Like Prometheus, but for logs.
GNU Affero General Public License v3.0
23.72k
stars
3.43k
forks
source link
Promtail 3.2.0 crash on docker with panic: runtime error: invalid memory address or nil pointer dereference #14405
Open
lluisd opened 2 weeks ago
Describe the bug On starting Promtail 3.2.0 I get this error, it was working fine with previous version.
panic: runtime error: invalid memory address or nil pointer dereference [signal SIGSEGV: segmentation violation code=0x1 addr=0x88 pc=0x2bf5735] goroutine 132 [running]: github.com/prometheus/prometheus/discovery/moby.(*DockerDiscovery).refresh(0xc000894440, {0x47b79d0, 0xc0007b3950}) /src/loki/vendor/github.com/prometheus/prometheus/discovery/moby/docker.go:284 +0xcd5 github.com/prometheus/prometheus/discovery/refresh.(*Discovery).refresh(0xc00065bf20, {0x47b79d0, 0xc0007b3950}) /src/loki/vendor/github.com/prometheus/prometheus/discovery/refresh/refresh.go:112 +0xcf github.com/prometheus/prometheus/discovery/refresh.(*Discovery).Run(0xc00065bf20, {0x47b79d0, 0xc0007b3950}, 0xc000312900) /src/loki/vendor/github.com/prometheus/prometheus/discovery/refresh/refresh.go:68 +0x66 created by github.com/prometheus/prometheus/discovery.(*Manager).startProvider in goroutine 1 /src/loki/vendor/github.com/prometheus/prometheus/discovery/manager.go:302 +0x2ec
I use that docker-compose for Promtail
`version: "3.9"
services: promtail: image: grafana/promtail:3.2.0 container_name: promtail command:
networks: loki_net: name: loki_net external: true`
and that promtail configuration: `server: http_listen_address: 0.0.0.0 http_listen_port: 9030 enable_runtime_reload: true
positions: filename: /tmp/positions.yaml
clients:
job_name: container_logs docker_sd_configs:
job_name: container_logs_nuc docker_sd_configs:
job_name: container_logs_minipc docker_sd_configs:
Environment: