收集网卡日志,sysinfo-snapshot,ibdiagnet

2023-07-28 16:30:48 昌辉 900


NVIDIA Mellanox 网卡收集日志 sysinfo-snapshot 方法。(基于Linux操作系统)


1、如果系统已经安装了网卡驱动,直接运行 驱动自带的命令 :sysinfo-snapshot.py  ,大概等待5分钟,

将会在 /tmp 目录生成一个 tgz包,把这个tgz包文件发送给售后支持团队。

[root@node01 ~]# sysinfo-snapshot.py

Sysinfo-snapshot is still in process...please wait till completed successfully

Gathering the information may take a while, especially in large networks

Your patience is appreciated

------------------------------------------------------------


Running sysinfo-snapshot has ended successfully!

Temporary destination directory is /tmp/

Out file name is /tmp/sysinfo-snapshot-v3.4.0-node01-20220307-063343.tgz



2、如果是驱动版本太低(几年前安装的系统或者驱动),或者系统暂时无法安装网卡驱动,

我们可以把新版本的 sysinfo-snapshot.py (Sinoinfo_Picsysinfo-snapshot_version_3_7_0.tgz)上传到服务器,单独运行新版本进行收集。

[root@node01 ~]# sysinfo-snapshot.py -v

Sysinfo-snapshot version: 3.4.0

[root@node01 ~]# tar xf sysinfo-snapshot_version_3_7_0.tgz

[root@node01 ~]# ls

sysinfo-snapshot_version_3_7_0.tgz  sysinfo-snapshot_version_3_7_0

[root@node01 ~]# cd sysinfo-snapshot_version_3_7_0

[root@node01 sysinfo-snapshot_version_3_7_0]# ls

config.csv  sysinfo-snapshot_v3.7.0.py

[root@node01 sysinfo-snapshot_version_3_7_0]# ./sysinfo-snapshot_v3.7.0.py

Sysinfo-snapshot is still in process...please wait till completed successfully

Gathering the information may take a while, especially in large networks
Your patience is appreciated
------------------------------------------------------------
Running sysinfo-snapshot has ended successfully!
Temporary destination directory is /tmp/
Out file name is /tmp/sysinfo-snapshot-v3.7.0-node01-20220307-064656.tgz


3、如果是IB网卡,我们可能还需要收集 ibdiagnet 日志,在IB子网中的任意一个服务器节点都可以运行

[root@node01 ~]# ibdiagnet --pc --pm_pause_time 600 -P all=1 --get_cable_info -r

等待600秒后,将在默认目录  /var/tmp/ibdianget2  下生成相关文件,如果想指定生成日志的文件夹,使用参数  -o

然后打包整个目录 ,把打包好的文件发送给售后支持团队。

[root@node01 ~]# tar czf node01-ibdiagnet.tar.gz /var/tmp/ibdiagnet2/*
tar: Removing leading `/' from member names
[root@sinoinfo ~]# ls node01-ibdiagnet.tar.gz
node01-ibdiagnet.tar.gz


4、(可选)收集IB子网拓扑,运行命令 ibnetdiscover ,把输出重定向到文件,把文件发送给售后支持团队

[root@node01 ~]# ibnetdiscover >ibnetdiscover.txt

与 NVIDIA产品相关的图片或视频(完整或部分)的版权均归 NVIDIA Corporation 所有。

微信公众号