> Have been there and gave up at the end[1]. Could reproduce (even though
> it took a bit longer) under most Linux versions (incl. using latest LSI
> drivers) and LSI 3081E-R HBA.
> Is it just mpt causing the errors or also mpt_sas?

This is anecdotal, but I would say that the LSI1068 cards and the
SASX28/SASX36 expander chips were far more prone to issues. I'm not
saying it can't or doesn't happen on LSI2008/SAS2X28-36 - I had
a prod box the other day that hard-hung because a Constellation went
south, had to have someone go yank the disk - but it sure
seems to happen a lot less with the newer kit and Phase10+
firmware. Got all sorts of mixes of SAS+SATA in large qty,
mostly pretty stable. 

Even having a single SASX28/36 expander in the mix seems to add
to the problems somewhat, for reasons that are not entirely clear. 

I have a couple around still. Hell, I have one box that is nothing
_but_ 1068s and old 846E1s - 4 of 'em to be exact, 86 old 1TB cuda.12
drives - which pretty much just works. Granted it doesn't work that
hard; it's archival material - but it runs and rarely errors. Still
Sol10U9 even.

*throws up hands*


zfs-discuss mailing list

Reply via email to