Re: ceph-deploy progress and CDS session
One thing that comes to mind is the ability to create (or activate) osd's with a custom crush specification from (say) a supplied file. Regards Mark On 03/08/13 06:02, Sage Weil wrote: There is a session at CDS scheduled to discuss ceph-deploy (4:40pm PDT on Monday). We'll be going over what we currently have in backlog for improvements, but if you have any opinions about what else ceph-deploy should or should not do or areas where it is problematic, please reply to this thread to let us know what you think, and/or join the CDS discussion hangout. For those who haven't noticed, we now have a full-time devoloper, Alfredo Deza, who is working on ceph-deploy. He's been making huge progress over the last couple of weeks improving error reporting, visibility into what ceph-deploy is doing, and fixing various bugs. We have a long list of things we want to do with the tool, but any feedback from users is helpful to make sure we're working on the right things first! -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ceph-users] ceph-deploy progress and CDS session
Hi, First I would like to state that with all its limitiation, I have managed to build multiple clusters with ceph-deploy and without it, I would have been totally lost. Things that I feel would improve it include: A debug mode where it lists everything it is doing. This will be helpful in the future when I move to a more integrated tool then ceph-deploy, as I could see exactly how ceph-deploy built my test cluster. To understand more types of linux storage devices. I have spent hours trying to make it understand multipath devices, as I happen to have a large number of these in my lab, but so far I have not made it work. Really good documentation on all the ceph-deploy options. Lastly, this is not just a ceph-deploy thing, but documentation explaining how things boot up, and interact. Ceph-deploy depends on tools like ceph-disk to mount OSD disks on the servers during boot, and I learned the hard way that if a OSD is on a LUN that is seen by more then one OSD node, you can corrupt data, as each OSD node tries to mount all the ODS it can find. There is a session at CDS scheduled to discuss ceph-deploy (4:40pm PDT on Monday). We'll be going over what we currently have in backlog for improvements, but if you have any opinions about what else ceph-deploy should or should not do or areas where it is problematic, please reply to this thread to let us know what you think, and/or join the CDS discussion hangout. Thank Eric -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
ceph-deploy progress and CDS session
There is a session at CDS scheduled to discuss ceph-deploy (4:40pm PDT on Monday). We'll be going over what we currently have in backlog for improvements, but if you have any opinions about what else ceph-deploy should or should not do or areas where it is problematic, please reply to this thread to let us know what you think, and/or join the CDS discussion hangout. For those who haven't noticed, we now have a full-time devoloper, Alfredo Deza, who is working on ceph-deploy. He's been making huge progress over the last couple of weeks improving error reporting, visibility into what ceph-deploy is doing, and fixing various bugs. We have a long list of things we want to do with the tool, but any feedback from users is helpful to make sure we're working on the right things first! sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html