Packstack rabbitmq restart, erlang_vm restart needed

asked 2018-02-26 03:17:37 -0600

AndyW gravatar image

Hi Am runnning Packstack all-in-one on a laptop as a small demo vehicule (inside a VM 4vCPU 10GB, 30G disk).
I suspend the laptop from time to time (yes, I realise this is not common use of openstack...) , it seems after a long suspend (a few days) that the RabbitMQ messaging on the system no longer works well, I see beam.smp consuming 100% CPU and Horizon or CLI is not responding.

What usually works to restart it is

sudo rabbitmqctl stop_app

sudo rabbitmqctl reset

sudo rabbitmqctl start_app

occasionally doing this throws up the additional message:
erlang VM I/O system is damaged , restart needed erlang_vm_restart needed

what is link between erlang and rabbitmq ?

Is there a simple command to do this erlang restart , or do I need to reboot the VM? Actually, even after this erlang error message during rabbitmq restart , Packstack installation seems to be working , horizon or CLI is responding correctly again , so is this a minor error message I can ignore?


edit retag flag offensive close merge delete


Erlang is the language used for writing RabbitMQ.

Message queues rely on precise message timestamps. Perhaps RabbitMQ is confused by the sudden change in the system time after the suspend.

Bernd Bausch gravatar imageBernd Bausch ( 2018-03-20 18:33:49 -0600 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2018-03-20 12:27:24 -0600

AndyW gravatar image

I've noticed in the above use of Packstack on a laptop (ubuntu with suspend,resume cycles) that when I resume the laptop, the VM running Packstack is inaccessible for about 5 minutes (via Horizon or CLI, I also have an ssh connection to the Centos VM which hangs as well) but finally things always seem to unblock and VM is running again ok. beam.smp is often running at 100% cpu or more during this phase, I suspect that rabbitmq is taking a long time to clean up old queues (old timestamps?). The only thing that seems to stop working at a certain resume cycle is console service (vnc) (error 1006) , not sure that problem is linked to (strange) way am running it.

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2018-02-26 03:17:37 -0600

Seen: 1,196 times

Last updated: Mar 20 '18