Use Systemtap to Identify Pod Exceptions

Terakhir diperbarui:2022-04-20 19:15:30

This article describes how to use SystemTap to troubleshoot pod issues.

Preparations

Different operating systems have different methods for installing SystemTap and its dependencies. Pick one that suits you.

Ubuntu

Run the following command to install SystemTap:
```
apt install -y systemtap
```
Run the following command to check for dependencies:
```
stap-prep
```

The following is a sample result:

Please install linux-headers-4.4.0-104-generic
You need package linux-image-4.4.0-104-generic-dbgsym but it does not seem to be available
Ubuntu -dbgsym packages are typically in a separate repository
Follow https://wiki.ubuntu.com/DebuggingProgramCrash to add this repository
apt install -y linux-headers-4.4.0-104-generic

The above result shows that you need to install dbgsym, which is not in the existing sources. Run the following command to add the third-party source:

   sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys C8CAB6595FDFF622

codename=$(lsb_release -c | awk  '{print $2}')
sudo tee /etc/apt/sources.list.d/ddebs.list << EOF
deb http://ddebs.ubuntu.com/ ${codename}      main restricted universe multiverse
deb http://ddebs.ubuntu.com/ ${codename}-security main restricted universe multiverse
deb http://ddebs.ubuntu.com/ ${codename}-updates  main restricted universe multiverse
deb http://ddebs.ubuntu.com/ ${codename}-proposed main restricted universe multiverse
EOF

sudo apt-get update

Run the following command after adding the source:
```
stap-prep
```

The following is a sample result:

Please install linux-headers-4.4.0-104-generic
Please install linux-image-4.4.0-104-generic-dbgsym

Run the following command to install the prompted packages:
```
apt install -y linux-image-4.4.0-104-generic-dbgsym
```

apt install -y linux-headers-4.4.0-104-generic

CentOS

Run the following command to install SystemTap:
```
yum install -y systemtap
```

For the purpose of this article, we assume that debuginfo is not added. Add the following to /etc/yum.repos.d/CentOS-Debug.repo and save.

[debuginfo]
name=CentOS-$releasever - DebugInfo
baseurl=http://debuginfo.centos.org/$releasever/$basearch/
gpgcheck=0
enabled=1
protect=1
priority=1

Run the following command to check for dependencies and install them:

Note：

The following command installs kernel-debuginfo.
```
stap-prep
```
Run the following command to check if the node has multiple versions of kernel-devel installed:
```
rpm -qa | grep kernel-devel
```

The returned result is as follows:

kernel-devel-3.10.0-327.el7.x86_64
kernel-devel-3.10.0-514.26.2.el7.x86_64
kernel-devel-3.10.0-862.9.1.el7.x86_64

If there are multiple versions, keep the one that corresponds to the kernel version. For example, if the current kernel version is 3.10.0-862.9.1.el7.x86_64, delete all version except kernel-devel-3.10.0-862.9.1.el7.x86_64.

Note：

You can use uname -r to view the kernel version.

Make sure kernel-debuginfo and kernel-devel are both installed and their versions correspond to the kernel version.

rpm -e kernel-devel-3.10.0-327.el7.x86_64 kernel-devel-3.10.0-514.26.2.el7.x86_64

Problem Analysis

You can use SystemTap to monitor a process in order to troubleshoot pod issues. This is how it works:

SystemTap translates the script into C code and calls gcc to compile the code into the Linux kernel module. It then uses modprobe to load the module into the kernel.
It uses the script to create kernel hooks and identify the causes of pod issues using the signals captured by the hooks.

Troubleshooting

Step 1: obtain the pids of the containers that restarted automatically in the pod due to exceptions

Run the following command to obtain the Container ID:
```
kubectl describe pod <pod name>
```

The returned result is as follows:

......
Container ID:  docker://5fb8adf9ee62afc6d3f6f3d9590041818750b392dff015d7091eaaf99cf1c945
......
Last State:     Terminated
Reason:       Error
Exit Code:    137
Started:      Thu, 05 Sep 2019 19:22:30 +0800
Finished:     Thu, 05 Sep 2019 19:33:44 +0800

Run the following command to query the pid of the main container process using the obtained Container ID:
```
docker inspect -f "{{.State.Pid}}" 5fb8adf9ee62afc6d3f6f3d9590041818750b392dff015d7091eaaf99cf1c945
```

The returned result is as follows:

Step 2: narrow the scope using the container exit code

Use the Exit Code in the result of Step 1 to obtain the status code of the last container exit. For the purpose of this article, we will use 137 as an example. The analysis is as follows:

If the process was killed by an external signal, the exit code should be between 129 and 255.
An exit code of 137 indicates that the process was killed by SIGKILL. However, we still cannot determine the reason why the process exited.

Step 3: use the SystemTap script to identify the reason

Assuming the issue is reproducible, you can use a SystemTap to troubleshoot the problem.

Create a file called sg.stp. Add the following content and save.

global target_pid = 7942
probe signal.send{
if (sig_pid == target_pid) {
 printf("%s(%d) send %s to %s(%d)\n", execname(), pid(), sig_name, pid_name, sig_pid);
 printf("parent of sender: %s(%d)\n", pexecname(), ppid())
 printf("task_ancestry:%s\n", task_ancestry(pid2task(pid()), 1));
}
}

Note：

Substitute pid with the value of the main container process pid obtained in Step 2. For the purpose of this article, we will use 7942 as an example:

Run the following command to execute the script:
```
stap sg.stp
```

When the container process is killed, the script captures the event and outputs the following:

pkill(23549) send SIGKILL to server(7942)
parent of sender: bash(23495)
task_ancestry:swapper/0(0m0.000000000s)=>systemd(0m0.080000000s)=>vGhyM0(19491m2.579563677s)=>sh(33473m38.074571885s)=>bash(33473m38.077072025s)=>bash(33473m38.081028267s)=>bash(33475m4.817798337s)=>pkill(33475m5.202486630s)

Solution

By observing task_ancestry, you can see the parent processes of the stopped process. In the example above, you can see a strange process called vGhyM0. This usually indicates that there is a trojan in the system. Take the necessary steps to clean it so your containers can function properly.

tencent cloud

Recent Pages