Clone Tools
  • last updated a few minutes ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
JCR-2323: Wrong Upload size on cluster mode

Problem analysis:

- On clustering mode : for files whose size exceeds the buffer value (by default 200K), the steps are as follows:

- The property jcr:data is built in the first node having the value of an object (more precisely of StremValuePropertyData type) that has an instance variable with the name "file" containing the file path in the value storage and point on a path in the swap file.

- When this property is applied to the second node and the object is intercepted, the logic responsible for the serialization of the object simply checks if the file exists in the system or not (which is not the case) and make it "null" if necessary.

Fix description

- When the path file of StremValuePropertyData is null, do forceLoad the property data from the DB storage.

  1. … 7 more files in changeset.
JCR-2323: Wrong Upload size on cluster mode

  1. … 6 more files in changeset.
JCR-2323 : Wrong Upload size on cluster mode (Part 2)

  1. … 5 more files in changeset.
JCR-2312 : Cannot start eXo JCR with MySQL in UTF-8

Fix description:

- Relimit character number of a varchar field if dialect is mysql and utf-8

  1. … 17 more files in changeset.
JCR-2312 : Cannot start eXo JCR with MySQL in UTF-8

  1. … 17 more files in changeset.
JCR-2312 : use the same dialect variable for jcr and lock

  1. … 2 more files in changeset.
JCR-2312 : Cannot start eXo JCR with MySQL in UTF-8

  1. … 18 more files in changeset.
JCR-2312 : Cannot start eXo JCR with MySQL in UTF-8

  1. … 12 more files in changeset.
JCR-2288: JCR-2309 Avoid replication for local changes

  1. … 1 more file in changeset.
JCR-2288: [Cluster] Adding nodes to the same parent node causes TimeoutException

Problem analysis

- The goal of this task is to get rid or limit the risk of getting TimeoutException when we concurrently add nodes to the same parent node in a cluster environment.

Fix description

- When we add a new child node, we have to invalidate the list of child nodes of the parent node in the cache which was done by a simple remove of the related cache entry. The problem with this approach is the fact that this remove was done in a global tx so when it is done concurrently we can face deadlock like in this particular use case. The fix simply isolates the invalidation by doing it outside the global tx in order to highly reduce the risk of such deadlocks. The second part of the fix ensures that if we do a local change, the related commands commit/rollback remain local because before the fix they were replicated even for local changes.

  1. … 9 more files in changeset.
JCR-2288: [Cluster] Adding nodes to the same parent node causes TimeoutException

  1. … 9 more files in changeset.
JCR-2288: Align the PR with the fix in the master

  1. … 3 more files in changeset.
JCR-2288: Fixed the lazy get implementation

  1. … 1 more file in changeset.
JCR-2288: Invalidation is done before to ensure consistency

  1. … 8 more files in changeset.
JCR-2288: [Cluster] Adding nodes to the same parent node causes TimeoutException

  1. … 6 more files in changeset.
JCR-2281: Loading all the properties into the cache causes undesirable side effects

Problem analysis:

- Some parts of the code (orderBefore + Remove + Same name sibling and import document view with UUID_COLLISION_REPLACE_EXISTING) could behave abnormally in case the cache contains all the properties of the target node, that is due to the fact that those parts of the code assume that we get the list of properties from the database which allows to have the correct path for the properties when we change the path of a node which is not true in case we get the list of properties from the cache because the path of the properties still have the old path.

Fix description:

- We make sure that the path of the properties matches with the expected/new path

  1. … 17 more files in changeset.
JCR-2281: Loading all the properties into the cache causes undesirable side effects

  1. … 17 more files in changeset.
JCR-2234: Make sure that the JDBC Cache store used by the LockManager is optimized and reliable (jcr)

    • -263
    • +0
    ./impl/core/query/ispn/AbstractIndexerCacheStore.java
    • -82
    • +0
    ./impl/core/query/ispn/ChangesKey.java
    • -253
    • +0
    ./impl/core/query/ispn/ISPNIndexInfos.java
    • -42
    • +0
    ./impl/core/query/ispn/IndexInfosKey.java
    • -43
    • +0
    ./impl/core/query/ispn/IndexUpdateKey.java
    • -276
    • +0
    ./impl/core/query/ispn/IndexerCacheStore.java
    • -48
    • +0
    ./impl/core/query/ispn/LocalIndexCacheStore.java
    • -181
    • +0
    ./impl/core/query/ispn/RsyncIndexInfos.java
  1. … 277 more files in changeset.
JCR-2156: Included fixes to properly support triggerEventsForDescendantsOnRename = false in case of a session move which is required to fix ECMS-5225

  1. … 7 more files in changeset.
JCR-2125: Included fixes to properly support triggerEventsForDescendantsOnRename = false in case of a session move which is required to fix ECMS-5225

  1. … 7 more files in changeset.
JCR-2156: Code cleanup and perf fix

  1. … 1 more file in changeset.
JCR-2156: Unexpected behavior while importing data with IMPORT_UUID_COLLISION_REPLACE_EXISTING

  1. … 11 more files in changeset.
JCR-2125: cleanup the code

JCR-2125: merged last changes

  1. … 1 more file in changeset.
JCR-2125: Unexpected behavior while importing data with IMPORT_UUID_COLLISION_REPLACE_EXISTING

  1. … 11 more files in changeset.
JCR-2144 : Backup service: exceptions with huge dataset

  1. … 12 more files in changeset.
JCR-2094 : Backup service: exceptions with huge dataset

  1. … 11 more files in changeset.
JCR-2094 :Backup service: exceptions with huge dataset

  1. … 11 more files in changeset.
JCR-2085: Propagation of permission from parent to children

  1. … 3 more files in changeset.
JCR-2084: Fixed acl test

  1. … 2 more files in changeset.