我在恢复单个节点配置(看起来是什么)之后遇到了一些问题。 您可以在jira here上查看该问题的完整详细信息。
以下是mongod --repair
2016-06-16T16:15:21.625+0300 I STORAGE [initandlisten] WiredTiger progress WT_SESSION.verify 56200
2016-06-16T16:15:21.630+0300 E STORAGE [initandlisten] WiredTiger (0) [1466082921:630660][8347:0x7f006e3b6cc0], file:collection-128-6816245440313053.wt, WT_SESSION.verify: read checksum error for 12288B block at offset 288509952: block header checksum of 2294773803 doesn't match expected checksum of 1228719121
2016-06-16T16:15:21.680+0300 E STORAGE [initandlisten] WiredTiger (0) [1466082921:680635][8347:0x7f006e3b6cc0], file:collection-128-6816245440313053.wt, WT_SESSION.verify: checkpoint ranges never verified: 3657
2016-06-16T16:15:21.697+0300 E STORAGE [initandlisten] WiredTiger (0) [1466082921:697338][8347:0x7f006e3b6cc0], file:collection-128-6816245440313053.wt, WT_SESSION.verify: file ranges never verified: 3657
2016-06-16T16:15:21.697+0300 I STORAGE [initandlisten] Verify failed on uri table:collection-128-6816245440313053. Running a salvage operation.
2016-06-16T16:15:21.701+0300 I STORAGE [initandlisten] WiredTiger progress WT_SESSION.salvage 100
...
2016-06-16T16:15:27.541+0300 I STORAGE [initandlisten] WiredTiger progress WT_SESSION.salvage 96200
2016-06-16T16:15:27.673+0300 I INDEX [initandlisten] build index on: db.collection properties: { v: 1, key: { _id: 1 }, name: "_id_", ns: "db.collection" }
2016-06-16T16:15:27.673+0300 I INDEX [initandlisten] building index using bulk method
2016-06-16T16:15:53.433+0300 I - [initandlisten] Invariant failure rs.get() src/mongo/db/catalog/database.cpp 190
2016-06-16T16:15:53.434+0300 I - [initandlisten]
2016-06-16T16:15:53.462+0300 F - [initandlisten] Got signal: 6 (Aborted).
I've restarted and got the same failure (after the repair went through the data again)
2016-06-16T17:18:48.833+0300 I STORAGE [initandlisten] Verify succeeded on uri table:collection-128-6816245440313053. Not salvaging.
2016-06-16T17:18:48.847+0300 I INDEX [initandlisten] build index on: db.collection properties: { v: 1, key: { _id: 1 }, name: "_id_", ns: "db.collection" }
2016-06-16T17:18:48.847+0300 I INDEX [initandlisten] building index using bulk method
2016-06-16T17:19:15.146+0300 I - [initandlisten] Invariant failure rs.get() src/mongo/db/catalog/database.cpp 190
2016-06-16T17:19:15.146+0300 I - [initandlisten]
***aborting after invariant() failure
2016-06-16T17:19:15.189+0300 F - [initandlisten] Got signal: 6 (Aborted).
用于修复的mongod版本:
$ mongod --version
db version v3.2.7
git version: 4249c1d2b5999ebbf1fdf3bc0e0e3b3ff5c0aaf2
OpenSSL version: OpenSSL 1.0.1f 6 Jan 2014
allocator: tcmalloc
modules: none
build environment:
distmod: ubuntu1404
distarch: x86_64
target_arch: x86_64
我还有哪些其他恢复方案? 我如何忽略破碎的集合?试图将其从文件夹中删除但是没有用。