 99
> 0 0 0 0 811888 3828 103476 0 0 0 0 171 153 0
1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 170 171 85 0
1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 174 95 1
0
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 173 147 0
1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 0 176 98 0
1
> 99
> 0 0 0 0 811852 3828 103512 0 0 0 1 175 96 0
1
> 99
> 1 0 0 0 811852 3828 103512 0 0 0 0 173 110 1
3
> 96
> 1 0 0 0 811704 3828 103516 0 0 0 0 179 145 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 194 121 1
1
> 98
> 0 0 0 0 811688 3828 103516 0 0 0 19 186 119 1
1
> 98
> 0 0 0 0 811688 3828 103516 0 0 0 0 174 149 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 172 86 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 96 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 171 149 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 91 0
1
> 99
> 0 0 1 0 811688 3828 103516 0 0 0 0 173 86 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 154 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 174 91 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 98 1
0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 1 0 0 0 811688 3828 103516 0 0 0 0 174 100 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 184 137 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 171 89 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 95 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 176 150 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 121 3
0
> 97
> 0 0 0 0 811688 3828 103516 0 0 0 15 171 116 3
0
> 97
> 0 0 0 0 811688 3828 103516 0 0 0 0 179 149 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 88 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 15 172 149 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 174 98 0
1
> 99
> 1 0 0 0 811688 3828 103516 0 0 0 0 178 127 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 179 153 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 88 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 173 88 0
1
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 175 158 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 1 175 96 1
0
> 99
> 0 0 0 0 811688 3828 103516 0 0 0 0 171 90 0
1
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 173 204 1
5
> 94
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 1 0 0 0 811924 3828 103516 0 0 0 0 174 90 1
0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 24 182 93 1
0
> 99
> 1 0 0 0 811924 3828 103516 0 0 0 0 173 142 0
1
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 171 93 1
0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 175 94 0
1
> 99
> 1 0 0 0 811924 3828 103516 0 0 0 2 173 92 1
0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 175 151 1
0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 0 173 87 1
0
> 99
> 0 0 1 0 811924 3828 103516 0 0 0 0 173 89 1
0
> 99
> 0 0 0 0 811924 3828 103516 0 0 0 1 171 142 0
1
> 99
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 5 0 1 0 815896 3656 102228 0 0 156 1076 129 917 6
34
> 60
> 2 0 1 0 812860 3656 102992 0 0 737 0 252 1015 36
40
> 25
> 0 0 0 0 813104 3656 102992 0 0 0 0 187 129 2
1
> 97
> 0 0 0 0 813180 3656 102996 0 0 4 0 168 93 0
1
> 99
> 6 8 1 0 802392 3656 103220 0 0 222 0 251 757 62
37
> 1
> 0 10 0 0 804588 3684 103252 0 0 133 40 214 871 50
43
> 8
> 2 8 1 0 804324 3712 103272 0 0 162 40 222 817 43
40
> 18
> 9 4 0 0 805400 3712 103288 0 0 4 0 196 276 13
8
> 79
> 9 0 1 0 804724 3812 103348 0 0 644 96 297 1255 41
59
> 0
> 14 0 1 0 804144 3816 103344 0 0 0 0 167 888 56
44
> 0
> 10 0 1 0 804448 3820 103340 0 0 0 0 171 873 57
43
> 0
> 0 1 0 0 812288 3828 103436 0 0 97 0 222 1051 53
42
> 5
> 0 0 0 0 811868 3828 103476 0 0 84 0 395 429 0
3
> 97
> 1 0 0 0 811868 3828 103476 0 0 0 0 167 91 0
1
> 99
> 1 0 0 0 811868 3828 103476 0 0 0 0 167 141 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 177 100 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 171 96 0
1
> 99
> 1 0 0 0 811868 3828 103476 0 0 0 0 171 132 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 170 100 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 167 90 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 171 93 1
0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 0 0 0 0 811868 3828 103476 0 0 0 0 170 148 1
0
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 176 83 0
1
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 167 86 0
1
> 99
> 0 0 0 0 811868 3828 103476 0 0 0 0 169 146 0
1
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 173 168 84 1
0
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 178 104 1
0
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 167 141 0
1
> 99
> 0 0 0 0 811832 3828 103512 0 0 0 0 173 92 1
2
> 97
> 0 0 0 0 811832 3828 103512 0 0 0 1 169 89 0
2
> 98
> 1 0 0 0 811832 3828 103512 0 0 0 0 173 138 2
3
> 95
> 1 0 0 0 811684 3828 103516 0 0 0 0 183 132 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 178 103 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 19 180 108 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 148 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 88 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 94 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 170 147 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 168 94 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 95 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 173 175 3
0
> 97
> 0 0 0 0 811668 3828 103516 0 0 0 15 169 89 1
0
> 99
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 0 0 0 0 811668 3828 103516 0 0 0 0 168 87 1
0
> 99
> 1 0 0 0 811668 3828 103516 0 0 0 0 178 156 2
1
> 97
> 0 0 0 0 811668 3828 103516 0 0 0 0 175 122 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 15 167 86 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 92 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 177 151 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 169 93 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 169 86 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 167 146 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 172 95 0
1
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 167 84 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 178 189 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 89 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 0 171 89 0
1
> 99
> 1 0 0 0 811668 3828 103516 0 0 0 0 171 124 1
0
> 99
> 0 0 0 0 811668 3828 103516 0 0 0 1 183 127 0
1
> 99
>
> bash# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/ram1 125011 85304 33154 73% /
> coreserver:/var/cores
> 74858752 475144 70580928 1%
> /var/.automount/cores
> /dev/sdj1 7827172 24 7429540 1% /disks/disk10.1
> /dev/sdi1 7827172 24 7429540 1% /disks/disk9.1
> /dev/sdl1 7827172 24 7429540 1% /disks/disk12.1
> /dev/sdk1 7827172 24 7429540 1% /disks/disk11.1
> /dev/sdh1 7827172 24 7429540 1% /disks/disk8.1
> /dev/sdf1 7827172 24 7429540 1% /disks/disk6.1
> /dev/sde1 7827172 24 7429540 1% /disks/disk5.1
> /dev/sdb1 7827172 24 7429540 1% /disks/disk2.1
> /dev/sdg1 7827172 24 7429540 1% /disks/disk7.1
> /dev/sda1 7827172 24 7429540 1% /disks/disk1.1
> /dev/sdd1 7827172 24 7429540 1% /disks/disk4.1
> /dev/sdc1 7827172 24 7429540 1% /disks/disk3.1
> bash#
>
>
> bash# vmstat 1
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 0 0 0 0 755440 3924 105192 0 0 83 8108 353 655 5
25
> 69
> 0 0 1 0 755440 3924 105192 0 0 0 0 167 70 0
0
> 100
> 0 0 0 0 755440 3924 105192 0 0 0 24 159 82 0
1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 162 133 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 156 68 0
1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 59 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 1 157 124 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 174 82 0
1
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 161 73 0
0
> 100
> 1 0 0 0 755052 3924 105192 0 0 0 0 159 101 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 1 155 92 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 57 0
0
> 100
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 67 1
0
> 99
> 0 0 0 0 755052 3924 105192 0 0 0 0 155 112 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 6 157 67 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 62 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 157 128 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 160 66 0
1
> 99
> 0 0 0 0 754440 3924 105192 0 0 0 1 157 72 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 117 0
0
> 100
> 0 0 0 0 754440 3924 105192 0 0 0 0 155 71 0
0
> 100
> procs memory swap io system
> cpu
> r b w swpd free buff cache si so bi bo in cs us
sy
> id
> 0 0 0 0 754440 3924 105192 0 0 0 0 157 66 0
0
> 100
> 1 0 0 0 754440 3924 105192 0 0 0 1 166 114 0
1
> 99
> 0 0 0 0 754440 3924 105192 0 0 0 0 158 93 0
0
> 100
>
> there are no hangs. On startup, I am doing parallel mje2fs accross all the
> drives. 3ware 4-port controller shows that LEDs are ON. I have tried
> replacing the controllers but that also does not help ...
