Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Simplify serializable test to avoid mystery deadlock | Matthew Draper | 2016-10-07 | 1 | -23/+21 |
| | |||||
* | Make PG deadlock error more deterministic | Matthew Draper | 2016-10-01 | 1 | -9/+12 |
| | | | | | We've seen occasional Travis failures mentioning deadlocks. I think they're escaping from this test. | ||||
* | normalizes indentation and whitespace across the project | Xavier Noria | 2016-08-06 | 1 | -6/+6 |
| | |||||
* | applies new string literal convention in activerecord/test | Xavier Noria | 2016-08-06 | 1 | -7/+7 |
| | | | | | The current code base is not uniform. After some discussion, we have chosen to go with double quotes by default. | ||||
* | The problem isn't the detection but the deadlock itself | David Heinemeier Hansson | 2016-08-04 | 1 | -2/+2 |
| | |||||
* | Introduce new ActiveRecord transaction error classes | Erol Fornoles | 2016-05-24 | 1 | -2/+27 |
| | |||||
* | Add AR::TransactionSerializationError for transaction serialization failures ↵ | Erol Fornoles | 2016-05-21 | 1 | -0/+72 |
or deadlocks |