UTHPC - Unavailable cluster nodes and job queue delays – Intsidentide üksikasjad

Osaline katkestus

Unavailable cluster nodes and job queue delays

Lahendatud
Halvenenud jõudlus
Algas 2 kuu eestKestis 7 päeva

Mõjutatud

rocket.hpc.ut.ee

Halvenenud jõudlus alates 10:11 AM kuni 6:35 AM

Värskendused
  • Lahendatud
    Lahendatud
    This incident has been resolved.
  • Jälgimisel
    Jälgimisel

    A workaround is currently being tested and deployed

  • Uuendus
    Uuendus

    The issue has been tracked to a recent bug in the RedHat kernel. Our team is working on a workaround

  • Tuvastatud
    Tuvastatud

    We are currently experiencing severe delays in the job queue and significant number of nodes are currently down. The technical team is actively working to resolve this incident.