>
> Thanks
> Manish
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: 2.4.17 SMP hangs ..
2002-11-21 6:03 ` Andrew Morton
@ 2002-11-26 15:22 ` Theodore Ts'o
2002-11-26 16:13 ` Andrew Morton
0 siblings, 1 reply; 7+ messages in thread
From: Theodore Ts'o @ 2002-11-26 15:22 UTC (permalink / raw)
To: Andrew Morton; +Cc: Manish Lachwani, linux-kernel
On Wed, Nov 20, 2002 at 10:03:50PM -0800, Andrew Morton wrote:
> Manish Lachwani wrote:
> >
> > I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
> > drives in parallel. However, the hangs only occur when the I/O rate from
> > vmstat is high:
> >
>
> Quite possibly it has not hung. You just need to wait half an
> hour or so.
>
> The algorithm isn't very good.
[Catching up lkml mail after the IETF meeting....]
Try setting the environment variable "MKE2FS_SYNC" to a value such as
10. This will cause mke2fs to force a sync after writing out every 10
block groups worth of inode tables.
If this fixes the problem, then it means that the kernel isn't
handling write throttling correctly, and the system is thrashing
itself to death. Write thottleing is one of these kernel bugs which
gets fixed and broken in the kernel multiple times. I've considered
making MKE2FS_SYNC the default, but I haven't, mainly because current
behaviour is a great way of pointing out this write throttling bugs in
the VM. (Stephen has fixed this bug multiple times over the years,
and he suggested that having a good test case for noticing when
someone has broken write throttling would be a Good Thing --- and it
seems to get broken fairly often, as people try to make improvements
to the VM layer.....)
- Ted
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: 2.4.17 SMP hangs ..
2002-11-26 15:22 ` Theodore Ts'o
@ 2002-11-26 16:13 ` Andrew Morton
0 siblings, 0 replies; 7+ messages in thread
From: Andrew Morton @ 2002-11-26 16:13 UTC (permalink / raw)
To: Theodore Ts'o; +Cc: Manish Lachwani, linux-kernel
Theodore Ts'o wrote:
>
> On Wed, Nov 20, 2002 at 10:03:50PM -0800, Andrew Morton wrote:
> > Manish Lachwani wrote:
> > >
> > > I am seeing system hangs with 2.4.17 SMP kernel when doing mke2fs accros 12
> > > drives in parallel. However, the hangs only occur when the I/O rate from
> > > vmstat is high:
> > >
> >
> > Quite possibly it has not hung. You just need to wait half an
> > hour or so.
> >
> > The algorithm isn't very good.
>
> [Catching up lkml mail after the IETF meeting....]
>
> Try setting the environment variable "MKE2FS_SYNC" to a value such as
> 10. This will cause mke2fs to force a sync after writing out every 10
> block groups worth of inode tables.
That will fix it.
> If this fixes the problem, then it means that the kernel isn't
> handling write throttling correctly, and the system is thrashing
> itself to death.
Nah, it's __block_fsync(). That function has to write buffers against
a particular device. So it searches the global buffer LRU for 32 buffers
against the nominated device, drops the lock, writes the buffers, then
searches again.
So the search complexity is O(n*n/32). Which means that when you have a
lot of dirty buffers against different devices on the queue the CPU cost
simply explodes.
One workaround is to use sync rather than fsync - because sync uses NODEV
and doesn't have to search past buffers from uninteresting devices.
Another is to do what you've done.
Probably, just syncing the buffers at i_dirty_data_buffers would suffice.
That would fix it.
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2002-11-26 16:06 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-11-21 5:28 2.4.17 SMP hangs Manish Lachwani
2002-11-21 6:03 ` Andrew Morton
2002-11-26 15:22 ` Theodore Ts'o
2002-11-26 16:13 ` Andrew Morton
2002-11-21 6:33 ` Andre Hedrick
2002-11-22 12:00 ` William Lee Irwin III
-- strict thread matches above, loose matches on Subject: below --
2002-11-22 22:30 Manish Lachwani
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.