-
Notifications
You must be signed in to change notification settings - Fork 670
kube-weave weave-npc failing to start #2653
Comments
Thanks for the report @chrislovecnm. Very odd that it fails on only one of the machines - I suspect you would have mentioned it already if it were the case, but is there anything different about that one? |
@chrislovecnm two possibilities come to mind so far, most likely first:
I wonder if perhaps we have somehow ended up with an orphaned ulogd process lying around on that machine?
|
All the machines are the same ami, let me poke around. We have not been able to recreate it ;( |
Thanks. I'm going to try to see if I can get a ulogd to survive the weave-npc pod exiting - that would clash once k8s restarts the pod. |
Winner winner chicken dinner... ulogd is already running admin@ip-172-20-155-50:~$ ps axu | grep ulogd
root 1911 0.0 0.0 13528 624 ? S< Nov17 0:00 /usr/sbin/ulogd -v
admin 25256 0.0 0.0 12732 2224 pts/0 S+ 15:18 0:00 grep ulogd docker info admin@ip-172-20-155-50:~$ sudo docker info
Containers: 17
Running: 12
Paused: 0
Stopped: 5
Images: 7
Server Version: 1.11.2
Storage Driver: overlay
Backing Filesystem: extfs
Logging Driver: json-file
Cgroup Driver: cgroupfs
Plugins:
Volume: local
Network: null host bridge
Kernel Version: 4.4.26-k8s
Operating System: Debian GNU/Linux 8 (jessie)
OSType: linux
Architecture: x86_64
CPUs: 8
Total Memory: 31.42 GiB
Name: ip-172-20-155-50
ID: YP6U:EMZ2:PGZU:7TFB:N7HC:EF66:R3GB:I66C:UWJQ:KFIH:YPW3:AIKV
Docker Root Dir: /var/lib/docker
Debug mode (client): false
Debug mode (server): false
Registry: https://index.docker.io/v1/
WARNING: No swap limit support
WARNING: No kernel memory limit support |
So killed
|
See #2617, which will be fixed in 1.8.1. |
Filed #2654 to help with the pid0 stuff that we MAY have ran into |
Further analysis: weave-kube runs with |
Hi there
I am launch a
kops
weave cluster with private networking, @bboreham knows what I mean. We have 20 nodes, and on one of themweave-npc
is failing to start. The issue is that ulogd is failing with:I have launched a pod onto the impacted node with a nodeSelector and am running from a single weave-npc pod.
Here is a full strace for ulogd
The text was updated successfully, but these errors were encountered: