archive mirror
 help / color / mirror / Atom feed
From: Johannes Thumshirn <>
To: Daniel Wagner <>
	Chaitanya Kulkarni <>,
Subject: Re: [PATCH] nvmet: Always remove processed AER elements from list
Date: Mon, 4 Nov 2019 10:50:34 +0100	[thread overview]
Message-ID: <20191104095034.GA3193@linux-lxv2> (raw)
In-Reply-To: <20191104081338.gdrk2a4mlov2io6s@beryllium.lan>

On Mon, Nov 04, 2019 at 09:13:38AM +0100, Daniel Wagner wrote:
> BTW, I got feedback how to produce the oops on our custerms setup:
> Test Steps:
> 1. Create two targets on node n1 as target side (tgt1 and tgt2)
> 2. Run perf to write data to tgt2:
>    sudo ./perf -q 1 -w read -o 4096 -t 60 -r 'trtype:RDMA adrfam:IPv4 traddr: trsvcid:4421'
> 3. Deleted tgt2 node n1 configuration during perf execution
>    and reboot the node n1 automatically.

I assume thhis "perf" is some kind of performance test. So it should be
possible to create a blktest recreating this issue.

I'd translate the "perf" line into something similar to 'fio -rw=read -bs 4k \
--time_based --timeout=60 --iodepth=1' and then running on a mpathed
nvme-loop and delete one of the controllers while the IO is running. This
should be a fairly usual multi-path test. The only thing I can't see yet
is, how we end up with unprocessed AENs here.


Linux-nvme mailing list

  reply	other threads:[~2019-11-04  9:50 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-03 20:13 [PATCH] nvmet: Always remove processed AER elements from list Chaitanya Kulkarni
2019-11-04  8:13 ` Daniel Wagner
2019-11-04  9:50   ` Johannes Thumshirn [this message]
2019-11-04 10:19     ` Daniel Wagner
2019-11-08 10:42       ` Daniel Wagner
2019-11-10  2:25         ` Chaitanya Kulkarni
2019-11-11  7:28           ` Daniel Wagner
2019-12-05  9:23           ` Daniel Wagner
2019-12-11  7:39             ` Chaitanya Kulkarni
2019-12-12  9:35               ` hch
2019-11-04 20:41   ` Chaitanya Kulkarni

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:

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191104095034.GA3193@linux-lxv2 \ \ \ \ \ \ \

* 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).