From: Sagi Grimberg <sagi@grimberg.me>
To: Victor Gladkov <Victor.Gladkov@kioxia.com>,
Hannes Reinecke <hare@suse.de>,
James Smart <james.smart@broadcom.com>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>
Subject: Re: [PATCH] nvme-fabrics: reject I/O to offline device
Date: Tue, 17 Dec 2019 13:46:46 -0800 [thread overview]
Message-ID: <73006c25-b6a8-fc36-0789-772e3ea59a02@grimberg.me> (raw)
In-Reply-To: <d7953accf06e418a893b9cc6017b981a@kioxia.com>
On 12/9/19 7:30 AM, Victor Gladkov wrote:
> On 12/8/19 14:18 PM, Hannes Reinecke wrote:
>>
>> On 12/6/19 11:18 PM, Sagi Grimberg wrote:
>>>
>>>>> ---
>>>>> diff --git a/drivers/nvme/host/fabrics.c
>>>>> b/drivers/nvme/host/fabrics.c index 74b8818..b58abc1 100644
>>>>> --- a/drivers/nvme/host/fabrics.c
>>>>> +++ b/drivers/nvme/host/fabrics.c
>>>>> @@ -549,6 +549,8 @@ blk_status_t nvmf_fail_nonready_command(struct
>>>>> nvme_ctrl *ctrl,
>>>>> {
>>>>> if (ctrl->state != NVME_CTRL_DELETING &&
>>>>> ctrl->state != NVME_CTRL_DEAD &&
>>>>> + !(ctrl->state == NVME_CTRL_CONNECTING &&
>>>>> + ((ktime_get_ns() - rq->start_time_ns) >
>>>>> jiffies_to_nsecs(rq->timeout))) &&
>>>>> !blk_noretry_request(rq) && !(rq->cmd_flags &
>>>>> REQ_NVME_MPATH))
>>>>> return BLK_STS_RESOURCE;
>>>>>
>>>>
>>>> Did you test this to ensure it's doing what you expect. I'm not sure
>>>> that all the timers are set right at this point. Most I/O's timeout
>>>> from a deadline time stamped at blk_mq_start_request(). But that
>>>> routine is actually called by the transports post the
>>>> nvmf_check_ready/fail_nonready calls. E.g. the io is not yet in
>>>> flight, thus queued, and the blk-mq internal queuing doesn't count
>>>> against the io timeout. I can't see anything that guarantees
>>>> start_time_ns is set.
>>>
>>> I'm not sure this behavior for failing I/O always desired? some
>>> consumers would actually not want the I/O to fail prematurely if we
>>> are not multipathing...
>>>
>>> I think we need a fail_fast_tmo set in when establishing the
>>> controller to get it right.
>>>
>> Agreed. This whole patch looks like someone is trying to reimplement
>> fast_io_fail_tmo / dev_loss_tmo.
>> As we're moving into unreliable fabrics I guess we'll need a similar mechanism.
>>
>> Cheers,
>>
>> Hannes
>
>
> Following your suggestions, I added a new session parameter called "fast_fail_tmo".
> The timeout is measured in seconds from the controller reconnect, any command beyond that timeout is rejected.
> The new parameter value may be passed during ‘connect’, and its default value is 30 seconds.
The default should be consistent with the existing behavior.
> A value of -1 means no timeout (in similar to current behavior).
>
> ---
> diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c
> index 74b8818..ed6b911 100644
> --- a/drivers/nvme/host/fabrics.c
> +++ b/drivers/nvme/host/fabrics.c
> @@ -406,6 +406,7 @@
> }
>
> ctrl->cntlid = le16_to_cpu(res.u16);
> + ctrl->start_reconnect_ns = ktime_get_ns();
>
> out_free_data:
> kfree(data);
> @@ -474,8 +475,12 @@
> bool nvmf_should_reconnect(struct nvme_ctrl *ctrl)
> {
> if (ctrl->opts->max_reconnects == -1 ||
> - ctrl->nr_reconnects < ctrl->opts->max_reconnects)
> + ctrl->nr_reconnects < ctrl->opts->max_reconnects){
> + if(ctrl->nr_reconnects == 0)
> + ctrl->start_reconnect_ns = ktime_get_ns();
> +
> return true;
> + }
>
> return false;
> }
> @@ -549,6 +554,8 @@
> {
> if (ctrl->state != NVME_CTRL_DELETING &&
> ctrl->state != NVME_CTRL_DEAD &&
> + !(ctrl->state == NVME_CTRL_CONNECTING && ctrl->opts->fail_fast_tmo_ns >= 0 &&
> + ((ktime_get_ns() - ctrl->start_reconnect_ns) > ctrl->opts->fail_fast_tmo_ns)) &&
> !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
> return BLK_STS_RESOURCE;
I cannot comprehend what is going on here...
We should have a dedicated delayed_work that transitions the controller
to a FAIL_FAST state and cancels the inflight requests again. This
work should be triggered when the error is detected.
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2019-12-17 21:47 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-01 7:59 [PATCH] nvme-fabrics: reject I/O to offline device Victor Gladkov
2019-12-02 22:26 ` Chaitanya Kulkarni
2019-12-02 22:47 ` James Smart
2019-12-03 10:04 ` Victor Gladkov
2019-12-03 16:19 ` James Smart
2019-12-04 8:28 ` Victor Gladkov
2019-12-06 0:38 ` James Smart
2019-12-06 22:18 ` Sagi Grimberg
2019-12-08 12:31 ` Hannes Reinecke
2019-12-09 15:30 ` Victor Gladkov
2019-12-17 18:03 ` James Smart
2019-12-17 21:46 ` Sagi Grimberg [this message]
2019-12-18 22:20 ` James Smart
2019-12-15 12:33 ` Victor Gladkov
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=73006c25-b6a8-fc36-0789-772e3ea59a02@grimberg.me \
--to=sagi@grimberg.me \
--cc=Victor.Gladkov@kioxia.com \
--cc=hare@suse.de \
--cc=james.smart@broadcom.com \
--cc=linux-nvme@lists.infradead.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;
as well as URLs for NNTP newsgroup(s).