Re: Re: Linux fs.aio-max-nr Leak?
Date: Tue, 30 Sep 2014 10:10:21 +0200
Message-ID: <148c58f9b87.11d9c8c1d111428.5701137068925160098_at_zoho.com>
I don't reject the chance of an Oracle bug, but performance degradation in flash/ssd storage is a fairly common issue in my experience and google is plenty of information about it.
Furthermore, since the OP wrote that fs.aio.max needed to increase (initial 1M, then 5M and now 10M), between an Oracle/Linux bug and a performance degradation I'm biased the latter.
- On Mon, 29 Sep 2014 21:30:10 +0200 Matthew Zito wrote ----
>Whoops, I accidentally sent this response just to Alessandro:
>
>"Why would you say that the problem is almost certainly due to flash storage? It seems a lot more likely to me that there's a bug in either Oracle or Linux where requests get retransmitted and the originals hang around."
>
>On Mon, Sep 29, 2014 at 12:09 PM, Alessandro Vercelli <alever22_at_zoho.com> wrote:
>Maybe the debate abount pros and cons of flash/SSD storage vs magnetic hard drives has been taken thousands of times without a winner, but I personally would use a non-magnetic disk only for os filesystems (and being compelled to use flash storage), first of all because of performance degradation.
> In the specific situation, I'm almost certain the problem is due to flash storage.
>
>
>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Tue Sep 30 2014 - 10:10:21 CEST