Skip to content

zincati::strategy lock-manager pre-reboot failure: server-side error, code 500: all 1 semaphore slots currently locked #81

@sgohl

Description

@sgohl

Hi,

I'm not sure this is a bug; there's so little documentation on how this works

I use Fedora CoreOS with a zincati config like this:

[identity]
rollout_wariness = 1.0
group = "default"

[updates]
strategy = "fleet_lock"

[updates.fleet_lock]
base_url = "http://192.168.55.200:3333/"

[[updates.fleet_lock.window]]
days = [ "Mon" ]
start_time = "01:00"
length_minutes = 30

And my airlock config is this:

            [status]
            enabled = true
            address = \"0.0.0.0\"
            port = 2222
            tls = false

            [service]
            address = \"0.0.0.0\"
            port = 3333
            tls = false

            [etcd3]
            endpoints = [ \"http://dutch_etcd:2379\" ]

            [lock]
            default_group_name = \"default\"
            default_slots = 99

            [[lock.groups]]
            name = \"cluster\"
            slots = 1

In belief that the default group has 99 slots, so I think 99 machines can reboot simultaneously.
However, I always get this error message

Sep 18 09:59:20 rdbops zincati[3964]: [ERROR zincati::strategy] lock-manager pre-reboot failure: server-side error, code 500: all 1 semaphore slots currently locked

How could I query the airlock api for possible servers that caused the semaphore lock?
Do I possibly misunderstand what a slot is?
What if default_slots is missing? Does it mean "only one server can reboot" or "no limit on servers that can reboot" ?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions