Open issues

Order by Priority
Order by Priority
  1. Bug
    MCOL-5852GROUP_CONCAT in query uses enormous amount of RAM causing OOM to kill PrimProc
  2. Bug
    MCOL-5801Queries Hung | PrimProc <-> Storagemanager | Unknown error code in response
  3. Bug
    MCOL-930Cannot execute queries
  4. Bug
    MCOL-5879mariadbd crashing randomly when running "SELECT calShowPartitions("
  5. New Feature
    MCOL-5874Evaluation for Cross Engine Join only single-table queries processing
  6. New Feature
    MCOL-5806Dynamic horizontal scaling for Small to Medium-sized installations.
  7. Bug
    MCOL-5756add forced select_handler mode for all cases
  8. Task
    MCOL-5165CAL0055: ERROR: ExeMgr has caught an exception. Resource temporarily unavailable
  9. New Feature
    MCOL-5797New memory consumption accounting mechanism to prevent OOM events-related crashes.
  10. Bug
    MCOL-5460concurrent CTE operation excessive mutex
  11. New Feature
    MCOL-4250Correlated Subquery, particularly inside and OUTER JOIN ON clause, is currently not supported
  12. New Feature
    MCOL-5466In place upgrades across major versions - support large columnstore deployment upgrades
  13. New Feature
    MCOL-4218Remove NFS/EFS/ Filestore prereq for S3 storage
  14. Bug
    MCOL-5889SubAdapterStep::execute() MCS-2035 Fix Requested
  15. Bug
    MCOL-4942Different results compared to innodb with union all and subselect with null in select and impossible where condition
  16. Task
    MCOL-5493Races and UB's in Columnstore (sanitizers)
  17. New Feature
    MCOL-5474Mutex lock contention doing block cache lookup when loading a block
  18. New Feature
    MCOL-5473Mutex lock contention doing VSS lookup loading a block
  19. Bug
    MCOL-5674CREATE OR REPLACE TABLE not supported, but it drops the table
  20. Bug
    MCOL-5892FATAL ERROR: ExeMgr has allocated too much memory! Percent allocation-96, allowed-95. ExeMgr is restarting
  21. New Feature
    MCOL-5876Evaluation for Cross Engine Join only multi-table queries processing
  22. Bug
    MCOL-5019creating different keys with cskeys on all nodes can break cluster(key not distributed properly)
  23. Task
    MCOL-5831Burza: replication mode
  24. Task
    MCOL-5898Long running test
  25. Task
    MCOL-5638Improve Client Error Messaging and async broadcasting config.
  26. Bug
    MCOL-5789`caldroppartitions` removes files but leaves ExtentMap entries causing read errors
  27. Bug
    MCOL-5760Dump failing with "Internal error: Lost connection to ExeMgr. Please contact your administrator (1815)" after upgrading to version 23.10.1
  28. Bug
    MCOL-5544StatisticsManager crashes on PP startup unable to read the file.
  29. Bug
    MCOL-5498Potential memory leak in GROUP_CONCAT()/JSON_ARRAYAGG() processing
  30. New Feature
    MCOL-5702Auto Resolution of DBRM Table Locks When Orphaned
  31. New Feature
    MCOL-5142Recursive CTE support
  32. Bug
    MCOL-5603Online alter can break data consistency
  33. Bug
    MCOL-5446CS Install/Upgrade chown mysql:mysql /var/lib/columnstore performance
  34. Task
    MCOL-5639Tooling/Process Improvements - How can we replicate/reproduce client issues quicker/easier
  35. Sub-Task
    MCOL-4945NOT IN subquery for some SELECTs and DMLs does not work as expected
  36. Bug
    MCOL-5540sudden massive "Invalid Range from HWM for lbid" was thrown
  37. New Feature
    MCOL-5504Error on unsupported function using CURRENT_TIMESTAMP on Columnstore
  38. Bug
    MCOL-5666PP freezes waiting for SM to answer over a socket
  39. Sub-Task
    MCOL-5727Deliver FoundationDB with CMAPI
  40. Bug
    MCOL-5557MariaDB columnstore table join innodb table caused ERROR 1815 (HY000): Internal error: MCS-2058: Unknown Error
  41. Sub-Task
    MCOL-4914Complete ColumnStore Cache Part 2
  42. Bug
    MCOL-5412Scalar vectoring code does not always work (certain where clause values combo)
  43. Bug
    MCOL-3329Pentaho server go in java out of memory cause Pentaho kettle plugin
  44. Bug
    MCOL-5088Docker: Cluster creation did not finish waiting for PM1 to be initialized -develop only
  45. Bug
    MCOL-3412MariaDB Columnstore 1.2.3-1 Cluster with replication for um1/um2 fails HA
  46. Bug
    MCOL-3300Next executed Query is returning wrong results after Error reception from some previous executed Query
  47. Bug
    MCOL-1770Columnstore restart fails when the Swap is over threshold
  48. Bug
    MCOL-1851CASE statement cannot handle null values
  49. New Feature
    MCOL-5781Create columnstore_repair | post dbrm restore or crash
  50. New Feature
    MCOL-5692mcs_backup_manager | dbrm_restore
Refresh results
Uploaded image for project: 'MariaDB ColumnStore'
  1. MariaDB ColumnStore
  2. MCOL-5852

GROUP_CONCAT in query uses enormous amount of RAM causing OOM to kill PrimProc

Details

    • Bug
    • Status: In Progress (View Workflow)
    • Blocker
    • Resolution: Unresolved
    • 23.10.2
    • 25.02
    • ExeMgr, PrimProc
    • 2025-1

    Description

      When a query has the `GROUP_CONCAT` keyword, it executes for a short while, and dies off with the below error:

      ERROR 1815 (HY000): Internal error: InetStreamSocket::readToMagic: Remote is closed

      This is due to `OOM` killing the `PrimProc` service.

      Parameter values of note, we have:

      mcsSetConfig HashJoin TotalUmMemory 50%  --previous value 25%, same issue.
      		 
      mcsSetConfig HashJoin AllowDiskBasedJoin N --previous value Y, same issue.

      Attachments

        Activity

          Ascending order - Click to sort in descending order
          michael.amadi Michael Amadi created issue -
          michael.amadi Michael Amadi made changes -
          Field Original Value New Value
          Summary GROUP_CONCAT in query uses enormous amount of AM causing OOM to kill PrimProc GROUP_CONCAT in query uses enormous amount of RAM causing OOM to kill PrimProc
          ricardo.galvan Ricardo Galvan made changes -
          Labels performance performance triage
          julien.fritsch Julien Fritsch made changes -
          Assignee Leonid Fedorov [ JIRAUSER48443 ]
          julien.fritsch Julien Fritsch made changes -
          Priority Major [ 3 ] Critical [ 2 ]
          drrtuy Roman made changes -
          Assignee Leonid Fedorov [ JIRAUSER48443 ] Alexey Antipovsky [ JIRAUSER47594 ]
          drrtuy Roman made changes -
          Status Open [ 1 ] Confirmed [ 10101 ]
          alexey.antipovsky Alexey Antipovsky made changes -
          Status Confirmed [ 10101 ] In Progress [ 3 ]
          allen.herrera Allen Herrera made changes -
          Priority Critical [ 2 ] Blocker [ 1 ]
          julien.fritsch Julien Fritsch made changes -
          Fix Version/s 25.02 [ 29991 ]
          leonid.fedorov Leonid Fedorov made changes -
          Sprint 2025-1 [ 784 ]
          julien.fritsch Julien Fritsch made changes -
          Labels performance triage performance

          People

            alexey.antipovsky Alexey Antipovsky
            michael.amadi Michael Amadi
            Votes:
            4 Vote for this issue
            Watchers:
            5 Start watching this issue

            Dates

              Created:
              Updated:

              Git Integration

                Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.

                1 of 612