我使用redis在我的网站上存储统计数据,通常的stream量大约是每秒1000-2000次redis调用,最近我把这个网站迁移到了另一个托pipe服务上,而且我正在尝试redis-server的问题, 12小时运行它突然间得到公羊饥饿,直到我崩溃,为了得到它,我需要重新启动redis服务。 我没有看到在redis日志或系统日志中的任何exception
5274:M 17 Aug 14:55:26.006 * 10000 changes in 900 seconds. Saving... 5274:M 17 Aug 14:55:26.023 * Background saving started by pid 12754 12754:C 17 Aug 14:55:30.823 * DB saved on disk 12754:C 17 Aug 14:55:30.838 * RDB: 24 MB of memory used by copy-on-write 5274:M 17 Aug 14:55:30.925 * Background saving terminated with success 5274:M 17 Aug 15:07:40.615 # User requested shutdown... ->this is where I restart it 5274:M 17 Aug 15:07:40.615 * Saving the final RDB snapshot before exiting. 5274:M 17 Aug 15:07:44.877 * DB saved on disk 5274:M 17 Aug 15:07:44.877 * Removing the pid file. 5274:M 17 Aug 15:07:44.878 # Redis is now ready to exit, bye bye...
这是ram的使用情况,大概是在gg之前,在崩溃之前,它突然上升到了13gb(剩余20gb)
CPU系统时间(用户时间在正常水平)