The logs of the "rkt-api" unit are a good place to check when you get rpc
errors because that's the service it's trying to communicate with.
I recall getting a similar error with older versions of rkt due to a bug in
the api-service leading to a panic. Updating to a newer rkt version (which
can be done by updating to a newer CoreOS release) might help, but if not
the api-service logs would help us figure out what's going on.
Best,
Euan
On Tuesday, August 9, 2016 at 7:42:09 PM UTC-7, Stefan Vetter wrote:
>
> Maybe I should just switch back to docker as it seams rkt does not work
> with kubelet properly... :-(
>
> Am Dienstag, 9. August 2016 18:09:00 UTC+2 schrieb Stefan Vetter:
>>
>> Hi all,
>>
>> first of all, I still have error messages when using a downloaded
>> kubelet, but pods get deployed.
>>
>>
>> Using kubernetes 1.3.2 and rkt 1.10.1, coreos 1068.8.0
>>
>>
>> Here are my findings (kubelet-wrapper):
>>
>> Aug 09 14:35:09 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:35:09.942522 4523 kubelet.go:2594] Failed cleaning pods: couldn't
>> list pods: rpc error: code = 4 desc = "context deadline exce
>>
>> eded"
>>
>> Aug 09 14:35:09 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:35:09.942455 4523 kubelet.go:2247] Error listing containers:
>> &errors.errorString{s:"couldn't list pods: rpc error: code = 4 de
>>
>> sc = \"context deadline exceeded\""}
>>
>> Aug 09 14:35:04 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:35:04.942120 4523 generic.go:197] GenericPLEG: Unable to retrieve
>> pods: couldn't list pods: rpc error: code = 4 desc = "contex
>>
>> t deadline exceeded"
>>
>> Aug 09 14:35:04 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:35:04.941377 4523 kubelet.go:927] Container garbage collection
>> failed: rpc error: code = 4 desc = "context deadline exceeded"
>>
>> Aug 09 14:35:04 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:35:04.941322 4523 rkt.go:1866] rkt: Failed to list pods: rpc error:
>> code = 4 desc = "context deadline exceeded"
>>
>> Aug 09 14:33:04 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:33:04.944400 4523 manager.go:235] Docker container factory
>> registration failed: failed to validate Docker info: failed to dete
>>
>> ct Docker info: Cannot connect to the Docker daemon. Is the docker daemon
>> running on this host?.
>>
>> Aug 09 14:33:04 node3.cluster1.kubernetes.cluster.int sudo[4487]: E0809
>> 14:33:04.940714 4523 kubelet.go:933] Image garbage collection failed:
>> unable to find data for container /
>>
>>
>>
>>
>> Here are my findings (without wrapper):
>>
>>
>> E0809 16:00:35.278281 46910 pod_workers.go:183] Error syncing pod
>> ff3cfae6-5e48-11e6-9c00-001999c214fd, skipping: failed to SyncPod: couldn't
>> list pods: rpc error: code = 4 desc = "context deadline exceeded"
>>
>> E0809 16:02:34.972125 46910 kubelet.go:2247] Error listing containers:
>> &errors.errorString{s:"couldn't list pods: rpc error: code = 4 desc =
>> \"context deadline exceeded\""}
>>
>> E0809 16:02:34.972156 46910 kubelet.go:2594] Failed cleaning pods:
>> couldn't list pods: rpc error: code = 4 desc = "context deadline exceeded"
>>
>> E0809 16:02:37.278562 46910 generic.go:197] GenericPLEG: Unable to
>> retrieve pods: couldn't list pods: rpc error: code = 4 desc = "context
>> deadline exceeded"
>>
>> E0809 16:02:39.183965 46910 rkt.go:65] detectRktContainers:
>> listRunningPods failed: rpc error: code = 2 desc = "grpc: the client
>> connection is closing"
>>
>>
>>
>>
>> Thanks for any very appreaciated help!
>>
>>
>>
>> Stefan
>>
>
