我不知道为什么我的芹菜工人每天都挂在同一时间，而且超过3个月没有进行任何代码更改。
芹菜工人似乎总是卡在revcfrom（x，... 查看论坛，这表明文件描述符可能存在死锁。这是两个工人pid = 30626，30627（他们分别从9和17卡住）的痕迹
uwsgi 29493 ubuntu 9u IPv4 56723782 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:41734->ip-172-31-23-149.us-west-2.compute.internal:mysql (ESTABLISHED)
uwsgi 29494 ubuntu 9u sock 0,8 0t0 56439057 protocol: TCP
uwsgi 29495 ubuntu 9u unix 0xffff8800e9ae6800 0t0 56339078 type=STREAM
celery 30627 ubuntu 9u IPv4 56710356 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:38140->ip-172-31-23-149.us-west-2.compute.internal:mysql (ESTABLISHED)
celery 30627 17217 ubuntu 9u IPv4 56710356 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:38140->ip-172-31-23-149.us-west-2.compute.internal:mysql (ESTABLISHED)
systemd 1 root 17u unix 0xffff8800e9d6dc00 0t0 56338935 type=DGRAM
systemd-j 386 root 17u unix 0xffff8800eac41800 0t0 11976 /run/systemd/journal/stdout type=STREAM
nginx 28994 www-data 17u IPv4 56828469 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:81->ip-172-31-25-87.us-west-2.compute.internal:6209 (ESTABLISHED)
uwsgi 29480 ubuntu 17u unix 0xffff8800ea3f6400 0t0 56339086 type=STREAM
uwsgi 29499 ubuntu 17u unix 0xffff8800ea3f6400 0t0 56339086 type=STREAM
celery 30626 ubuntu 17u IPv4 56829569 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:37158->ip-172-31-23-149.us-west-2.compute.internal:mysql (ESTABLISHED)
celery 30626 12626 ubuntu 17u IPv4 56829569 0t0 TCP ip-172-31-29-33.us-west-2.compute.internal:37158->ip-172-31-23-149.us-west-2.compute.internal:mysql (ESTABLISHED)
我很难理解这一点，为什么nginx访问相同的东西？尽管该经纪人还活着，但我只是想不出为什么它会卡住并且总是在每天中午左右发生，所以CPU使用率也没有增加。谢谢那些有建议的人！