Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

On two Supermicro 847E16-RJBOD1 enclosures, I have trouble accessing SATA disks on the rear backplane with 6gb/s. The only significant difference between rear and front backplanes is total cable length from HBA to backplane, which is about 1m + 30cm vs. 1m + 70cm.

SAS signalling shouldn't be affected at these lengths, and I always assumed HBA and backplane will connect with SAS signalling whatever disk type attached. Is this assumption wrong? Do SATA disks limit HBA to backplane cable lengths?

share|improve this question
up vote 3 down vote accepted

SATA and SAS are point-to-point links. So the total distance is irrelevant, only the link lengths.

The SAS Protocol is used to connect to SAS devices. SATA is used to talk to SATA devices. SAS HBAs and Active Backplanes (aka SAS Expanders) must be able to speak SATA, so it can communicate to SATA drives; and this SATA connection is sent back over the SAS backhauls via the SATA Tunneling Protocol.

I haven't checked the specs yet, but it's likely that the expanders are SAS 1.0 or 1.1 and don't support 6Gb SATA (they would only support 1.5Gb SATA). I have further seen issues with these "older" generation SuperMicro Expanders when using SATA disks the whole chip slows to 1.5Gb instead of just the one lane. This would mean you could only get 1.5Gbx4 to all the drives in the rear of the chassis, and only under ideal conditions.

share|improve this answer
Thanks for confirming my original assumption. But the chips do negotiate 6 gb/s, I have to throttle down to 3 gb/s to avoid disk connectivity loss. The whole enclosure is 6 gb/s capable (hence the E16 in name). It must be a cable length problem, or a very specific expander incompatibility. I consider removing the dedicated external connection for the rear backplane and instead wire the backplanes. That would mean loss of parallelity but increase in stability. – korkman Oct 4 '12 at 14:10
It's definitely not a length issue, but may be a damaged cable/board. Just looked it over and it's the newer gen LSI ICs, which I haven't heard of any compatibility issues. – Chris S Oct 5 '12 at 2:01
Actually my idea of cable length issues was very wrong from the beginning: the rear backplane has shorter cabling than the front backplane. I can't test much on these in-production arrays right now, but I can rule out bad cables because in total 4 internal and 4 external cables were tested. Must be a hard drive incompatibility. – korkman Oct 8 '12 at 17:21

The limit doesn't just exist on the HBA side but on the disk side as well, and with 1.7m you are well beyond the specification limit of 1m.

share|improve this answer
Both SAS and SATA are point-to-point connections, each of the connection must be less than the electrical limit. Also, most of his connections are SAS, which are supposed to be able to run 10m (though half that is a more realistic number). – Chris S Oct 4 '12 at 13:25

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.