Storage Performance Development Kit (SPDK)
 help / color / mirror / Atom feed
From: Kiran Dikshit <kdikshit at cloudsimple.com>
To: spdk@lists.01.org
Subject: Re: [SPDK] Need help for iSCSI configuration for NVMe disc
Date: Wed, 31 May 2017 15:47:03 +0530	[thread overview]
Message-ID: <A1612022-B1F4-442D-9306-35879ACB3AC3@cloudsimple.com> (raw)
In-Reply-To: FF7FC980937D6342B9D289F5F3C7C2625B58CFD3@SHSMSX103.ccr.corp.intel.com

[-- Attachment #1: Type: text/plain, Size: 6221 bytes --]

ah got it,  thank you Liu that works ;-)



> On 31-May-2017, at 3:20 PM, Liu, Changpeng <changpeng.liu(a)intel.com> wrote:
> 
> Hi Kiran,
>  
> It’s:
> TransportID "trtype:PCIe traddr:0000:0a:00.0" Nvme3
>  
> Seems that you wrote with TransportID “trtype:PCIe traddr:0000.0a:00.0” Nvme3, very small differences, J
>   <>
> From: Kiran Dikshit [mailto:kdikshit(a)cloudsimple.com] 
> Sent: Wednesday, May 31, 2017 5:40 PM
> To: Liu, Changpeng <changpeng.liu(a)intel.com>
> Cc: Storage Performance Development Kit <spdk(a)lists.01.org>
> Subject: Re: [SPDK] Need help for iSCSI configuration for NVMe disc
>  
> Hi Liu,
>  
> I did make the changes but still see the error.
> Attached is the modified config file and the error log for the same.
>  
>  
>  
> -Kiran
>  
>  
> On 31-May-2017, at 8:32 AM, Liu, Changpeng <changpeng.liu(a)intel.com <mailto:changpeng.liu(a)intel.com>> wrote:
>  
> Hi Kiran,
>  
> Can you try set TransportID “trtype:PCIe traddr:0000.0a:00.0” Nvme3  in your configuration file, and also
> change your “LUN0 Nvme3” with “LUN0 Nvme3n1” ?
>  
> Nvme3 is not a block device name
>  
>   <>
> From: SPDK [mailto:spdk-bounces(a)lists.01.org <mailto:spdk-bounces(a)lists.01.org>] On Behalf Of Harris, James R
> Sent: Friday, May 26, 2017 11:42 PM
> To: Kiran Dikshit <kdikshit(a)cloudsimple.com <mailto:kdikshit(a)cloudsimple.com>>
> Cc: Storage Performance Development Kit <spdk(a)lists.01.org <mailto:spdk(a)lists.01.org>>
> Subject: Re: [SPDK] Need help for iSCSI configuration for NVMe disc
>  
> Hi Kiran,
>  
> Could you attach the following:
>  
> 1)      A log of “lspci –vvv”
> 2)      Your full SPDK iscsi.conf file
> 3)      A full log of the iscsi_tgt output.  Please enable debug (./configure --enable-debug) and add “-t all” to your command line to enable some extra debug logging
>  
> Thanks,
>  
> -Jim
>  
>  
> From: Kiran Dikshit <kdikshit(a)cloudsimple.com <mailto:kdikshit(a)cloudsimple.com>>
> Date: Thursday, May 25, 2017 at 7:07 PM
> To: James Harris <james.r.harris(a)intel.com <mailto:james.r.harris(a)intel.com>>
> Cc: Storage Performance Development Kit <spdk(a)lists.01.org <mailto:spdk(a)lists.01.org>>
> Subject: Re: [SPDK] Need help for iSCSI configuration for NVMe disc
>  
> Hi Harris, 
>  
> After the previous mail, i did check out the masters and DPDK version 17.05 and found the same issue. 
>  
> SPDK git commit ID : 5597e835c8981d6500d00c0f88fd10edae173621
>  
> Thank you
> Kiran
>  
> On 26-May-2017, at 12:38 AM, Harris, James R <james.r.harris(a)intel.com <mailto:james.r.harris(a)intel.com>> wrote:
>  
> Hi Kiran,
>  
> Can you confirm which SPDK git commit ID you are working from?  Since there have been changes in this transport ID format, I want to make sure I provide you the correct details based on the SPDK version you are using.
>  
> Thanks,
>  
> -Jim
>  
>  
> From: Kiran Dikshit <kdikshit(a)cloudsimple.com <mailto:kdikshit(a)cloudsimple.com>>
> Date: Wednesday, May 24, 2017 at 10:50 PM
> To: James Harris <james.r.harris(a)intel.com <mailto:james.r.harris(a)intel.com>>
> Cc: Storage Performance Development Kit <spdk(a)lists.01.org <mailto:spdk(a)lists.01.org>>
> Subject: Re: [SPDK] Need help for iSCSI configuration for NVMe disc
