Hi Bob, The way I workaround that issue is checking if the file for each hard disk already exists, if not, I create it. Take a look here: https://github.com/Nafiux/ceph-workshop/blob/master/Vagrantfile#L41-L57 That allows me to do vagrant up/halt multiple times without any problem. Thanks! On Fri, Feb 14, 2020 at 10:42 AM Bob Wassell <bob@softiron.com> wrote:

I’ve found that having > 1 drive controller in vagrant is problematic, although i do agree it would be preferred logically for > 1 osd per node. That being said, have you all run into the problems that i’ve seen with > 1 hdd controller? Namely, the inability to use vagrant's up command after a machine has been provisioned and halted. At least from my perspective, once you provision you can’t use vagrant for up / halt and just use VB commands. VB does not check for drive controller’s existence, and gets confused. Is that consistent with what you’ve found? I’ve spent some time on this issue, so i’m curious as to if you’ve found any “magic grits” to solve it. This issue is on both *nix and windows too btw. (It is certainly a VB issue, not vagrant’s falt imo)

Thank you in advance for your thoughts, anything that jumpstarts newbs past a ceph installation is a great thing!

Bob

[image: SoftIron Logo]

*Bob Wassell*

Solutions Architect | Soft <http://www.softiron.com>Iron <http://www.softiron.com>

+1 610 505 9861

bob@softiron.com

On Feb 14, 2020, at 11:50 AM, Ignacio Ocampo <nafiux@gmail.com> wrote:

Hi all,

A group of friends and my self are documenting a hands-on workshop about Ceph https://github.com/Nafiux/ceph-workshop, for learning purposes.

The idea is to provide visibility step-by-step on common scenarios, from basic usage, to disaster and recovery scenarios.

We will hold a workshop next weekend, and some of the ideas for learning are:

- Configure and learn how to consume Block Storage Devices - Configure and learn how to consume File Systems Storage - Configure and learn how to consume Object Storage - Simulate a disaster and recovery event by killing a node and setting up a new one - Simulate a node migration

Any idea or feedback is welcome! From the way we've decided to install Ceph (ceph-ansible), the way we're configuring the cluster, suggestions on basic day-to-day operations we should learn, etc.

Thanks for your support!

-- Ignacio Ocampo _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io

-- Ignacio Ocampo