Hi,
We are testing the multipath behavior on centos7.3 and linux version Linux
EC056108 3.10.
We have two paths connected to dm-1(sda and sdb).
We are using fio tool to read data from /dev/dm-1 and we are seeing data on
both paths(sda and sdb)
Below is the fio command which we are using
./fio
On 07/19/2017 08:56 PM, Mike Snitzer wrote:
> On Wed, Jul 19 2017 at 12:22pm -0400,
> Milan Broz wrote:
>
>> On 07/19/2017 05:39 PM, Mike Snitzer wrote:
>>> On Wed, Jul 19 2017 at 11:30P -0400,
>>> Mikulas Patocka wrote:
>>>
Hi Mike
Some
On Wed, 19 Jul 2017, Mike Snitzer wrote:
> On Wed, Jul 19 2017 at 2:39pm -0400,
> John Stoffel wrote:
>
> > > "Mikulas" == Mikulas Patocka writes:
> >
> > Mikulas> When using block size greater than 512 bytes, the dm-integrity
> > target
> >
On Wed, 19 Jul 2017, John Stoffel wrote:
> I'd like to argue that you should never use BUG_ON at all, esp since
> if you have integrity running on just one critical device, but have
> other devices that work just fine, bringing down the entire system
> because you don't think things are ok is
On Wed, 19 Jul 2017, John Stoffel wrote:
> > "Mikulas" == Mikulas Patocka writes:
>
> Mikulas> When using block size greater than 512 bytes, the dm-integrity target
> Mikulas> allocates journal space inefficiently, it allocates one entry for
> each
> Mikulas>
On Thu, 20 Jul 2017, Mikulas Patocka wrote:
>
>
> On Wed, 19 Jul 2017, John Stoffel wrote:
>
> > I'd like to argue that you should never use BUG_ON at all, esp since
> > if you have integrity running on just one critical device, but have
> > other devices that work just fine, bringing down
Dear Yang,
On Thu, 2017-07-20 at 11:36 +0800, Yang Feng wrote:
> Add args min_avg_latency of logarithmic scale, for
> prioritizers/path_latency.c.
> Min average latency is not constant 1us, and is set by user.
> Certainly, max average
> latency value is still 100s. It make support better for
Dear Yang,
On Thu, 2017-07-20 at 11:36 +0800, Yang Feng wrote:
> 1. The SCSI-to-NVMe translations have been removed in the patch
> "nvme:
> Remove SCSI translations" in the linux-nvme, so the native NVMe Ioctl
> command should be supported in the multipath-tools.
> 2. In the
> "Mikulas" == Mikulas Patocka writes:
Mikulas> On Wed, 19 Jul 2017, John Stoffel wrote:
>> I'd like to argue that you should never use BUG_ON at all, esp since
>> if you have integrity running on just one critical device, but have
>> other devices that work just fine,
> "Mikulas" == Mikulas Patocka writes:
Mikulas> On Wed, 19 Jul 2017, John Stoffel wrote:
>> > "Mikulas" == Mikulas Patocka writes:
>>
Mikulas> When using block size greater than 512 bytes, the dm-integrity target
Mikulas> allocates journal
Dear Martin,
Thank you very much for your guidance and reviews.
Please find my replys as follows.
The up-to-date patch will be sent later.
Regards,
-Yang
On 2017/7/21 2:07, Martin Wilck wrote:
> Dear Yang,
>
> On Thu, 2017-07-20 at 11:36 +0800, Yang Feng wrote:
>> Add args min_avg_latency of
Dear Martin,
Thanks a lot for your reviews.
Please find my replys as follows.
And the up-to-date patch will be sent later.
Regards,
-Yang
On 2017/7/21 1:50, Martin Wilck wrote:
> Dear Yang,
>
> On Thu, 2017-07-20 at 11:36 +0800, Yang Feng wrote:
>> 1. The SCSI-to-NVMe translations have been
This three patches support coalescing heterogenous paths by referencing
another path identifier. This is useful in the scenario of migrating data
for heterogenous arrays without interrupting uplayer transaction.
The 1/3 patch and 2/3 patch is used to prepare for the 3/3 patch which
covers the
The helper get_next_string is useful and generic. So move from
exclusive weightedpath module to util module. It will be used
in the next second patch.
Signed-off-by: Junxiong Guan
---
libmultipath/prioritizers/weightedpath.c | 10 +-
libmultipath/util.c
This patch supports migrating data for heterogenous arrays without
interrupting uplayer transaction.
The usage is as follows:
The new attribute named uid_reference provides a unique path identifier
for some paths by referencing another path's identifier. Therefor those
paths can be coalesced.
The flags which will be used in the next patch indicates the internel
state of the path such as whether the wwid of the path is referenced
by another path. With this flag, we can avoid overflow of getting wwid
of a certain path by recurring method in the next patch.
Signed-off-by: Junxiong Guan
16 matches
Mail list logo