3 ms·

Well, they did end up making it a pretty good database. They saved the good parts (great clients, dev experience) and got rid of the bad (original storage engin

Well, they did end up making it a pretty good database. They saved the good parts (great clients, dev experience) and got rid of the bad (original storage engine, instability) add and have a legit database. Good on them.

winrid 5y ago
While I want to agree, until this year there were still deadlock issues that could take down whole clusters [0][1]. [0] https://jira.mongodb.org/browse/SERVER-57476 https://jira.mongodb.org/browse/SERVER-57476 [1] https://jira.mongodb.org/browse/SERVER-56779 https://jira.mongodb.org/browse/SERVER-56779
- tchvil 5y ago
  A JIRA user is needed to see them I guess. Do you mind to tell what are the issues ?
  - ddorian43 5y ago
    I could open the issues in icognito mode. 1: Operation may block on prepare conflict while holding oplog slot, stalling replication indefinitely 2: Do not use the collection distributed lock for chunk merges
    - winrid 5y ago
      Yep. #2 is kind of a hush-hush way of them fixing a bug that can take a lock on a collection during a chunk merge and fail to release it, resulting in writes to primaries to stop until they are kill -9'd.
5e92cb50239222b 5y ago
This was posted not much more than a year ago. https://jepsen.io/analyses/mongodb-4.2.6 https://jepsen.io/analyses/mongodb-4.2.6 https://news.ycombinator.com/item?id=23290844 https://news.ycombinator.com/item?id=23290844 https://news.ycombinator.com/item?id=23285249 https://news.ycombinator.com/item?id=23285249
- DonnyV 5y ago
  Looks like it was a bug and Mongodb fixed it.