我有一个用于崩溃报告的Django网站,利用Celery定期更新网站时出现新的崩溃。一切都运行良好,除了在3个小时之间的任意间隔之后5天,两个预定任务不再开火。虽然较高的崩溃率似乎缩短了它可能发生的时间间隔,但似乎没有多少押韵或理由为什么它们会停止。
区分这两项任务的是:
他们都使用以下任务装饰器:
@periodic_task(run_every=crontab(hour='*/3'),
ignore_result=True,
queue=WorkQueue.LOCAL,
options={'queue': WorkQueue.LOCAL})
我注意到的事情:
有什么想法吗?如果我需要提供更多信息,请告诉我。
编辑:在调度程序删除任务之前,来自工作人员的日志片段:
[2014-09-07 00:06:00,874: WARNING/Worker-1] Submitted new modules for report 00e6d5af-21de-4981-9c9f-dcc002140906
[2014-09-07 00:06:00,874: WARNING/Worker-1] Retrieving report from db.
[2014-09-07 00:06:00,874: WARNING/Worker-1] Marking complete/processed 00e6d5af-21de-4981-9c9f-dcc002140906
[2014-09-07 00:06:00,921: INFO/MainProcess] Task sirocco.crashes.tasks.analyze_windows_crash[5313821c-3a43-47fa-9a27-c6a
6fba41b03] succeeded in 8.81399989128s: None
[2014-09-07 00:06:04,180: WARNING/Worker-1] CDB exited with code 0 in 12.043 seconds
[2014-09-07 00:06:04,180: WARNING/Worker-1] Created new digest 38ec8b36230e783298d53dcc58ad655502d57b755c7d23e9bb9b34892
8bdb941
[2014-09-07 00:06:04,197: WARNING/Worker-1] Attempting to submit 821d6cd2-07fb-491d-886c-574682140906
[2014-09-07 00:06:04,197: WARNING/Worker-1] [821d6cd2-07fb-491d-886c-574682140906] Submit analysis to DB
[2014-09-07 00:06:04,211: WARNING/Worker-1] Got existing signature 205
[2014-09-07 00:06:04,243: WARNING/Worker-1] Submitting modules for 821d6cd2-07fb-491d-886c-574682140906:
[2014-09-07 00:06:05,569: WARNING/Worker-1] Submitted new modules for report 821d6cd2-07fb-491d-886c-574682140906
[2014-09-07 00:06:05,569: WARNING/Worker-1] Retrieving report from db.
[2014-09-07 00:06:05,569: WARNING/Worker-1] Marking complete/processed 821d6cd2-07fb-491d-886c-574682140906
[2014-09-07 00:06:05,615: INFO/MainProcess] Task sirocco.crashes.tasks.analyze_windows_crash[b59f9846-56d7-40fd-a235-96a
4349f20cc] succeeded in 13.5089998245s: None