Mongodb 因地址无效访问而崩溃 - 分段错误信号 11 - 版本 2.6

我正在尝试在 MongoDB 中插入文档。我当前的版本是 2.6 生产版本。我的应用程序能够插入一些文档，但在某个时间点后开始出现以下错误。我每次都会遇到同样的错误。我正在从 cmd 提示符运行 mongod。我在 2.6.0 rc 和 2.6.1 rc 中遇到了同样的问题，但在 2.5.4 中工作正常。 下面是日志跟踪。

2014-04-08T20:04:01.373+0000 [conn12] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:40 reslen:40 18ms
2014-04-08T20:04:01.612+0000 [conn15] update deadlinedb.DeadlineSettings query: { _id: "history_log" } update: { $push: { Entries: "2014/04/08 15:04:00 vibhu MOBILE-019 (MOBILE-019\Vibhu): Submitted new job: (tyler.jordan,"Among assumed stumbles v051",92:none:none,2014/04/08 15:04:00,)." }, $inc: { EntryCount: 1 } } nscanned:1 nscannedObjects:1 nMatched:1 nModified:1 keyUpdates:0 numYields:0 locks(micros) w:5621 21ms
2014-04-08T20:04:01.612+0000 [conn15] command deadlinedb.$cmd command: update { update: "DeadlineSettings", ordered: true, updates: [ { q: { _id: "history_log" }, u: { $push: { Entries: "2014/04/08 15:04:00 vibhu MOBILE-019 (MOBILE-019\Vibhu): Submitted new job: (tyler.jordan,"Among assumed stumbles v051",92:none:none,2014/04/08 15:04:00,)." }, $inc: { EntryCount: 1 } }, multi: false, upsert: true } ] } keyUpdates:0 numYields:0  reslen:55 22ms
2014-04-08T20:04:01.681+0000 [conn12] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:23 reslen:40 16ms
2014-04-08T20:04:01.727+0000 [conn12] update deadlinedb.LimitGroups query: { _id: "534456307f87b2196c09841b" } update: { $set: { IsSub: false }, $currentDate: { LastWriteTime: { $type: "date" } } } nscanned:1 nscannedObjects:1 nMatched:1 nModified:1 keyUpdates:1 numYields:0 locks(micros) w:161 17ms
2014-04-08T20:04:01.727+0000 [conn12] command deadlinedb.$cmd command: update { update: "LimitGroups", ordered: true, updates: [ { q: { _id: "534456307f87b2196c09841b" }, u: { $set: { IsSub: false }, $currentDate: { LastWriteTime: { $type: "date" } } }, multi: false, upsert: false } ] } keyUpdates:0 numYields:0  reslen:55 17ms
2014-04-08T20:04:02.676+0000 [conn15] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:21 reslen:40 11ms
2014-04-08T20:04:02.676+0000 [conn12] update deadlinedb.Jobs query: { _id: "534456317f87b2196c098426" } update: { $set: { IsSub: false }, $currentDate: { LastWriteTime: { $type: "date" } } } nscanned:1 nscannedObjects:1 nMatched:1 nModified:1 keyUpdates:1 numYields:0 locks(micros) w:89 17ms
2014-04-08T20:04:02.676+0000 [conn12] command deadlinedb.$cmd command: update { update: "Jobs", ordered: true, updates: [ { q: { _id: "534456317f87b2196c098426" }, u: { $set: { IsSub: false }, $currentDate: { LastWriteTime: { $type: "date" } } }, multi: false, upsert: false } ] } keyUpdates:0 numYields:0  reslen:55 17ms
2014-04-08T20:04:03.165+0000 [conn12] insert deadlinedb.JobTasks query: { _id: "534456327f87b2196c09842b_11", JobID: "534456327f87b2196c09842b", TaskID: 11, Frames: "11-11", Slave: "", Stat: 2, Errs: 0, Start: new Date(-62135596800000), StartRen: new Date(-62135596800000), Comp: new Date(-62135596800000), WtgStrt: false } ninserted:1 keyUpdates:0 numYields:0 locks(micros) w:150875 151ms
2014-04-08T20:04:03.178+0000 [conn12] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:259 reslen:40 166ms
2014-04-08T20:04:03.526+0000 [conn15] insert deadlinedb.JobTasks query: { _id: "534456327f87b2196c098430_0", JobID: "534456327f87b2196c098430", TaskID: 0, Frames: "0-9", Slave: "", Stat: 2, Errs: 0, Start: new Date(-62135596800000), StartRen: new Date(-62135596800000), Comp: new Date(-62135596800000), WtgStrt: false } ninserted:1 keyUpdates:0 numYields:0 locks(micros) w:12295 12ms
