---
title: 运行报错failed to bind memory，但应用仍可以正常结束
description: "在单节点或者多节点运行MPI作业，指定-bind-to core，报错“failed to bind memory”。"
url: https://www.hikunpeng.com/document/detail/zh/kunpenghpcs/hpckit/devg/userg_huaweimpi_0052.html
sourcePath: /source/zh/kunpenghpcs/hpckit/devg/userg_huaweimpi_0052.html
indexId: cb0b2a7caa93b438dea6f1381b37fadbc6ec252a9d718f587eb2c84421aa1c2a60
---
# 运行报错failed to bind memory，但应用仍可以正常结束

#### 现象描述

在单节点或者多节点运行MPI作业，指定-bind-to core，报错“failed to bind memory”。

```
--------------------------------------------------------------------------
WARNING: Open MPI tried to bind a process but failed.  This is a
warning only; your job will continue, though performance may
be degraded.
Local host:        node175
Application name:  /root/tqa/osu-micro-benchmarks-7.1-1/c/mpi/collective/blocking/osu_bcast
Error message:     failed to bind memory
Location:          rtc_hwloc.c:447
--------------------------------------------------------------------------
[node175:06049] 63 more processes have sent help message help-orte-odls-default.txt / memory not bound
[node175:06049] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
```


#### 可能原因

节点上没有插满内存条，导致部分NUMA node没有绑定内存，可以通过numactl --hardware命令检查。

```
available: 4 nodes (0-3)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
node 0 size: 31566 MB
node 0 free: 30548 MB
node 1 cpus: 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
node 1 size: 0 MB
node 1 free: 0 MB
node 2 cpus: 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
node 2 size: 31049 MB
node 2 free: 29424 MB
node 3 cpus: 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
node 3 size: 0 MB
node 3 free: 0 MB
node distances:
node   0   1   2   3
0:  10  12  35  37
1:  12  10  37  40
2:  35  37  10  12
3:  37  40  12  10
```

其中node1和node2均没有绑定内存。


#### 恢复步骤

1. 进程规模较小的情况下，可以尝试指定rankfile，避免进程绑定在没有内存的NUMA上。
2. 插满内存条。