[Condor-users] Condor fault tolerance

[Janito Ferreira Filho <jvffprog@xxxxxxxxxxx>]

Hello,

what kind of fault tolerance does Condor offer? If a job is running on a node, and a power failure occurrs, does Condor detect that the node is off and does it re-start it on another node? I tried this on a cluster, and even though Condor detected the node restarted, it still listed the job as running, even though it wasn't. Is this common behavior? Thanks,

JVFF

---

See all the ways you can stay connected to friends and family