You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This commit was created on GitHub.com and signed with GitHub’s verified signature.
Chart 0.25.2 [ Corezoid 6.13.0 ]
Applications versions:
capi - 9.0.0.1
mult - 3.10.0.1
web - 6.13.0
http-worker - 5.0.0.1
usercode - 10.0.1
worker - 6.0.0.2
syncapi - 3.8.1
web_superadm - 2.6.3
conf_agent_server - 3.0.1
conf_agent_admin - 2.6.3
limits - 2.5.1
Updated applications
capi - 9.0.0.1
mult (conveyor_api_multipart) - 3.10.0.1
web (webadm) - 6.13.0
http-worker - 5.0.0.1
usercode (cce) - 10.0.1
worker - 6.0.0.2
conf_agent_server - 3.0.1
New features
Added the conveyor_task_store block to the capi, worker, http-worker and conf-agent-server configmaps — required by the Corezoid 6.13.0 task store. It points at the root conveyor database via POSTGRES_DBHOST (resolves to pgbouncer-service when global.db.bouncer is enabled); shard pools and the Redis cache are reused from the existing pgsql / redis2 settings
Added the global.usercode.sys_ptrace toggle (default true): adds the SYS_PTRACE capability to the usercode container so eu-stack / gdb can dump native stacks of a stuck cnode. Disable it on clusters with a baseline/restricted Pod Security admission policy.
postgres_schema bumped to 6.13.0: creates task_store_settings on the main conveyor database and tasks_payload + tasks_commit_log on every shard.