[histogramCE] "Data distribution values must be monotonically increasing" with non-ascii character, x64

XMLWordPrintableJSON

    • Type: Bug
    • Resolution: Unresolved
    • Priority: Major - P3
    • None
    • Affects Version/s: None
    • Component/s: None
    • Query Optimization
    • ALL
    • Hide 
      db.foo.drop();
const string_docs = [];

for (let i = 32; i < 127; i++) {
	string_docs.push({a: String.fromCharCode(i)});
}

string_docs.push({a: 'Алабала'});

db.foo.insert(string_docs);
db.foo.runCommand({analyze: "foo", key: "a"});
      Show
      db.foo.drop(); const string_docs = []; for (let i = 32; i < 127; i++) { string_docs.push({a: String .fromCharCode(i)}); } string_docs.push({a: 'Алабала' }); db.foo.insert(string_docs); db.foo.runCommand({analyze: "foo" , key: "a" });
    • None
    • 3
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      If there is a non-ascii character in the dataset along with ascii characters, analyze will fail as follows. This issue seems to be reproducible only on x64, and not on arm.

      Enterprise test> db.foo.runCommand({analyze: "foo", key: "a"});
MongoServerError: PlanExecutor error during aggregation :: caused by :: Data distribution values must be monotonically increasing, however encountered 126 before -48.43823654584429

              Assignee:
              Unassigned
              Reporter:
              Philip Stoev
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated: