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
level=error ts=2025-01-06T22:41:49.341622742Z caller=memcached_client.go:183 msg="error setting memcache servers to host" host=grafana-tempo-memcached err="lookup _memcached-client._tcp.grafana-tempo-memcached on 10.43.0.10:53: read udp 10.42.5.127:47106->10.43.0.10:53: read: connection refused"
level=warn ts=2025-01-06T22:42:54.346528774Z caller=memcached_client.go:257 msg="error updating memcache servers" err="lookup _memcached-client._tcp.grafana-tempo-memcached on 10.43.0.10:53: read udp 10.42.5.127:56758->10.43.0.10:53: read: connection refused"
level=warn ts=2025-01-06T22:43:54.346734867Z caller=memcached_client.go:257 msg="error updating memcache servers" err="lookup _memcached-client._tcp.grafana-tempo-memcached on 10.43.0.10:53: read udp 10.42.5.127:55790->10.43.0.10:53: read: connection refused"
level=error ts=2025-01-06T22:44:04.402619173Z caller=main.go:122 msg="error running Tempo" err="failed to init module services: error initialising module: store: failed to create store: unexpected error from ListObjects on tempo-trace: Get \"http://monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000/tempo-trace/?delimiter=%2F&encoding-type=url&prefix=\": dial tcp: lookup monitoring-minio.cattle-monitoring-system.svc.cluster.local on 10.43.0.10:53: read udp 10.42.5.127:32852->10.43.0.10:53: read: connection refused"
The 10.43.0.10 is the IP of kube-dns service.
Coredns is up and all other non-grafana charts deployed in the cluster do not have this issue
including mimir from bitnami.
A curl to monitoring-minio from any other pod with curl gives something similar to the below which shows a successful connection
> curl -v http://monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000/tempo-trace/?delimiter=%2F&encoding-type=url&prefix=
* Host monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000 was resolved.
* IPv6: (none)
* IPv4: 10.43.89.23
* Trying 10.43.89.23:9000...
* Connected to monitoring-minio.cattle-monitoring-system.svc.cluster.local (10.43.89.23) port 9000
> GET /tempo-trace/?delimiter=%2F HTTP/1.1
> Host: monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000
> User-Agent: curl/8.6.0
> Accept: */*>< HTTP/1.1 403 Forbidden
< ...truncated...
<?xml version="1.0" encoding="UTF-8"?>* Connection #0 to host monitoring-minio.cattle-monitoring-system.svc.cluster.local left intact<Error><Code>AccessDenied</Code><Message>Access Denied.</Message><BucketName>tempo-trace</BucketName><Resource>/tempo-trace/</Resource><RequestId>18183C29DDA7405B</RequestId><Hos
>
Even the pods that are active like gateway cannot reach any service/pod and cannot be reached either eg.
From within gateway pod.
~ $ curl -v http://monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000/tempo-trace/?delimiter=%2F&encoding-type=url&prefix=
~ $ /bin/sh: encoding-type=url: not found
* Could not resolve host: monitoring-minio.cattle-monitoring-system.svc.cluster.local
* Could not resolve host: monitoring-minio.cattle-monitoring-system.svc.cluster.local
* closing connection #0
curl: (6) Could not resolve host: monitoring-minio.cattle-monitoring-system.svc.cluster.local
[2]+ Done(127) encoding-type=url
[1]+ Done(6) curl -v http://monitoring-minio.cattle-monitoring-system.svc.cluster.local:9000/tempo-trace/?delimiter=%2F
Reaching memcached From a 3rd party pod
> curl -v grafana-tempo-memcached.cattle-monitoring-system.svc.cluster.local:11211
* Host grafana-tempo-memcached.cattle-monitoring-system.svc.cluster.local:11211 was resolved.
* IPv6: (none)
* IPv4: 10.43.233.164
* Trying 10.43.233.164:11211...
* connect to 10.43.233.164 port 11211 from 10.42.2.17 port 42042 failed: Connection refused
* Failed to connect to grafana-tempo-memcached.cattle-monitoring-system.svc.cluster.local port 11211 after 12 ms: Couldn't connect to server* Closing connectioncurl: (7) Failed to connect to grafana-tempo-memcached.cattle-monitoring-system.svc.cluster.local port 11211 after 12 ms: Couldn't connect to server
Reaching bitnami mimir's memcached however (from a 3rd party pod) is not a problem—whereas it's also a connection refused error when i tried it for grafana's mimir-distributed.
> curl -v bitnami-mimir-memcachedchunks.cattle-monitoring-system.svc.cluster.local:11211
* Host bitnami-mimir-memcachedchunks.cattle-monitoring-system.svc.cluster.local:11211 was resolved.
* IPv6: (none)
* IPv4: 10.43.126.169
* Trying 10.43.126.169:11211...
* Connected to bitnami-mimir-memcachedchunks.cattle-monitoring-system.svc.cluster.local (10.43.126.169) port 11211
> GET / HTTP/1.1
> Host: bitnami-mimir-memcachedchunks.cattle-monitoring-system.svc.cluster.local:11211
> User-Agent: curl/8.6.0
> Accept: */*>* Empty reply from server
* Closing connection
curl: (52) Empty reply from server
I'll just focus on tempo for this issue as I already gave up on mimir and deployed bitnami
chart(successfully) instead.
What is the expected behavior?
Application workloads come up and can reach each other.
What do you see instead?
Connection refused in pod-service communication.
Describe the bug
There seems to a network config preventing inbound and outbound communication among the pods. eg
The below workloads won't come up because they can't reach dependent services.
Sampling the
compactor
logs.The
10.43.0.10
is the IP of kube-dns service.Coredns is up and all other non-grafana charts deployed in the cluster do not have this issue
including mimir from bitnami.
A curl to
monitoring-minio
from any other pod with curl gives something similar to the below which shows a successful connectionEven the pods that are active like
gateway
cannot reach any service/pod and cannot be reached either eg.From within
gateway
pod.mimir-distributed
.Cluster Details
architecture: amd64
version: v1.31.3
provider: k3s
Chart Details
Helm Values
The text was updated successfully, but these errors were encountered: