CPU 死锁问题 死机问题

山小峰

[i=s] 本帖最后由 山小峰 于 2025-3-9 16:38 编辑 [/i]<br /> <br />

root@Qing:~# Message from syslogd@trim-b17f at Mar 9 16:13:28 ... kernel:[ 3956.120640] watchdog: BUG: soft lockup - CPU#3 stuck for 175s! [kcompactd0:50]

Message from syslogd@trim-b17f at Mar 9 16:13:28 ... kernel:[ 3956.120640] watchdog: BUG: soft lockup - CPU#3 stuck for 175s! [kcompactd0:50]

root@Qing:~# Message from syslogd@trim-b17f at Mar 9 16:15:36 ... kernel:[ 4084.121266] watchdog: BUG: soft lockup - CPU#3 stuck for 294s! [kcompactd0:50]

Message from syslogd@trim-b17f at Mar 9 16:15:36 ... kernel:[ 4084.121266] watchdog: BUG: soft lockup - CPU#3 stuck for 294s! [kcompactd0:50]

deepseek解答

Linux内核软死锁（Soft Lockup）问题分析与解决方案

问题描述

系统日志中出现 watchdog: BUG: soft lockup - CPU#3 stuck for 175s! [kcompactd0:50] 报错，表明CPU#3上的进程 kcompactd0 在内核态长时间占用CPU未释放控制权，触发了内核的软死锁检测机制。此问题会导致系统响应变慢，但通常不会完全死机

1

20

41

。

核心原因分析

软死锁的根本原因是某个内核线程或进程（如本例中的 kcompactd0）在内核模式下长时间占用CPU且未释放资源，导致其他任务无法调度。可能的具体原因包括：

1. CPU高负载：长时间高负载任务（如内存压缩操作）导致CPU无法响应看门狗中断 31

   46

   。

2. 硬件问题：

   • 电源供电不足或电压不稳定（常见于物理服务器） 1

     20

     。

   • CPU过热或BIOS设置不当（如开启超频、C-State节能模式、Spread Spectrum等） 1

     20

     41

     。

3. 内核或驱动Bug：内存管理模块（如 kcompactd0 相关的内存压缩机制）存在缺陷 41

   47

   。

4. 虚拟机配置问题：虚拟机的vCPU数量超过物理核心数，或宿主机资源争抢 1

   20

   。

czw

Message from syslogd@home-1 at Mar 22 18:43:17 ...
kernel:[ 8407.864889] watchdog: BUG: soft lockup - CPU#3 stuck for 7797s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:43:45 ...
kernel:[ 8435.864807] watchdog: BUG: soft lockup - CPU#3 stuck for 7823s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:44:41 ...
kernel:[ 8491.864607] watchdog: BUG: soft lockup - CPU#3 stuck for 7875s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:45:09 ...
kernel:[ 8519.864490] watchdog: BUG: soft lockup - CPU#3 stuck for 7901s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:48:53 ...
kernel:[ 8743.863241] watchdog: BUG: soft lockup - CPU#3 stuck for 8110s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:49:49 ...
kernel:[ 8799.862856] watchdog: BUG: soft lockup - CPU#3 stuck for 8162s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:50:17 ...
kernel:[ 8827.862655] watchdog: BUG: soft lockup - CPU#3 stuck for 8188s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:50:45 ...
kernel:[ 8855.862447] watchdog: BUG: soft lockup - CPU#3 stuck for 8214s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:53:33 ...
kernel:[ 9023.861106] watchdog: BUG: soft lockup - CPU#3 stuck for 8371s! [mount:2817]

Message from syslogd@home-1 at Mar 22 18:54:01 ...
kernel:[ 9051.860868] watchdog: BUG: soft lockup - CPU#3 stuck for 8397s! [mount:2817]

山小峰

czw 发表于 2025-3-22 18:55
Message from syslogd@home-1 at Mar 22 18:43:17 ...
kernel:[ 8407.864889] watchdog: BUG: soft lockup ...

看来不止我一个人的问题啊

czw

山小峰 发表于 2025-3-23 21:45
看来不止我一个人的问题啊

Reinstall and emby is using an earlier version of 4.8.9. Then notice that the previous docker folder is deleted clean, so that it does not appear for a short time.相同的问题。

唉 输入法 经常自己变成英语翻译的。  奇了怪了