gluster-devel
[Top][All Lists]
Advanced

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

Re: [Gluster-devel] Hang in glfs_init_wait()


From: Anand Avati
Subject: Re: [Gluster-devel] Hang in glfs_init_wait()
Date: Wed, 18 Jul 2012 21:29:08 -0700



On Wed, Jul 18, 2012 at 9:21 PM, Bharata B Rao <address@hidden> wrote:
On Wed, Jul 18, 2012 at 10:35 PM, Anand Avati <address@hidden> wrote:
> Why do you want to re-init an already existing object?

As you note below, I don't want to re-init but rather do fini and init :)


My question came from a different angle. glfs_fini() will be destroying the entire object. You need to get hold of a new object with glfs_new() and do glfs_init() on the new object. I see no real reason in providing a glfs_fini() api which will result in partial destruction of glfs_t and also lets you to glfs_init() on it again. You might as well leave the glfs_t as-is and avoid an extra glfs_init() for the second access.

 
> Rather, why do you
> want to fini an object which you know you will init again? There is a cost
> to establishing and tearing down glfs_t. Why not initialize it once and use
> it for the lifetime of the process?

I agree, but the short answer is that for protocols like gluster where
I am not required to specify any image format,  QEMU ends up opening
the image twice which results in connection init and termination
twice.


As I said, why not just use a new glfs_t for the second access attempt?

Avati

reply via email to

[Prev in Thread] Current Thread [Next in Thread]