From chiaranjun@gmail.com Sat Aug  8 08:41:42 2020
From: chiaranjun@gmail.com
To: ceph-users@ceph.io
Subject: [ceph-users] i would love more this content
Date: Sat, 08 Aug 2020 08:41:35 +0000
Message-ID: <159687609546.20.3827592004700396812@mailman-web>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6615661460801056088=="

--===============6615661460801056088==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

Nice to be visiting your blog again, it has been months for me. Well this art=
icle that i've been waited for so long. I need this article to complete my as=
signment in the college, and it has same topic with your article. Thanks, gre=
at share.Here are some ineteresting content i have to share as well;
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-for-sale-nea=
r-me/"rel=3D"dofollow">pomeranian for sale near me</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-chihuahua-mi=
x/"rel=3D"dofollow">pomeranian chihuahua mix</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-for-sale/"re=
l=3D"dofollow">pomeranian for sale</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-on-sale/"rel=
=3D"dofollow">pomeranian on sale</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-black-and-wh=
ite/"rel=3D"dofollow">pomeranian black and white</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-miniature/"r=
el=3D"dofollow">pomeranian miniature</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-black/"rel=
=3D"dofollow">pomeranian black</a>
<a href=3D"https://pomeranianhaustierhaus.com/product/pomeranian-adoption/"re=
l=3D"dofollow">pomeranian adoption</a>
<a href=3D"https://pomeranianhaustierhaus.com/pomeranian-puppy/"rel=3D"dofoll=
ow">pomeranian puppy</a>

--===============6615661460801056088==--


From ekalk@socket.net Sat Aug  8 10:49:02 2020
From: Edward kalk <ekalk@socket.net>
To: ceph-users@ceph.io
Subject: [ceph-users]  SED drives , poor performance 
Date: Sat, 08 Aug 2020 05:48:54 -0500
Message-ID: <A97D5B2D-E023-4C52-864B-4D0FB80D3E40@socket.net>
In-Reply-To: <159687609546.20.3827592004700396812@mailman-web>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============4134266355696433827=="

--===============4134266355696433827==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

Im getting poor performance with 5 of my OSDs, Seagate Constellation ES SED (=
1) 10k  SAS 2TB 3.5 drives.
disk write latency keeps drifting high ,
100ms-230ms on writes. the other 30 OSDs are performing well. avg latency 10-=
20ms

We observe stats via =E2=80=9Ciostat -xtc 2=E2=80=9D on CEPH server. w_await =
showing writes to disk latency.=20

**anyone else have poor latency on SEDs?

The disks were included on a server we acquired. Didn=E2=80=99t know they wou=
ld be SEDs.
I=E2=80=99ve read that seagate SED vs. non SED shouldn=E2=80=99t perform diff=
erently, but looks like they are awful.  All the SEDs are in the newly added =
CEPH node. A Dell R510 with H700 raid card. all disks in entire cluster are R=
0, direct write to disk, no cache.
The new server and disk control spec out at about 10x the ability and IOPs of=
 other servers, so I suspect the SEDs.

-Ed

--===============4134266355696433827==--


From M.Roos@f1-outsourcing.eu Sat Aug  8 10:56:13 2020
From: Marc Roos <M.Roos@f1-outsourcing.eu>
To: ceph-users@ceph.io
Subject: [ceph-users] Re: SED drives , poor performance
Date: Sat, 08 Aug 2020 12:56:01 +0200
Message-ID: <"H0000071001780a2.1596884161.sx.f1-outsourcing.eu*"@MHS>
In-Reply-To: <A97D5B2D-E023-4C52-864B-4D0FB80D3E40@socket.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============6652808000650223965=="

--===============6652808000650223965==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

 
Maybe to obvious suggestion, but what about disabling SED on one of 
these drives?


-----Original Message-----
Cc: ceph-users@ceph.io
Subject: [ceph-users] SED drives , poor performance

Im getting poor performance with 5 of my OSDs, Seagate Constellation ES 
SED (1) 10k  SAS 2TB 3.5 drives.
disk write latency keeps drifting high , 100ms-230ms on writes. the 
other 30 OSDs are performing well. avg latency 10-20ms

We observe stats via “iostat -xtc 2” on CEPH server. w_await showing 
writes to disk latency. 

**anyone else have poor latency on SEDs?

The disks were included on a server we acquired. Didn’t know they would 
be SEDs.
I’ve read that seagate SED vs. non SED shouldn’t perform differently, 
but looks like they are awful.  All the SEDs are in the newly added CEPH 
node. A Dell R510 with H700 raid card. all disks in entire cluster are 
R0, direct write to disk, no cache.
The new server and disk control spec out at about 10x the ability and 
IOPs of other servers, so I suspect the SEDs.

-Ed
_______________________________________________
ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an 
email to ceph-users-leave@ceph.io


--===============6652808000650223965==--


From ekalk@socket.net Sat Aug  8 11:11:36 2020
From: Edward kalk <ekalk@socket.net>
To: ceph-users@ceph.io
Subject: [ceph-users] Re: SED drives , poor performance 
Date: Sat, 08 Aug 2020 06:11:28 -0500
Message-ID: <BE30405F-560C-40EA-89EC-E264BF25E33E@socket.net>
In-Reply-To: <"H0000071001780a2.1596884161.sx.f1-outsourcing.eu*"@MHS>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============7640125266624960000=="

--===============7640125266624960000==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

been reading up on the Seagate Constellation ES SED, don=E2=80=99t see anythi=
ng saying that can be done. I plan to swap one with a spare non-SED I have ne=
xt week to see if perf goes normal.

