qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected
@ 2014-10-30 11:23 Fam Zheng
  2014-10-30 11:43 ` Peter Lieven
  0 siblings, 1 reply; 5+ messages in thread
From: Fam Zheng @ 2014-10-30 11:23 UTC (permalink / raw)
  To: qemu-devel
  Cc: Kevin Wolf, Paolo Bonzini, Peter Lieven, Stefan Hajnoczi,
	Ronnie Sahlberg

Before, when a write protected iSCSI target is attached as scsi-disk
with BDRV_O_RDWR, we report it as writable, while in fact all writes
will fail.

One way to improve this is to report write protect flag as true to
guest, but a even better way is to refuse using a write protected LUN to
guest.

Target write protect flag is checked with a mode sense query.

Signed-off-by: Fam Zheng <famz@redhat.com>
---
v2: Improve error message.
    Fall back to a warning if mode sense failed.
    Check unmarshal return value.
---
 block/iscsi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/block/iscsi.c b/block/iscsi.c
index 233f462..dcacbca 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -1219,6 +1219,44 @@ static void iscsi_attach_aio_context(BlockDriverState *bs,
               qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
 }
 
+static bool iscsi_is_write_protected(IscsiLun *iscsilun)
+{
+    struct scsi_task *task;
+    struct scsi_mode_sense *ms = NULL;
+
+    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
+            1, SCSI_MODESENSE_PC_CURRENT,
+            0x3F,
+            0, 255);
+
+    if (task == NULL) {
+        error_report("Failed to send MODE_SENSE6 command: %s",
+                     iscsi_get_error(iscsilun->iscsi));
+        goto out;
+    }
+
+    if (task->status != SCSI_STATUS_GOOD) {
+        error_report("MODE_SENSE6 failed: %s",
+                     iscsi_get_error(iscsilun->iscsi));
+        goto out;
+    }
+    ms = scsi_datain_unmarshall(task);
+    if (!ms) {
+        error_report("MODE_SENSE6 failed: %s",
+                     iscsi_get_error(iscsilun->iscsi));
+        goto out;
+    }
+out:
+    if (task) {
+        scsi_free_scsi_task(task);
+    }
+    if (!ms) {
+        error_report("Assuming write enabled");
+        return false;
+    }
+    return ms->device_specific_parameter & 0x80;
+}
+
 /*
  * We support iscsi url's on the form
  * iscsi://[<username>%<password>@]<host>[:<port>]/<targetname>/<lun>
@@ -1339,6 +1377,14 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
     scsi_free_scsi_task(task);
     task = NULL;
 
+    /* Check the write protect flag of the LUN if we want to write */
+    if ((flags & BDRV_O_RDWR)
+            && iscsi_is_write_protected(iscsilun)) {
+        error_setg(errp, "Cannot open a write protected LUN as read-write");
+        ret = -EPERM;
+        goto out;
+    }
+
     iscsi_readcapacity_sync(iscsilun, &local_err);
     if (local_err != NULL) {
         error_propagate(errp, local_err);
-- 
1.9.3

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected
  2014-10-30 11:23 [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected Fam Zheng
@ 2014-10-30 11:43 ` Peter Lieven
  2014-10-30 12:09   ` Paolo Bonzini
  0 siblings, 1 reply; 5+ messages in thread
From: Peter Lieven @ 2014-10-30 11:43 UTC (permalink / raw)
  To: Fam Zheng, qemu-devel
  Cc: Kevin Wolf, Paolo Bonzini, Stefan Hajnoczi, Ronnie Sahlberg

