Re: iSCSI target: Handling in-flight requests during ctld shutdown

From: Alexander Motin <mav_at_FreeBSD.org>
Date: Thu, 30 Dec 2021 23:06:13 UTC
On 30.12.2021 17:41, John Baldwin wrote:
> On 12/30/21 12:10 PM, Alexander Motin wrote:
>> The checks were in all backends, until in 2c7dc6bae9fd I've moved them
>> to ctl_datamove_done_process().  Backends don't have any special
>> knowledge now to process specific data move error codes.  It was
>> complete code duplication between backends and I've tried to organize
>> that a bit.  Unfortunately we do not have formalized data move error
>> statuses in CTL to do more.  It would be much more interesting (and
>> complicated ;)) if iSCSI (for example) could report data transfer error
>> without killing connection, but that area in general is not very
>> specified in SCSI, since usually each transport has own means of 
>> recovery.
> 
> Hmmm, but this commit seems to have actually broken this case.  That is,
> now the possibly uninitialized data is written to the backing store:
> 
>          /*
> -        * We set status at this point for read commands, and write
> -        * commands with errors.
> +        * We set status at this point for read and compare commands.
>           */
> -       if (io->io_hdr.flags & CTL_FLAG_ABORT) {
> -               ;
> -       } else if ((io->io_hdr.port_status != 0) &&
> -           ((io->io_hdr.status & CTL_STATUS_MASK) == CTL_STATUS_NONE ||
> -            (io->io_hdr.status & CTL_STATUS_MASK) == CTL_SUCCESS)) {
> -               ctl_set_internal_failure(&io->scsiio, /*sks_valid*/ 1,
> -                   /*retry_count*/ io->io_hdr.port_status);
> -       } else if (io->scsiio.kern_data_resid != 0 &&
> -           (io->io_hdr.flags & CTL_FLAG_DATA_MASK) == CTL_FLAG_DATA_OUT &&
> -           ((io->io_hdr.status & CTL_STATUS_MASK) == CTL_STATUS_NONE ||
> -            (io->io_hdr.status & CTL_STATUS_MASK) == CTL_SUCCESS)) {
> -               ctl_set_invalid_field_ciu(&io->scsiio);
> -       } else if ((io->io_hdr.port_status == 0) &&
> -           ((io->io_hdr.status & CTL_STATUS_MASK) == CTL_STATUS_NONE)) {
> +       if ((io->io_hdr.flags & CTL_FLAG_ABORT) == 0 &&
> +           (io->io_hdr.status & CTL_STATUS_MASK) == CTL_STATUS_NONE) {
>                  lbalen = ARGS(beio->io);
>                  if (lbalen->flags & CTL_LLF_READ) {
>                          ctl_set_success(&io->scsiio);
> 
> As ctl_datamove_done_process() will call ctl_set_internal_failure(), but
> ctl_datamove_done() will still call be_move_done which runs the code above.
> The code above doesn't check port_status anywhere, so it will still happily
> write the stale data to the backend unless CTL_FLAG_ABORT is set.  Do the
> backends need to be checking port_status == 0 again in addition to
> CTL_FLAG_ABORT then?

No. Backends check for ((io->io_hdr.status & CTL_STATUS_MASK) == 
CTL_STATUS_NONE) .  Notice that you've received that HARDWARE ERROR 
status, not garbage.

>>>    I do see checks for CTL_FLAG_ABORT, and the handler
>>> for
>>> the CTL_TASK_I_T_NEXUS_RESET does set CTL_FLAG_ABORT on pending 
>>> requests.
>>>
>>> For the tasks in sciscsi_session_terminate_tasks(), those should already
>>> have
>>> CTL_FLAG_ABORT set anyway, but it wouldn't hurt if it were set again by
>>> cfiscsi_data_wait_abort().  For the the cfiscsi_task_management_done
>>> case I'm
>>> less certain, but I suspect there too that returning an internal error
>>> status
>>> back to the initiator is not expected and that it would be better to
>>> just set
>>> CTL_FLAG_ABORT and drop any response?
>>
>> cfiscsi_data_wait_abort() does only one specific thing -- it aborts
>> waiting for data and completes the data move operation, allowing backend
>> to continue (or at least free resources, etc).  It is only
>> cfiscsi_session_terminate_tasks() knows that it should abort data moves
>> while it aborts tasks, since it knows that the connection is dead.
>> cfiscsi_task_management_done() is another special case, since iSCSI
>> explicitly mentions that there will be no further exchanges related to
>> that task, so the data move should explicitly return.  But in general
>> case I think it would be nice to not combine those two facts of aborting
>> task and aborting data transfer, since the last may not necessary imply
>> the first, as the first does not imply the last.
>>
>> It is tempting to just set the flag, but I suspect it may cause more
>> problems (I am worrying about HA peer update).  I'd prefer the race
>> inside the transport to be handled inside the transport.  May be we
>> could just add another flag to be set under the session lock inside
>> cfiscsi_session_terminate_tasks() after it aborted all the tasks, but
>> before it start to abort transfers itself and use it inside
>> cfiscsi_datamove_out() instead of cs_terminating.
> 
> I think though that even that flag wouldn't quite help as you would still
> have the problem that the internal_error response would still be sent on
> the wire if we don't abort the entire task vs aborting just the move.

No.  cfiscsi_datamove_out() called before the new flag is set would 
still try to send R2T over the dying connection to be aborted by the 
cfiscsi_session_terminate_tasks() few milliseconds later. 
cfiscsi_data_wait_abort() would only be needed if 
cfiscsi_session_terminate_tasks() has already passed through the data 
waiters list and waiting for the last tasks completion.

> It may be that all I really need to do is fix the one case in
> cfiscsi_datamove_out() to set CTL_FLAG_ABORT rather than changing
> cfiscsi_data_move_abort()?

I still think it won't be needed with the new flag I proposed, that 
would be cleaner IMO.

-- 
Alexander Motin