-Ed

> On Aug 8, 2020, at 5:56 AM, Marc Roos <M.Roos@f1-outsourcing.eu> wrote:
>=20
> =EF=BB=BF
> Maybe to obvious suggestion, but what about disabling SED on one of=20
> these drives?
>=20
>=20
> -----Original Message-----
> Cc: ceph-users@ceph.io
> Subject: [ceph-users] SED drives , poor performance
>=20
> Im getting poor performance with 5 of my OSDs, Seagate Constellation ES=20
> SED (1) 10k  SAS 2TB 3.5 drives.
> disk write latency keeps drifting high , 100ms-230ms on writes. the=20
> other 30 OSDs are performing well. avg latency 10-20ms
>=20
> We observe stats via =E2=80=9Ciostat -xtc 2=E2=80=9D on CEPH server. w_awai=
t showing=20
> writes to disk latency.=20
>=20
> **anyone else have poor latency on SEDs?
>=20
> The disks were included on a server we acquired. Didn=E2=80=99t know they w=
ould=20
> be SEDs.
> I=E2=80=99ve read that seagate SED vs. non SED shouldn=E2=80=99t perform di=
fferently,=20
> but looks like they are awful.  All the SEDs are in the newly added CEPH=20
> node. A Dell R510 with H700 raid card. all disks in entire cluster are=20
> R0, direct write to disk, no cache.
> The new server and disk control spec out at about 10x the ability and=20
> IOPs of other servers, so I suspect the SEDs.
>=20
> -Ed
> _______________________________________________
> ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an=20
> email to ceph-users-leave@ceph.io
>=20
>=20
>=20

--===============7640125266624960000==--


From darren.soothill@suse.com Sun Aug  9 06:51:29 2020
From: Darren Soothill <darren.soothill@suse.com>
To: ceph-users@ceph.io
Subject: [ceph-users] Re: SED drives , poor performance
Date: Sun, 09 Aug 2020 06:51:18 +0000
Message-ID: <127A1F0C-E203-4C75-8991-98F72BA7FE58@suse.com>
In-Reply-To: <BE30405F-560C-40EA-89EC-E264BF25E33E@socket.net>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1257834155312203159=="

--===============1257834155312203159==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

This is where an understanding of how an encrypting drive works needs to be t=
aken into account.

An SED always encrypts all data written to it it uses a long encryption key t=
hat is built into the drive.

When you =E2=80=9Cturn on encryption=E2=80=9D what you are doing is setting a=
 pin to allow access to the encryption key to read your data back.

I would be looking at some of the heparin parameters to see if there are any =
differences in these parameters. Things like drive caching being on for one d=
rive and off for another.

Having tested many different drives over the years I have never seen a perfor=
mance difference between SED and non SED when all parameters on the drive are=
 the same.

Sent from my iPhone

> On 8 Aug 2020, at 12:12, Edward kalk <ekalk@socket.net> wrote:
>=20
> =EF=BB=BFbeen reading up on the Seagate Constellation ES SED, don=E2=80=99t=
 see anything saying that can be done. I plan to swap one with a spare non-SE=
D I have next week to see if perf goes normal.
>=20
> -Ed
>=20
>> On Aug 8, 2020, at 5:56 AM, Marc Roos <M.Roos@f1-outsourcing.eu> wrote:
>>=20
>> =EF=BB=BF
>> Maybe to obvious suggestion, but what about disabling SED on one of=20
>> these drives?
>>=20
>>=20
>> -----Original Message-----
>> Cc: ceph-users@ceph.io
>> Subject: [ceph-users] SED drives , poor performance
>>=20
>> Im getting poor performance with 5 of my OSDs, Seagate Constellation ES=20
>> SED (1) 10k  SAS 2TB 3.5 drives.
>> disk write latency keeps drifting high , 100ms-230ms on writes. the=20
>> other 30 OSDs are performing well. avg latency 10-20ms
>>=20
>> We observe stats via =E2=80=9Ciostat -xtc 2=E2=80=9D on CEPH server. w_awa=
it showing=20
>> writes to disk latency.=20
>>=20
>> **anyone else have poor latency on SEDs?
>>=20
>> The disks were included on a server we acquired. Didn=E2=80=99t know they =
would=20
>> be SEDs.
>> I=E2=80=99ve read that seagate SED vs. non SED shouldn=E2=80=99t perform d=
ifferently,=20
>> but looks like they are awful.  All the SEDs are in the newly added CEPH=20
>> node. A Dell R510 with H700 raid card. all disks in entire cluster are=20
>> R0, direct write to disk, no cache.
>> The new server and disk control spec out at about 10x the ability and=20
>> IOPs of other servers, so I suspect the SEDs.
>>=20
>> -Ed
>> _______________________________________________
>> ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an=20
>> email to ceph-users-leave@ceph.io
>>=20
>>=20
>>=20
> _______________________________________________
> ceph-users mailing list -- ceph-users@ceph.io
> To unsubscribe send an email to ceph-users-leave@ceph.io

--===============1257834155312203159==--


From M.Roos@f1-outsourcing.eu Sun Aug  9 09:14:41 2020
From: Marc Roos <M.Roos@f1-outsourcing.eu>
To: ceph-users@ceph.io
Subject: [ceph-users] Re: SED drives , poor performance
Date: Sun, 09 Aug 2020 11:14:27 +0200
Message-ID: <"H00000710017817e.1596964467.sx.f1-outsourcing.eu*"@MHS>
In-Reply-To: <127A1F0C-E203-4C75-8991-98F72BA7FE58@suse.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1152011715744382150=="

--===============1152011715744382150==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit

 


> When you “turn on encryption” what you are doing is setting a pin
> to allow access to the encryption key to read your data back.

Ooh interesting, I guess I have to read about this, to understand why 
one even needs this. It is not like one would stop using luks or so.



--===============1152011715744382150==--