From: Lars Seipel <lars.seipel@h-da.de>
To: Marc Cote <marc.cote.mathieu@gmail.com>
Cc: ceph-devel@vger.kernel.org
Subject: Re: iSCSI gateway
Date: Mon, 4 Dec 2017 19:37:12 +0100 [thread overview]
Message-ID: <20171204183712.GA26726@horsthansen.slrz.net> (raw)
In-Reply-To: <CAMMWhM_uhhNM6rJvOpEQgCtegE+Tp8rMp00EW2z7s5egjjTbWw@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 1957 bytes --]
On Fri, Nov 24, 2017 at 11:29:12AM -0500, Marc Cote wrote:
> Good day Ladies and Gentlemen,
>
> We are embarking on the adventure of testing Ceph and iSCSI gateway
> with Windows clients.
>
> After many attempt, we keep running into issue after issue. Is
> there a white paper on a known tested setup with solid install
> procedure, including the list of repo and the order (if needed).
>
> Right now we have been working on Luminous, should we go with Jewel?
Nah, Luminous is the way to go here.
I started playing with the iSCSI gateway, too. It's certainly a quite
bumpy road, at least for now.
You should be able to get things running up to the point where you try
to add RBD volumes to the gateways. This will fail with a "Kernel not
supported" error message, caused by the running kernel not supporting
the qfull_time_out attribute.
While the Ceph docs state you need Linux 4.14+, the required patches
never ended up in 4.14, as the author discovered an issue with them.
Mike posted a new series to target-devel but they won't land in mainline
before 4.16.
I initially interpreted the "RHEL/CentOS 7.4; or Linux kernel v4.14 or
newer" part as saying that the EL7 kernel had those patches
applied/backported. Grepping through the big monolithic tarball
contained within kernel-3.10.0-693.5.2.el7.src.rpm, that doesn't seem to
be the case.
I might have missed something (just started playing with the iSCSI
stuff), but AFAICT the Ceph docs are simply incorrect on this matter.
So, if you want to test the iSCSI gateway, the way forward is probably
to pick up those TCMU kernel patches and start testing.
What are you running your gateways on? EL7?
--
Lars Seipel
IT-Services
lars.seipel@h-da.de
h_da
Hochschule Darmstadt
University of Applied Sciences
fbi
Fachbereich Informatik
Department of Computer Science
Schöfferstraße 8b, D10/0.37
D-64295 Darmstadt
☎ +49 6151 16-38423
[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 4777 bytes --]
prev parent reply other threads:[~2017-12-04 18:43 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-24 16:29 iSCSI gateway Marc Cote
2017-12-04 18:37 ` Lars Seipel [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171204183712.GA26726@horsthansen.slrz.net \
--to=lars.seipel@h-da.de \
--cc=ceph-devel@vger.kernel.org \
--cc=marc.cote.mathieu@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.