[linux-lvm] read is blocking if reading snapshots with 16k blocks on mp systems

[Martin Petermann]

> On Thu, Jan 10, 2002 at 07:58:37PM +0100, Martin Petermann wrote:
> > Hi all,
> >
> > I had some performance problems, than I wrote a small program to see
where
> > it was hanging:
> >
> > -my configuration:
> >
> > I'm using a SuSE 7.1 and SuSE 7.2 mp system (two processors).
> I've created a snapshot in the following way:
>
>
> > The maximum time read was hanging was 40 seconds. I detected similar
> > problems reading 3*4096 bytes.
>
> Can you tell us more about your system - I can't get this to over 3
seconds even
> on a dog slow IDE system with next-to no memory.

Hi Patrick,

this system has the following configuration:

- two pentium III processors (933 MHz), 256 KB cache
- 500 MB RAM
- the physical volumes reside on a hardware RAID (IBM ServRAID, RAID level
5)
- FC connected disks

I'll have to do some more investigations. I first thought multiprocessing
was the cause for this problem. But even in single processor mode(boot:
linux maxcpus=0 noapic) the problem exists. If people find out a problem on
such a configuration it's not easy to find out why.

Regards
   Martin