我们的mongodb进程在每个星期六上午10点被杀死并重新启动。我们间歇性地遇到用户数据丢失的问题。要恢复，我们必须重新创建所有用户，包括siteUserAdmin。
在Linux上运行：
2.6.32-279.14.1.el6.x86_64 #1 SMP Mon Oct 15 13:44:51 EDT 2012 x86_64 x86_64 x86_64 GNU/Linux
以下是db log中的内容：
2015-10-02T18:24:45.413+0100 I NETWORK [initandlisten] connection accepted from xxx.211.4.30:47728 #300 (159 connections now open)
2015-10-02T18:24:45.439+0100 I ACCESS [conn300] Successfully authenticated as principal user on mydb
2015-10-02T18:24:45.446+0100 I NETWORK [conn300] end connection xxx.211.4.30:47728 (158 connections now open)
2015-10-03T10:00:35.013+0100 I CONTROL [signalProcessingThread] got signal 15 (Terminated), will terminate after current cmd ends
2015-10-03T10:00:35.115+0100 I CONTROL [signalProcessingThread] now exiting
2015-10-03T10:00:35.115+0100 I NETWORK [signalProcessingThread] shutdown: going to close listening sockets...
2015-10-03T10:00:35.129+0100 I NETWORK [signalProcessingThread] closing listening socket: 9
2015-10-03T10:00:35.129+0100 I NETWORK [signalProcessingThread] closing listening socket: 10
2015-10-03T10:00:35.195+0100 I NETWORK [signalProcessingThread] removing socket file: /tmp/mongodb-27017.sock
2015-10-03T10:00:35.195+0100 I NETWORK [signalProcessingThread] shutdown: going to flush diaglog...
2015-10-03T10:00:35.195+0100 I NETWORK [signalProcessingThread] shutdown: going to close sockets...
2015-10-03T10:00:35.196+0100 I NETWORK [conn9] end connection xxx.31.111.22:33093 (157 connections now open)
2015-10-03T10:00:35.196+0100 I NETWORK [conn107] end connection xxx.163.125.7:62991 (157 connections now open)
2015-10-03T10:00:35.196+0100 I NETWORK [conn22] end connection xxx.11.66.91:39150 (157 connections now open)
2015-10-03T10:00:35.196+0100 I NETWORK [conn128] end connection xxx.39.75.39:52306 (157 connections now open)
...
2015-10-03T10:00:35.199+0100 I NETWORK [conn247] end connection xxx.31.203.10:48608 (155 connections now open)
2015-10-03T10:00:35.199+0100 I NETWORK [conn245] end connection xxx.31.203.10:48602 (154 connections now open)
2015-10-03T10:00:35.199+0100 I NETWORK [conn264] end connection xxx.31.203.10:54803 (153 connections now open)
2015-10-03T10:00:35.199+0100 I NETWORK [conn248] end connection xxx.31.203.10:48203 (154 connections now open)
2015-10-03T10:00:35.216+0100 I STORAGE [signalProcessingThread] shutdown: waiting for fs preallocator...
2015-10-03T10:00:35.233+0100 I STORAGE [signalProcessingThread] shutdown: final commit...
2015-10-03T10:00:35.263+0100 I JOURNAL [signalProcessingThread] journalCleanup...
2015-10-03T10:00:35.263+0100 I JOURNAL [signalProcessingThread] removeJournalFiles
2015-10-03T10:00:35.352+0100 I JOURNAL [signalProcessingThread] Terminating durability thread ...
2015-10-03T10:00:35.438+0100 I JOURNAL [journal writer] Journal writer thread stopped
2015-10-03T10:00:35.438+0100 I JOURNAL [durability] Durability thread stopped
2015-10-03T10:00:35.438+0100 I STORAGE [signalProcessingThread] shutdown: closing all files...
2015-10-03T10:00:35.455+0100 I STORAGE [signalProcessingThread] closeAllFiles() finished
2015-10-03T10:00:35.463+0100 I STORAGE [signalProcessingThread] shutdown: removing fs lock...
2015-10-03T10:00:35.463+0100 I CONTROL [signalProcessingThread] dbexit: rc: 0
2015-10-03T10:20:04.575+0100 I CONTROL ***** SERVER RESTARTED *****
2015-10-03T10:20:04.629+0100 I JOURNAL [initandlisten] journal dir=xxx/journal
2015-10-03T10:20:04.629+0100 I JOURNAL [initandlisten] recover : no journal files present, no recovery needed
2015-10-03T10:20:04.655+0100 I JOURNAL [durability] Durability thread started
2015-10-03T10:20:04.655+0100 I JOURNAL [journal writer] Journal writer thread started
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] MongoDB starting : pid=17795 port=27017 dbpath=xxx/db2 64-bit host=xxx
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten]
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] ** WARNING: You are running on a NUMA machine.
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] ** We suggest launching mongod like this to avoid performance problems:
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] ** numactl --interleave=all mongod [other options]
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten]
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] db version v3.0.2
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] git version: 6201872043ecbbc0a4cc169b5482dcf385fc464f
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] build info: Linux build6.nj1.10gen.cc 2.6.32-431.3.1.el6.x86_64 #1 SMP Fri Jan 3 21:39:27 UTC 2014 x86_64 BOOST_LIB_VERSION=1_49
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] allocator: tcmalloc
2015-10-03T10:20:04.656+0100 I CONTROL [initandlisten] options: { config: "xxx/mongodb-linux-x86_64-3.0.2/bin/mongodb.conf", processManagement: { fork: true }, security: { authorization: "enabled" }, storage: { dbPath: "xxx/db2", journal: { enabled: true } }, systemLog: { destination: "file", logAppend: true, path: "xxx/mongod.log" } }
2015-10-03T10:20:04.827+0100 I NETWORK [initandlisten] waiting for connections on port 27017
2015-10-03T10:20:05.045+0100 I NETWORK [initandlisten] connection accepted from xxx.11.239.14:36968 #1 (1 connection now open)
2015-10-03T10:20:05.071+0100 I NETWORK [initandlisten] connection accepted from xxx.3.50.33:36240 #2 (2 connections now open)
2015-10-03T10:20:05.139+0100 I NETWORK [initandlisten] connection accepted from xxx.39.144.16:47416 #3 (3 connections now open)
...
2015-10-03T10:20:16.730+0100 I NETWORK [initandlisten] connection accepted from xxx.31.203.10:51281 #67 (67 connections now open)
2015-10-03T10:20:16.843+0100 I NETWORK [initandlisten] connection accepted from xxx.31.203.10:51282 #68 (68 connections now open)
2015-10-03T10:20:16.862+0100 I ACCESS [conn68] authenticate db: mydb { authenticate: 1, user: "user", nonce: "xxx", key: "xxx" }
2015-10-03T10:20:16.862+0100 I ACCESS [conn68] Failed to authenticate user@mydb with mechanism MONGODB-CR: AuthenticationFailed UserNotFound Could not find user user@mydb
有点不知所措，了解其原因并欢迎提出好的建议。