On 30.10.2014 12:23, Fam Zheng wrote:
> Before, when a write protected iSCSI target is attached as scsi-disk
> with BDRV_O_RDWR, we report it as writable, while in fact all writes
> will fail.
>
> One way to improve this is to report write protect flag as true to
> guest, but a even better way is to refuse using a write protected LUN to
> guest.
>
> Target write protect flag is checked with a mode sense query.
>
> Signed-off-by: Fam Zheng <famz@redhat.com>
> ---
> v2: Improve error message.
>      Fall back to a warning if mode sense failed.
>      Check unmarshal return value.
> ---
>   block/iscsi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>   1 file changed, 46 insertions(+)
>
> diff --git a/block/iscsi.c b/block/iscsi.c
> index 233f462..dcacbca 100644
> --- a/block/iscsi.c
> +++ b/block/iscsi.c
> @@ -1219,6 +1219,44 @@ static void iscsi_attach_aio_context(BlockDriverState *bs,
>                 qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
>   }
>   
> +static bool iscsi_is_write_protected(IscsiLun *iscsilun)
> +{
> +    struct scsi_task *task;
> +    struct scsi_mode_sense *ms = NULL;
> +
> +    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
> +            1, SCSI_MODESENSE_PC_CURRENT,
> +            0x3F,
> +            0, 255);
> +
> +    if (task == NULL) {
> +        error_report("Failed to send MODE_SENSE6 command: %s",
> +                     iscsi_get_error(iscsilun->iscsi));
> +        goto out;
> +    }
> +
> +    if (task->status != SCSI_STATUS_GOOD) {
> +        error_report("MODE_SENSE6 failed: %s",
> +                     iscsi_get_error(iscsilun->iscsi));
> +        goto out;
> +    }
> +    ms = scsi_datain_unmarshall(task);
> +    if (!ms) {
> +        error_report("MODE_SENSE6 failed: %s",
> +                     iscsi_get_error(iscsilun->iscsi));
> +        goto out;
> +    }
> +out:
> +    if (task) {
> +        scsi_free_scsi_task(task);
> +    }
> +    if (!ms) {

ms points to freed memory after scsi_free_scsi_task.
furthermore the requests likely fails with task->status != SCSI_STATUS_GOOD
if the modesense implementation is broken etc.

I would rework iscsi_is_write_protected to:

static bool iscsi_is_write_protected(IscsiLun *iscsilun)
{
     struct scsi_task *task;
     struct scsi_mode_sense *ms = NULL;
     bool wrprotected = false;

     task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
             1, SCSI_MODESENSE_PC_CURRENT,
             0x3F,
             0, 255);

     if (task == NULL || task->status != SCSI_STATUS_GOOD) {
         goto fail;
     }
     ms = scsi_datain_unmarshall(task);
     if (!ms) {
         goto fail;
     }
     wrprotected = ms->device_specific_parameter & 0x80;
     goto out;
fail:
     error_report("MODE_SENSE6 failed: %s. Assuming write enabled",
                          iscsi_get_error(iscsilun->iscsi));
out:
     if (task) {
         scsi_free_scsi_task(task);
     }
     return wrprotected;
}

Peter


> +        error_report("Assuming write enabled");
> +        return false;
> +    }
> +    return ms->device_specific_parameter & 0x80;
> +}
> +
>   /*
>    * We support iscsi url's on the form
>    * iscsi://[<username>%<password>@]<host>[:<port>]/<targetname>/<lun>
> @@ -1339,6 +1377,14 @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags,
>       scsi_free_scsi_task(task);
>       task = NULL;
>   
> +    /* Check the write protect flag of the LUN if we want to write */
> +    if ((flags & BDRV_O_RDWR)
> +            && iscsi_is_write_protected(iscsilun)) {
> +        error_setg(errp, "Cannot open a write protected LUN as read-write");
> +        ret = -EPERM;
> +        goto out;
> +    }
> +
>       iscsi_readcapacity_sync(iscsilun, &local_err);
>       if (local_err != NULL) {
>           error_propagate(errp, local_err);


-- 

Mit freundlichen Grüßen

Peter Lieven

...........................................................

   KAMP Netzwerkdienste GmbH
   Vestische Str. 89-91 | 46117 Oberhausen
   Tel: +49 (0) 208.89 402-50 | Fax: +49 (0) 208.89 402-40
   pl@kamp.de | http://www.kamp.de

   Geschäftsführer: Heiner Lante | Michael Lante
   Amtsgericht Duisburg | HRB Nr. 12154
   USt-Id-Nr.: DE 120607556

...........................................................

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected
  2014-10-30 11:43 ` Peter Lieven
@ 2014-10-30 12:09   ` Paolo Bonzini
  2014-10-30 12:31     ` Fam Zheng
  2014-10-30 12:55     ` Peter Lieven
  0 siblings, 2 replies; 5+ messages in thread
From: Paolo Bonzini @ 2014-10-30 12:09 UTC (permalink / raw)
  To: Peter Lieven, Fam Zheng, qemu-devel
  Cc: Kevin Wolf, Stefan Hajnoczi, Ronnie Sahlberg



