linux-nvme.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: swise@opengridcomputing.com (Steve Wise)
Subject: target crash / host hang with nvme-all.3 branch of nvme-fabrics
Date: Thu, 16 Jun 2016 15:01:35 -0500	[thread overview]
Message-ID: <01b701d1c809$e2b4b6d0$a81e2470$@opengridcomputing.com> (raw)
In-Reply-To: <5763044A.9090206@grimberg.me>

> Umm, I think this might be happening because we get to delete_ctrl when
> one of our queues has a NULL ctrl. This means that either:
> 1. we never got a chance to initialize it, or
> 2. we already freed it.
> 
> (1) doesn't seem possible as we have a very short window (that we're
> better off eliminating) between when we start the keep-alive timer (in
> alloc_ctrl) and the time we assign the sq->ctrl (install_queue).
> 
> (2) doesn't seem likely either to me at least as from what I followed,
> delete_ctrl should be mutual exclusive with other deletions, moreover,
> I didn't see an indication in the logs that any other deletions are
> happening.
> 
> Steve, is this something that started happening recently? does the
> 4.6-rc3 tag suffer from the same phenomenon?

In case it matters, here is my target config file.  I'm only connecting to 1
device (test-hynix0).  But each device is exported over 2 ports of a Chelsio
iWARP T580, and 1 port of a MLX IB mlx4 QDR adapter.

[root at stevo2 ~]# cat /etc/nvmet.json
{
  "hosts": [],
  "ports": [
    {
      "addr": {
        "adrfam": "ipv4",
        "traddr": "10.0.1.14",
        "treq": "not specified",
        "trsvcid": "4420",
        "trtype": "rdma"
      },
      "portid": 1,
      "referrals": [],
      "subsystems": [
        "test-ram0",
        "test-ram1",
        "test-ram2",
        "test-ram3",
        "test-hynix0",
        "test-nullb0",
        "test-nullb1"
      ]
    },
    {
      "addr": {
        "adrfam": "ipv4",
        "traddr": "10.0.2.14",
        "treq": "not specified",
        "trsvcid": "4420",
        "trtype": "rdma"
      },
      "portid": 2,
      "referrals": [],
      "subsystems": [
        "test-ram0",
        "test-ram1",
        "test-ram2",
        "test-ram3",
        "test-hynix0",
        "test-nullb0",
        "test-nullb1"
      ]
    },
    {
      "addr": {
        "adrfam": "ipv4",
        "traddr": "10.0.7.14",
        "treq": "not specified",
        "trsvcid": "4420",
        "trtype": "rdma"
      },
      "portid": 7,
      "referrals": [],
      "subsystems": [
        "test-ram0",
        "test-ram1",
        "test-ram2",
        "test-ram3",
        "test-hynix0",
        "test-nullb0",
        "test-nullb1"
      ]
    }
  ],
  "subsystems": [
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "00000000-0000-0000-0000-123400000000",
            "path": "/dev/ram0"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-ram0"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "00000000-0000-0000-0000-567800000000",
            "path": "/dev/ram1"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-ram1"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "00000000-0000-0000-0000-9abc00000000",
            "path": "/dev/ram2"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-ram2"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "00000000-0000-0000-0000-def100000000",
            "path": "/dev/ram3"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-ram3"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "5aa133f4-7c41-f141-b469-901e58461e7b",
            "path": "/dev/nvme0n1"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-hynix0"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "569232db-0f5a-8d4f-a077-7935117e2f3f",
            "path": "/dev/nullb0"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-nullb0"
    },
    {
      "allowed_hosts": [],
      "attr": {
        "allow_any_host": "1"
      },
      "namespaces": [
        {
          "device": {
            "nguid": "9eb7de4b-3ee9-4641-a6e3-e2569414e5a0",
            "path": "/dev/nullb1"
          },
          "enable": 1,
          "nsid": 1
        }
      ],
      "nqn": "test-nullb1"
    }
  ]
}

  parent reply	other threads:[~2016-06-16 20:01 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-16 14:53 target crash / host hang with nvme-all.3 branch of nvme-fabrics Steve Wise
2016-06-16 14:57 ` Christoph Hellwig
2016-06-16 15:10   ` Christoph Hellwig
2016-06-16 15:17     ` Steve Wise
2016-06-16 19:11     ` Sagi Grimberg
2016-06-16 20:38       ` Christoph Hellwig
2016-06-16 21:37         ` Sagi Grimberg
2016-06-16 21:40           ` Sagi Grimberg
2016-06-21 16:01           ` Christoph Hellwig
2016-06-22 10:22             ` Sagi Grimberg
2016-06-16 15:24   ` Steve Wise
2016-06-16 16:41     ` Steve Wise
2016-06-16 15:56   ` Steve Wise
2016-06-16 19:55     ` Sagi Grimberg
2016-06-16 19:59       ` Steve Wise
2016-06-16 20:07         ` Sagi Grimberg
2016-06-16 20:12           ` Steve Wise
2016-06-16 20:27             ` Ming Lin
2016-06-16 20:28               ` Steve Wise
2016-06-16 20:34                 ` 'Christoph Hellwig'
2016-06-16 20:49                   ` Steve Wise
2016-06-16 21:06                     ` Steve Wise
2016-06-16 21:42                       ` Sagi Grimberg
2016-06-16 21:47                         ` Ming Lin
2016-06-16 21:53                           ` Steve Wise
2016-06-16 21:46                       ` Steve Wise
2016-06-27 22:29                       ` Ming Lin
2016-06-28  9:14                         ` 'Christoph Hellwig'
2016-06-28 14:15                           ` Steve Wise
2016-06-28 15:51                             ` 'Christoph Hellwig'
2016-06-28 16:31                               ` Steve Wise
2016-06-28 16:49                                 ` Ming Lin
2016-06-28 19:20                                   ` Steve Wise
2016-06-28 19:43                                     ` Steve Wise
2016-06-28 21:04                                       ` Ming Lin
2016-06-29 14:11                                         ` Steve Wise
2016-06-27 17:26                   ` Ming Lin
2016-06-16 20:35           ` Steve Wise
2016-06-16 20:01       ` Steve Wise [this message]
2016-06-17 14:05       ` Steve Wise
     [not found]       ` <005f01d1c8a1$5a229240$0e67b6c0$@opengridcomputing.com>
2016-06-17 14:16         ` Steve Wise
2016-06-17 17:20           ` Ming Lin
2016-06-19 11:57             ` Sagi Grimberg
2016-06-21 14:18               ` Steve Wise
2016-06-21 17:33                 ` Ming Lin
2016-06-21 17:59                   ` Steve Wise
     [not found]               ` <006e01d1cbc7$d0d9cc40$728d64c0$@opengridcomputing.com>
2016-06-22 13:42                 ` Steve Wise
2016-06-27 14:19                   ` Steve Wise
2016-06-28  8:50                     ` 'Christoph Hellwig'
2016-07-04  9:57                       ` Yoichi Hayakawa

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='01b701d1c809$e2b4b6d0$a81e2470$@opengridcomputing.com' \
    --to=swise@opengridcomputing.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).