official way to load aoe module?

[Etzion Bar-Noy]

Inline

On Sun, Aug 22, 2010 at 4:41 PM, Amos Shapira <amos.shapira at gmail.com>wrote:

> On 22 August 2010 23:22, Oleg Goldshmidt <pub at goldshmidt.org> wrote:
> >
> > On Sun, Aug 22, 2010 at 2:25 PM, Amos Shapira <amos.shapira at gmail.com>
> wrote:
> >
> > > We are a little concerned about the situation of two guests mounting
> > > the ext3 and starting to manipulate the sqlite files on it in
> > > parallel.
> >
> > I think you should be *very* concerned about the situation where 2
> > guests mount an ext3 partition and start to manipulate files
> > *sequentially*. It looks like you *are* concerned (rightly), since you
> > wrote that only one client *mounts* the partition at a time.
>
> Yes. But apart from hoping that RHCS does its job right, there is
> nothing preventing other guests from mounting the same partition in
> parallel.
>
> Of course there is - LUN masking. Its purpose is exactly this. You expose
the LUN *only* to the nodes which require direct access to it.

> >
> > > Another option was to allow all guests to mount the file
> > > system read/write but carefully configure each guest to "own"
> > > different files or directories of sqlite files on the FS.
> >
> > What if one starts, e.g., creating files or appending content to
> > existing files (and allocating new blocks, etc., in the process)? The
> > other clients won't be aware of it.
>
> That's why we looked at cluster-aware file systems in form of GFS but
> decided the performance hit is too great to go with it. A brief look
> at OCFS installation steps gave an impression that it isn't trivial or
> well supported on CentOS 5.
>
Incorrect impression. Less than 5 minutes work, being extra slow, with
coffee included. Simple enough?

>
> >
> > I admit I have not thought long and hard about it, but it sounds
> > dangerous to me.
>
> It is. As was pointed out earlier in this thread - a large part of the
> "file system" is about how the file system module "caches" information
> in memory and synchronises it on the disk. If it's not a cluster-aware
> file system then parallel mounting is equivalent to opening the LV or
> device by an application and randomly starting to write data on it.
>
True. But cluster-aware FS should be considered carefully. For some
purposes, they ease management. For some others, they complicate it.
GFS has always been misunderstood by me. It has little benefits, and major
drawbacks. Can't see any reason to use it, from the existing variety of
clustered FS around.

Ez


>
> Cheers,
>
> --Amos
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.huji.ac.il/pipermail/linux-il/attachments/20100822/9dcaf4f6/attachment.html>