Back
Valerian Saliou
Load previous page…
re-create all kubernetes persistent volumes for
#crisp enrichment api +
#mirage databases, after finding out that they got created with a reclaim policy of delete instead of retain, putting the data at risk on the long term — TLDR if you use K8S always suffix the block storage class with -retain if your cloud provider supports it and if you don't want the data to disappear with a mistake
coffee late morning to chill down after intense 14 hours a day migration to kubernetes of
#crisp service for the last 3 days
start moving small
#crisp enrichment api from traditionally managed vps to infra as code on kubernetes to progressively reduce server mess
attempt to obtain lei from government website to open ibkr corporate account, and already fail at it
#life
finish migrating
#mirage kubernetes intel and nvidia gpu instances to scaleway, getting last-generation NVIDIA L40S + L4 GPUs, running much smoother now! (previously: old A40 and A16)
finally ready to migrate
#mirage kubernetes gpu cluster from vultr to scaleway!