On 10/30/2014 12:43 PM, Peter Lieven wrote:
> On 30.10.2014 12:23, Fam Zheng wrote:
>> Before, when a write protected iSCSI target is attached as scsi-disk
>> with BDRV_O_RDWR, we report it as writable, while in fact all writes
>> will fail.
>>
>> One way to improve this is to report write protect flag as true to
>> guest, but a even better way is to refuse using a write protected LUN to
>> guest.
>>
>> Target write protect flag is checked with a mode sense query.
>>
>> Signed-off-by: Fam Zheng <famz@redhat.com>
>> ---
>> v2: Improve error message.
>>      Fall back to a warning if mode sense failed.
>>      Check unmarshal return value.
>> ---
>>   block/iscsi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>   1 file changed, 46 insertions(+)
>>
>> diff --git a/block/iscsi.c b/block/iscsi.c
>> index 233f462..dcacbca 100644
>> --- a/block/iscsi.c
>> +++ b/block/iscsi.c
>> @@ -1219,6 +1219,44 @@ static void
>> iscsi_attach_aio_context(BlockDriverState *bs,
>>                 qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
>>   }
>>   +static bool iscsi_is_write_protected(IscsiLun *iscsilun)
>> +{
>> +    struct scsi_task *task;
>> +    struct scsi_mode_sense *ms = NULL;
>> +
>> +    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
>> +            1, SCSI_MODESENSE_PC_CURRENT,
>> +            0x3F,
>> +            0, 255);
>> +
>> +    if (task == NULL) {
>> +        error_report("Failed to send MODE_SENSE6 command: %s",
>> +                     iscsi_get_error(iscsilun->iscsi));
>> +        goto out;
>> +    }
>> +
>> +    if (task->status != SCSI_STATUS_GOOD) {
>> +        error_report("MODE_SENSE6 failed: %s",
>> +                     iscsi_get_error(iscsilun->iscsi));
>> +        goto out;
>> +    }
>> +    ms = scsi_datain_unmarshall(task);
>> +    if (!ms) {
>> +        error_report("MODE_SENSE6 failed: %s",
>> +                     iscsi_get_error(iscsilun->iscsi));
>> +        goto out;
>> +    }
>> +out:
>> +    if (task) {
>> +        scsi_free_scsi_task(task);
>> +    }
>> +    if (!ms) {
> 
> ms points to freed memory after scsi_free_scsi_task.
> furthermore the requests likely fails with task->status != SCSI_STATUS_GOOD
> if the modesense implementation is broken etc.

This is a mix of your and Fam's code.  Looks good?

static bool iscsi_is_write_protected(IscsiLun *iscsilun)
{
    struct scsi_task *task;
    struct scsi_mode_sense *ms = NULL;
    bool wrprotected = false;

    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
                                 1, SCSI_MODESENSE_PC_CURRENT,
                                 0x3F, 0, 255);
    if (task == NULL) {
        error_report("Failed to send MODE_SENSE(6) command: %s",
                     iscsi_get_error(iscsilun->iscsi));
        goto out;
    }

    if (task->status != SCSI_STATUS_GOOD) {
        error_report("MODE_SENSE(6) failed: %s",
                     iscsi_get_error(iscsilun->iscsi));
        goto out;
    }
    ms = scsi_datain_unmarshall(task);
    if (!ms) {
        error_report("Failed to unmarshall MODE_SENSE(6) data: %s",
                     iscsi_get_error(iscsilun->iscsi));
        goto out;
    }
    wrprotected = ms->device_specific_parameter & 0x80;

out:
    if (task) {
        scsi_free_scsi_task(task);
    }
    return wrprotected;
}



Paolo

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected
  2014-10-30 12:09   ` Paolo Bonzini
@ 2014-10-30 12:31     ` Fam Zheng
  2014-10-30 12:55     ` Peter Lieven
  1 sibling, 0 replies; 5+ messages in thread
From: Fam Zheng @ 2014-10-30 12:31 UTC (permalink / raw)
  To: Paolo Bonzini
  Cc: Kevin Wolf, Stefan Hajnoczi, Peter Lieven, qemu-devel,
	Ronnie Sahlberg

