tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Raidframe low I/O for parity rebuild, but when accessed through rraid* devices



Le 13/06/2015 21:02, Jean-Yves Migeon a écrit :
Le 13/06/2015 19:30, Michael L. Hitch a écrit :
On Sat, 13 Jun 2015, Jean-Yves Migeon wrote:

I think that the slowness is I/O disk bound there: both disks were
used elsewhere first, so both have quite different block content.
Given that these disks are rather slow (5400 RPM ones), I suppose that
they dislike 64k read => compare => then 64k write on one, hence the
parity rebuild takes quite some time.

   I'd check if the write cache is disabled on the 2nd disk.  That
typically makes the writes much slower.  You can use dkctl or atactl to
check the current cache status and change it.

The cache (writeback) was effectively enabled, turning it off made a
slight improvement (less than 1MiB/s). Still better than nothing though.

And... the jmicron controller (on which the PATA is plugged) on the MB was configured in AHCI mode by a jumper, putting it in "compatible IDE" made the PATA disk a lot more happy, now it runs close to a whopping 25MiB. Hmm.

--
Jean-Yves Migeon


Home | Main Index | Thread Index | Old Index