ceph-crash collector pods get restarted on every Rook operator upgrade
ceph-crash collector pods get restarted on every Rook operator upgrade: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). I'm running a Rook-Ceph cluster in production and recently bumped the Rook operator to a newer minor version. The Ceph image itself stayed exactly the same…
The task
I'm running a Rook-Ceph cluster in production and recently bumped the Rook operator to a newer minor version. The Ceph image itself stayed exactly the same — only the operator changed. Right after the upgrade I noticed that every rook-ceph-crash-collector- pod on every node was…
Part of XiaomiMiMo/MiMo-V2.6-RL-oss.