>  
> Hi Harris, 
>  
> I did try with the TransportID “trtype:PCIe traddr:0000.0a:00.0” Nvme3  option i am still seeing the same issue. 
>  
> Thank you
> Kiran
>  
> On 24-May-2017, at 9:50 PM, Harris, James R <james.r.harris(a)intel.com <mailto:james.r.harris(a)intel.com>> wrote:
>  
> 
> On May 24, 2017, at 4:10 AM, Kiran Dikshit <kdikshit(a)cloudsimple.com <mailto:kdikshit(a)cloudsimple.com>> wrote:
> 
> Hi All,
> 
> I am trying to configure iSCSI for a NVMe drive and getting the below error’s. I have unbound the NVMe drive from kernel using the setup.sh.
> 
> tgt_node.c: 718:spdk_iscsi_tgt_node_construct: ***ERROR*** Could not construct SCSI device
> tgt_node.c: 984:spdk_cf_add_iscsi_tgt_node: ***ERROR*** tgt_node1: add_iscsi_target_node error
> tgt_node.c:1010:spdk_iscsi_init_tgt_nodes: ***ERROR*** spdk_cf_add_iscsi_tgt_node() failed
> iscsi_subsystem.c: 965:spdk_iscsi_subsystem_init: ***ERROR*** spdk_iscsi_init_tgt_nodes() failed
> app.c: 404:spdk_app_init: ***ERROR*** spdk_subsystem_init() failed
> 
> 
> 
> Below is the section of my iscsi.conf.in <http://iscsi.conf.in/> file for reference.
> 
> # NVMe configuration options
> [Nvme]
> # NVMe Device Whitelist
> # Users may specify which NVMe devices to claim by their PCI
> # domain, bus, device, and function. The format is dddd:bb:dd.f, which is
> # the same format displayed by lspci or in /sys/bus/pci/devices. The second
> # argument is a "name" for the device that can be anything. The name
> # is referenced later in the Subsystem section.
> #
> # Alternatively, the user can specify ClaimAllDevices. All
> # NVMe devices will be claimed and named Nvme0, Nvme1, etc.
> # BDF 0000:00:00.0 Nvme0
>  BDF 0000:0a:00.0 Nvme3
> 
> Hi Kiran,
> 
> The format of the NVMe section changed in the last SPDK release, so that we could add support for remote NVMe-oF devices as well.
> 
> Can you try the following instead?
> 
> [Nvme]
>  TransportID “trtype:PCIe traddr:0000.0a:00.0” Nvme3
> 
> The comment for spdk_nvme_transport_id_parse() in include/spdk/nvme.h describes the format in more detail, including how to specify NVMe-oFtargets.
> 
> Thanks,
> 
> -Jim
> 
> 
> 
> [TargetNode1]
>  TargetName disk2
>  TargetAlias "Data Disk2"
>  Mapping PortalGroup1 InitiatorGroup1
>  AuthMethod Auto
>  AuthGroup AuthGroup1
>  UseDigest Auto
>  LUN0 Nvme3
>  QueueDepth 64
> 
> 
> Thank you
> Kiran
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org <mailto:SPDK(a)lists.01.org>
> https://lists.01.org/mailman/listinfo/spdk <https://lists.01.org/mailman/listinfo/spdk>
> 
> _______________________________________________
> SPDK mailing list
> SPDK(a)lists.01.org <mailto:SPDK(a)lists.01.org>
> https://lists.01.org/mailman/listinfo/spdk <https://lists.01.org/mailman/listinfo/spdk>

[-- Attachment #2: attachment.html --]
[-- Type: text/html, Size: 50833 bytes --]

             reply	other threads:[~2017-05-31 10:17 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-31 10:17 Kiran Dikshit [this message]
  -- strict thread matches above, loose matches on Subject: below --
2017-05-31  9:50 [SPDK] Need help for iSCSI configuration for NVMe disc Liu, Changpeng
2017-05-31  9:40 Kiran Dikshit
2017-05-31  3:02 Liu, Changpeng
2017-05-29  4:44 Kiran Dikshit
2017-05-29  4:42 Kiran Dikshit
2017-05-26 15:41 Harris, James R
2017-05-26  2:07 Kiran Dikshit
2017-05-25 19:08 Harris, James R
2017-05-25  5:50 Kiran Dikshit
2017-05-24 16:20 Harris, James R
2017-05-24 11:10 Kiran Dikshit

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=A1612022-B1F4-442D-9306-35879ACB3AC3@cloudsimple.com \
    --to=spdk@lists.01.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox