Pete Wyckoff wrote:
Another data point, if it is easy for you to setup: are things just
as slow using bmi_tcp rather than bmi_ib?
No, bmi_tcp works great (both over our regular ethernet interfaces and
our IBoIP interfaces). It's just bmi_ib that causes the slowness.
On Thu, May 18, 2006 at 05:10:34PM +0200, Phil Carns wrote:
> perf-mon.patch:
>
> karma1.patch:
> --
> karma2.patch:
> --
> karma3.patch:
> --
> karma4.patch:
> --
> karma5.patch:
> ---
Looks good. Comitted all of them (e
[EMAIL PROTECTED] wrote on Tue, 23 May 2006 13:21 -0500:
> >Another nice way to debug is to avoid the kernel module completely.
> >Just use the userspace tools and see if it also tends to be slow:
>
> I forget to mention in the previous email that I had already done this.
> Using pvfs-ls or pvfs
Pete Wyckoff wrote:
That doesn't sound familiar to me. I guess the first thing to do is
enable debugging and get a server log with timestamps so we can try
to figure it out. You may have to mail the log off-list as they
tend to be big.
Will do. I'll be sending you the log off-list either tod
Hi Avery,
Nice catch. Looks like the job_trove_bstream_read_list call sets
parameters and returns, so the result.bytes field is actually out of
scope later (as part of the one element array passed in). It doesn't
always segfault because the value remains unchanged, even though its
unal
[EMAIL PROTECTED] wrote on Tue, 23 May 2006 11:10 -0500:
> The pvfs2-migrate-collection binary worked perfectly. Thanks!
>
> Unfortunately, now that I can actually start up pvfs2-server, I've found
> out that my infiniband problem still exists. Any sort of file read
> operation causes pvfs2-se
On Thu, May 18, 2006 at 05:36:24PM +0200, Phil Carns wrote:
> kernel-lookup-err.patch:
>
> kernel-ro-2.4.patch:
>
> kernel-root-chmod.patch:
> -
> mount-pvfs2-dup-mnt.patch:
> --
These all look fine. Thanks for
Murali Vilayannur wrote:
As Rob pointed out, the pvfs2-migrate-collection utility in src/apps/admin
does migrate a 1.4.0 trove volume to the CVS trove format correctly. So I
guess you could just run that and let us know it you encounter any
problems...
Thanks,
Murali
The pvfs2-migrate-collectio
Hi Avery,
Yup.. Your fix is correct, but let's wait for Sam's comments..
Why exactly is result.bytes different from s_op->u.small_io.result_bytes?
I wonder how this did not get triggered in the normal scenarios (perhaps
it needed a sparse data file or something?)
Thanks,
Murali
On Mon, 22 May 200
Robert Latham wrote:
On Thu, May 18, 2006 at 05:10:34PM +0200, Phil Carns wrote:
Index: pvfs2_src/src/server/perf-mon.sm
===
--- pvfs2_src/src/server/perf-mon.sm(revision 1541)
+++ pvfs2_src/src/server/perf-mon.sm(revision
10 matches
Mail list logo