GlusterFS and RDMA support
By joe
- 1 minutes read - 190 words[update] In 3.3.1/3.3.2
This appeared in the 3.3.0 docs. “NOTE: with 3.3.0 release, transport type ‘rdma’ and ‘tcp,rdma’ are not fully supported.” On page 133 of the Admin Guide. We’ve been noting breakage with support since the 3.0.x days. I think there were varying factions within the company that wanted pure tcp, and some wanted RDMA included. The latter is what HPC folks use for their storage. GlusterFS is going in a decidedly non-HPC direction, which is fine. No issues. Looks like this is a deprecation, and bit-rot may be setting in. Just a delay due to resourcing we are being told. I am guessing, haven’t seen anything official from Red Hat yet. Not sure if we will. This might kill at least one install for GlusterFS that we were working on with a customer. Not to mention being somewhat embarrassing to explain, as they bought Infiniband hardware with GlusterFS in mind (about a year ago), and have been getting ready to deploy after experimenting with 3.2.x. And another place where we thought it would be a good fit for about 1PB managed data. This is not evolving well for our customers. Ugh.