charm-init in metallb-0 ERROR unable to connect to API: dial tcp xxx:17070: i/o timeout
Bug #2056122 reported by
Michael Fischer
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| MetalLB Operator |
Invalid
|
Undecided
|
Unassigned | ||
Bug Description
running command "juju deploy metallb --channel 1.29/stable --trust"
The metallb pod is stuck, unable to communicate outside of the cluster.
ERROR unable to connect to API: dial tcp 192.168.
Charmed Kubernetes was deployed in IPVS proxy mode per the documentation at https:/
Revision history for this message
| Michael Fischer (michaelandrewfischer) wrote | #1 |
| Changed in operator-metallb: | |
| status: | New → Invalid |
To post a comment you must log in.
The issue was caused by Calico's default cidr value being in conflict with the subnet used in the MAAS region.
In the overlay, adding a calico option for cidr config value with a non-conflicting subnet of host machines has resolved the issue preventing the cluster from communicating outside of the cluster.