Details
-
Task
-
Status: Closed
-
Major - P3
-
Resolution: Fixed
-
None
-
4
-
true
Description
We'd like to publish a full list of changes for 2.4.10 (and future "dot" releases) in addition to the 3 to 4 highlights on the Release Notes page.
Suggestion: Have two links there:
- Full list of changes (links to the new sub-page)
- 2.4.10 issues in Jira (as before)
Here the full list of changes:
# 2.4.10 Release Notes
|
|
- Indexes: Fixed issue that can cause index corruption when building indexes concurrently (:issue:`SERVER-12990`)
|
- Indexes: Fixed issue that can cause index corruption when shutting down secondary node during index build (:issue:`SERVER-12956`)
|
- Indexes: Mongod now recognizes incompatible “future” text and geo index versions and exits gracefully (:issue:`SERVER-12914`)
|
- Indexes: Fixed issue that can cause secondaries to fail replication when building the same index multiple times concurrently (:issue:`SERVER-12662`)
|
- Indexes: Fixed issue that can cause index corruption on the tenth index in a collection if the index build fails (:issue:`SERVER-12481`)
|
- Indexes: Introduced versioning for text and geo indexes to ensure backwards compatibility (:issue:`SERVER-12175`)
|
- Indexes: Disallowed building indexes on the system.indexes collection, which can lead to initial sync failure on secondaries (:issue:`SERVER-10231`)
|
- Sharding: Avoid frequent immediate balancer retries when config servers are out of sync (:issue:`SERVER-12908`)
|
- Sharding: Add indexes to locks collection on config servers to avoid long queries in case of large numbers of collections (:issue:`SERVER-12548`)
|
- Sharding: Fixed issue that can corrupt the config metadata cache when sharding collections concurrently (:issue:`SERVER-12515`)
|
- Sharding: Don't move chunks created on collections with a hashed shard key if the collection already contains data (:issue:`SERVER-9259`)
|
- Replication: Fixed issue where node appears to be down in a replica set during a compact operation (:issue:`SERVER-12264`)
|
- Replication: Fixed issue that could cause delays in elections when a node is not vetoing an election (:issue:`SERVER-12170`)
|
- Replication: Step down all primaries if multiple primaries are detected in replica set to ensure correct election result (:issue:`SERVER-10793`)
|
- Replication: Upon clock skew detection, secondaries will switch to sync directly from the primary to avoid sync cycles (:issue:`SERVER-8375`)
|
- Runtime: The SIGXCPU signal is now caught and mongod writes a log message and exits gracefully (:issue:`SERVER-12034`)
|
- Runtime: Fixed issue where mongod fails to start on Linux when /sys/dev/block directory is not readable (:issue:`SERVER-9248`)
|
- Windows: No longer zero-fill newly allocated files on systems other than Windows 7 or Windows Server 2008 R2 (:issue:`SERVER-8480`)
|
- GridFS: Chunk size is decreased to 255 KB (from 256 KB) to avoid overhead with usePowerOf2Sizes option (:issue:`SERVER-13331`)
|
- SNMP: Fixed MIB file validation under smilint (:issue:`SERVER-12487`)
|
- Shell: Fixed issue in V8 memory allocation that could cause long-running shell commands to crash (:issue:`SERVER-11871`)
|
- Shell: Fixed memory leak in the md5sumFile shell utility method (:issue:`SERVER-11560`)
|
|
This uses a format - Tag: Description. Alternatively, we could group the same tags together:
Tag
|
Description
|
Description
|
...
|
I'll leave the layout up to you. Whatever looks most webscale.