连接本地端口失败

[stalkercn]

sscms7.3.0版本，ubuntu 20.04操作系统，最近使用一键发布的时候有一定几率nginx返回502错误，查看nginx错误日志发现连接失败，但sscms服务在本机127.0.0.1:5000上，连接本机端口怎么会失败呢？

2025/08/06 15:28:05 [error] 1352#1352: *658450 connect() failed (111: Unknown error) while connecting to upstream, client: 10.10.40.9, server: xxxxxx.com, request: "GET /api/admin/cms/create/createStatus?siteId=189 HTTP/1.0", upstream: "http://127.0.0.1:5000/api/admin/cms/create/createStatus?siteId=189", host: "xxxxxx.com", referrer: "https://xxxxxx.com/ss-admin/cms/createStatus/?siteId=189"

[stalkercn]

查看了一下sscms进程的日志，发现访问一键发布的时候出现了如下日志

Aug 21 08:52:14 lq-www dotnet[17676]:    at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[[SSCMS.Core.Services.PathManager+<GetChannelUrlAsync>d__92, SSCMS.Core, Version=7.3.0.0, Culture=neutral, PublicKeyToken=null]](<GetChannelUrlAsync>d__92 ByRef)
Aug 21 08:52:14 lq-www dotnet[17676]:    at SSCMS.Core.Services.PathManager+<GetChannelUrlAsync>d__92.MoveNext()
Aug 21 08:52:14 lq-www systemd[1]: sscms.service: Main process exited, code=killed, status=6/ABRT
Aug 21 08:52:14 lq-www systemd[1]: sscms.service: Failed with result 'signal'.
Aug 21 08:52:14 lq-www systemd[1]: sscms.service: Consumed 1min 59.448s CPU time.
Aug 21 08:52:24 lq-www systemd[1]: sscms.service: Scheduled restart job, restart counter is at 5.
Aug 21 08:52:24 lq-www systemd[1]: Stopped SSCMS.
Aug 21 08:52:24 lq-www systemd[1]: sscms.service: Consumed 1min 59.448s CPU time.
Aug 21 08:52:24 lq-www systemd[1]: Started SSCMS.```

[stalkercn]

调试了一下，找到问题原因了，管理员不当操作导致的，不知道算不算bug。
复现方法：把一个栏目的栏目类型修改为“链接到指定栏目”，然后把指定栏目设置为本栏目，前端和后端都没有对这个进行防呆拦截，导致无限循环了。