2014-04-08T20:04:03.529+0000 [conn15] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:12 reslen:40 16ms
2014-04-08T20:04:03.710+0000 [conn15] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:22 reslen:40 30ms
2014-04-08T20:04:04.042+0000 [conn12] insert deadlinedb.JobTasks query: { _id: "534456337f87b2196c098435_22", JobID: "534456337f87b2196c098435", TaskID: 22, Frames: "132-137", Slave: "", Stat: 2, Errs: 0, Start: new Date(-62135596800000), StartRen: new Date(-62135596800000), Comp: new Date(-62135596800000), WtgStrt: false } ninserted:1 keyUpdates:0 numYields:0 locks(micros) w:15158 15ms
2014-04-08T20:04:04.043+0000 [conn12] command deadlinedb.$cmd command: insert { $msg: "query not recording (too large)" } keyUpdates:0 numYields:0 locks(micros) w:49 reslen:40 17ms
2014-04-08T20:04:04.072+0000 [conn15] SEVERE: Invalid access at address: 0x20069d1e02
2014-04-08T20:04:04.081+0000 [conn15] SEVERE: Got signal: 11 (Segmentation fault).
Backtrace:0x11bd301 0x11bc6de 0x11bc7cf 0x7f659066acb0 0x796892 0x796fbf 0x79773d 0x78f974 0x78fd42 0xc220ee 0xc430b9 0xc3ce61 0xc48ca6 0xb8f1c9 0xb993e8 0x76b76f 0x117367b 0x7f6590662e9a 0x7f658f9753fd
 mongod(_ZN5mongo15printStackTraceERSo+0x21) [0x11bd301]
 mongod() [0x11bc6de]
 mongod() [0x11bc7cf]
 /lib/x86_64-linux-gnu/libpthread.so.0(+0xfcb0) [0x7f659066acb0]
 mongod(_ZNK5mongo11mutablebson8Document4Impl12writeElementINS_16BSONArrayBuilderEEEvjPT_PKNS_10StringDataE+0x52) [0x796892]
 mongod(_ZNK5mongo11mutablebson8Document4Impl13writeChildrenINS_16BSONArrayBuilderEEEvjPT_+0x4f) [0x796fbf]
 mongod(_ZNK5mongo11mutablebson8Document4Impl12writeElementINS_14BSONObjBuilderEEEvjPT_PKNS_10StringDataE+0x22d) [0x79773d]
 mongod(_ZN5mongo11mutablebson8Document11makeElementENS0_12ConstElementEPKNS_10StringDataE+0x64) [0x78f974]
 mongod(_ZN5mongo11mutablebson8Document27makeElementWithNewFieldNameERKNS_10StringDataENS0_12ConstElementE+0x22) [0x78fd42]
 mongod(_ZNK5mongo11ModifierPop3logEPNS_10LogBuilderE+0x10e) [0xc220ee]
 mongod(_ZN5mongo12UpdateDriver6updateERKNS_10StringDataEPNS_11mutablebson8DocumentEPNS_7BSONObjEPNS_11FieldRefSetE+0x319) [0xc430b9]
 mongod(_ZN5mongo6updateERKNS_13UpdateRequestEPNS_7OpDebugEPNS_12UpdateDriverEPNS_14CanonicalQueryE+0xc71) [0xc3ce61]
 mongod(_ZN5mongo14UpdateExecutor7executeEv+0x66) [0xc48ca6]
 mongod(_ZN5mongo14receivedUpdateERNS_7MessageERNS_5CurOpE+0x729) [0xb8f1c9]
 mongod(_ZN5mongo16assembleResponseERNS_7MessageERNS_10DbResponseERKNS_11HostAndPortE+0xec8) [0xb993e8]
 mongod(_ZN5mongo16MyMessageHandler7processERNS_7MessageEPNS_21AbstractMessagingPortEPNS_9LastErrorE+0x9f) [0x76b76f]
 mongod(_ZN5mongo17PortMessageServer17handleIncomingMsgEPv+0x4fb) [0x117367b]
 /lib/x86_64-linux-gnu/libpthread.so.0(+0x7e9a) [0x7f6590662e9a]
 /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f658f9753fd]

你能尽快看一下这个吗？如果需要更多信息，请告诉我。

Thanks!

Vibhu

只是因为没有人添加答案，我会的。

这是 Mongo 2.6.0 中的一个错误，您可以在以下位置查看报告蒙戈·吉拉 https://jira.mongodb.org/browse/SERVER-13516.

问题摘要 对超过 128 个 BSON 的阵列发出更新 文档的元素（不限于数组本身）是 更新期间遍历可能会导致mongod崩溃 过程。受影响的更新运算符包括 {$pop: {: -1}} （来自 前面）并使用 .$ 或 . 进行位置更新。在哪里 是一个数值。

USER IMPACT 用户使用 $pop 更新大型数组 或位置运算符受影响最大，但限制为 128 BSON elements 适用于整个文档，而不仅仅是数组。这 失败的更新将使主节点（以及任何主节点）崩溃 随后当选，如果应用程序重复更新 这些节点之一）。部署将遭受停机，直到 集群可以降级，但不会丢失/数据损坏 发生。

没有什么可以做的2.6.0版本。解决方法是升级到2.6.1或更高或降级至2.4.x branch.