public inbox for [email protected]
 help / color / mirror / Atom feed
From: Sagi Grimberg <[email protected]>
To: Hannes Reinecke <[email protected]>, Christoph Hellwig <[email protected]>
Cc: Kanchan Joshi <[email protected]>,
	[email protected], [email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected],
	[email protected], [email protected]
Subject: Re: [PATCH for-next 4/4] nvme-multipath: add multipathing for uring-passthrough commands
Date: Wed, 13 Jul 2022 18:59:13 +0300	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>


>>>>> Amongst all the other issue we've found the prime problem with 
>>>>> SG_IO is that it needs to be directed to the 'active' path.
>>>>> For the device-mapper has a distinct callout (dm_prepare_ioctl), 
>>>>> which essentially returns the current active path device. And then 
>>>>> the device-mapper core issues the command on that active path.
>>>>>
>>>>> All nice and good, _unless_ that command triggers an error.
>>>>> Normally it'd be intercepted by the dm-multipath end_io handler, 
>>>>> and would set the path to offline.
>>>>> But as ioctls do not use the normal I/O path the end_io handler is 
>>>>> never called, and further SG_IO calls are happily routed down the 
>>>>> failed path.
>>>>>
>>>>> And the customer had to use SG_IO (or, in qemu-speak, LUN 
>>>>> passthrough) as his application/filesystem makes heavy use of 
>>>>> persistent reservations.
>>>>
>>>> How did this conclude Hannes?
>>>
>>> It didn't. The proposed interface got rejected, and now we need to 
>>> come up with an alternative solution.
>>> Which we haven't found yet.
>>
>> Lets assume for the sake of discussion, had dm-mpath set a path to be
>> offline on ioctl errors, what would qemu do upon this error? blindly
>> retry? Until When? Or would qemu need to learn about the path tables in
>> order to know when there is at least one online path in order to retry?
>>
> IIRC that was one of the points why it got rejected.
> Ideally we would return an errno indicating that the path had failed, 
> but further paths are available, so a retry is in order.
> Once no paths are available qemu would be getting a different error 
> indicating that all paths are failed.

There is no such no-paths-available error.

> 
> But we would be overloading a new meaning to existing error numbers, or 
> even inventing our own error numbers. Which makes it rather awkward to use.

I agree that this sounds awkward.

> Ideally we would be able to return this as the SG_IO status, as that is 
> well capable of expressing these situations. But then we would need to 
> parse and/or return the error ourselves, essentially moving sg_io 
> funtionality into dm-mpath. Also not what one wants.

uring actually should send back the cqe for passthru, but there is no
concept like "Path error, but no paths are available".

> 
>> What is the model that a passthru consumer needs to follow when
>> operating against a mpath device?
> 
> The model really is that passthru consumer needs to deal with these errors:
> - No error (obviously)
> - I/O error (error status will not change with a retry)
> - Temporary/path related error (error status might change with a retry)
> 
> Then the consumer can decide whether to invoke a retry (for the last 
> class), or whether it should pass up that error, as maybe there are 
> applications with need a quick response time and can handle temporary 
> failures (or, in fact, want to be informed about temporary failures).
> 
> IE the 'DNR' bit should serve nicely here, keeping in mind that we might 
> need to 'fake' an NVMe error status if the connection is severed.

uring passthru sends the cqe status to userspace IIRC. But nothing in
there indicates about path availability. That would be something that
userspace would need to reconcile on its own from traversing sysfs or
alike...

      reply	other threads:[~2022-07-13 15:59 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20220711110753epcas5p4169b9e288d15ca35740dbb66a6f6983a@epcas5p4.samsung.com>
2022-07-11 11:01 ` [PATCH for-next 0/4] nvme-multipathing for uring-passthrough Kanchan Joshi
     [not found]   ` <CGME20220711110800epcas5p3d338dd486fd778c5ba5bfe93a91ec8bd@epcas5p3.samsung.com>
2022-07-11 11:01     ` [PATCH for-next 1/4] io_uring, nvme: rename a function Kanchan Joshi
2022-07-14 13:55       ` Ming Lei
     [not found]   ` <CGME20220711110812epcas5p33aa90b23aa62fb11722aa8195754becf@epcas5p3.samsung.com>
2022-07-11 11:01     ` [PATCH for-next 2/4] nvme: compact nvme_uring_cmd_pdu struct Kanchan Joshi
2022-07-12  6:32       ` Christoph Hellwig
     [not found]   ` <CGME20220711110824epcas5p22c8e945cb8c3c3ac46c8c2b5ab55db9b@epcas5p2.samsung.com>
2022-07-11 11:01     ` [PATCH for-next 3/4] io_uring: grow a field in struct io_uring_cmd Kanchan Joshi
2022-07-11 17:00       ` Sagi Grimberg
2022-07-11 17:19         ` Jens Axboe
2022-07-11 17:18       ` Jens Axboe
2022-07-11 17:55         ` Sagi Grimberg
2022-07-11 18:22           ` Sagi Grimberg
2022-07-11 18:24             ` Jens Axboe
2022-07-11 18:58               ` Sagi Grimberg
2022-07-12 11:40               ` Kanchan Joshi
2022-07-14  3:40             ` Ming Lei
2022-07-14  8:19               ` Kanchan Joshi
2022-07-14 15:30                 ` Daniel Wagner
2022-07-15 11:07                   ` Kanchan Joshi
2022-07-18  9:03                     ` Daniel Wagner
     [not found]   ` <CGME20220711110827epcas5p3fd81f142f55ca3048abc38a9ef0d0089@epcas5p3.samsung.com>
2022-07-11 11:01     ` [PATCH for-next 4/4] nvme-multipath: add multipathing for uring-passthrough commands Kanchan Joshi
2022-07-11 13:51       ` Sagi Grimberg
2022-07-11 15:12         ` Stefan Metzmacher
2022-07-11 16:58           ` Sagi Grimberg
2022-07-11 18:54           ` Kanchan Joshi
2022-07-11 18:37         ` Kanchan Joshi
2022-07-11 19:56           ` Sagi Grimberg
2022-07-12  4:23             ` Kanchan Joshi
2022-07-12 21:26               ` Sagi Grimberg
2022-07-13  5:37                 ` Kanchan Joshi
2022-07-13  9:03                   ` Sagi Grimberg
2022-07-13 11:28                     ` Kanchan Joshi
2022-07-13 12:17                       ` Sagi Grimberg
2022-07-14 15:14                   ` Ming Lei
2022-07-14 23:05                     ` Kanchan Joshi
2022-07-15  1:35                       ` Ming Lei
2022-07-15  1:46                         ` Ming Lei
2022-07-15  4:24                           ` Kanchan Joshi
2022-07-12  6:52       ` Christoph Hellwig
2022-07-12 11:33         ` Kanchan Joshi
2022-07-12 20:13         ` Sagi Grimberg
2022-07-13  5:36           ` Christoph Hellwig
2022-07-13  8:04             ` Sagi Grimberg
2022-07-13 10:12               ` Christoph Hellwig
2022-07-13 11:00                 ` Sagi Grimberg
2022-07-13 11:28                   ` Christoph Hellwig
2022-07-13 12:16                     ` Sagi Grimberg
2022-07-13 11:49                   ` Hannes Reinecke
2022-07-13 12:43                     ` Sagi Grimberg
2022-07-13 13:30                       ` Hannes Reinecke
2022-07-13 13:41                         ` Sagi Grimberg
2022-07-13 14:07                           ` Hannes Reinecke
2022-07-13 15:59                             ` Sagi Grimberg [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 \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /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