On Thu, 10/30 13:09, Paolo Bonzini wrote:
> 
> 
> On 10/30/2014 12:43 PM, Peter Lieven wrote:
> > On 30.10.2014 12:23, Fam Zheng wrote:
> >> Before, when a write protected iSCSI target is attached as scsi-disk
> >> with BDRV_O_RDWR, we report it as writable, while in fact all writes
> >> will fail.
> >>
> >> One way to improve this is to report write protect flag as true to
> >> guest, but a even better way is to refuse using a write protected LUN to
> >> guest.
> >>
> >> Target write protect flag is checked with a mode sense query.
> >>
> >> Signed-off-by: Fam Zheng <famz@redhat.com>
> >> ---
> >> v2: Improve error message.
> >>      Fall back to a warning if mode sense failed.
> >>      Check unmarshal return value.
> >> ---
> >>   block/iscsi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
> >>   1 file changed, 46 insertions(+)
> >>
> >> diff --git a/block/iscsi.c b/block/iscsi.c
> >> index 233f462..dcacbca 100644
> >> --- a/block/iscsi.c
> >> +++ b/block/iscsi.c
> >> @@ -1219,6 +1219,44 @@ static void
> >> iscsi_attach_aio_context(BlockDriverState *bs,
> >>                 qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
> >>   }
> >>   +static bool iscsi_is_write_protected(IscsiLun *iscsilun)
> >> +{
> >> +    struct scsi_task *task;
> >> +    struct scsi_mode_sense *ms = NULL;
> >> +
> >> +    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
> >> +            1, SCSI_MODESENSE_PC_CURRENT,
> >> +            0x3F,
> >> +            0, 255);
> >> +
> >> +    if (task == NULL) {
> >> +        error_report("Failed to send MODE_SENSE6 command: %s",
> >> +                     iscsi_get_error(iscsilun->iscsi));
> >> +        goto out;
> >> +    }
> >> +
> >> +    if (task->status != SCSI_STATUS_GOOD) {
> >> +        error_report("MODE_SENSE6 failed: %s",
> >> +                     iscsi_get_error(iscsilun->iscsi));
> >> +        goto out;
> >> +    }
> >> +    ms = scsi_datain_unmarshall(task);
> >> +    if (!ms) {
> >> +        error_report("MODE_SENSE6 failed: %s",
> >> +                     iscsi_get_error(iscsilun->iscsi));
> >> +        goto out;
> >> +    }
> >> +out:
> >> +    if (task) {
> >> +        scsi_free_scsi_task(task);
> >> +    }
> >> +    if (!ms) {
> > 
> > ms points to freed memory after scsi_free_scsi_task.
> > furthermore the requests likely fails with task->status != SCSI_STATUS_GOOD
> > if the modesense implementation is broken etc.
> 
> This is a mix of your and Fam's code.  Looks good?
> 
> static bool iscsi_is_write_protected(IscsiLun *iscsilun)
> {
>     struct scsi_task *task;
>     struct scsi_mode_sense *ms = NULL;
>     bool wrprotected = false;
> 
>     task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
>                                  1, SCSI_MODESENSE_PC_CURRENT,
>                                  0x3F, 0, 255);
>     if (task == NULL) {
>         error_report("Failed to send MODE_SENSE(6) command: %s",
>                      iscsi_get_error(iscsilun->iscsi));
>         goto out;
>     }
> 
>     if (task->status != SCSI_STATUS_GOOD) {
>         error_report("MODE_SENSE(6) failed: %s",
>                      iscsi_get_error(iscsilun->iscsi));
>         goto out;
>     }
>     ms = scsi_datain_unmarshall(task);
>     if (!ms) {
>         error_report("Failed to unmarshall MODE_SENSE(6) data: %s",
>                      iscsi_get_error(iscsilun->iscsi));
>         goto out;
>     }
>     wrprotected = ms->device_specific_parameter & 0x80;
> 
> out:
>     if (task) {
>         scsi_free_scsi_task(task);
>     }
>     return wrprotected;
> }

Looks good. Thanks!

Fam

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected
  2014-10-30 12:09   ` Paolo Bonzini
  2014-10-30 12:31     ` Fam Zheng
@ 2014-10-30 12:55     ` Peter Lieven
  1 sibling, 0 replies; 5+ messages in thread
From: Peter Lieven @ 2014-10-30 12:55 UTC (permalink / raw)
  To: Paolo Bonzini, Fam Zheng, qemu-devel
  Cc: Kevin Wolf, Stefan Hajnoczi, Ronnie Sahlberg

On 30.10.2014 13:09, Paolo Bonzini wrote:
>
> On 10/30/2014 12:43 PM, Peter Lieven wrote:
>> On 30.10.2014 12:23, Fam Zheng wrote:
>>> Before, when a write protected iSCSI target is attached as scsi-disk
>>> with BDRV_O_RDWR, we report it as writable, while in fact all writes
>>> will fail.
>>>
>>> One way to improve this is to report write protect flag as true to
>>> guest, but a even better way is to refuse using a write protected LUN to
>>> guest.
>>>
>>> Target write protect flag is checked with a mode sense query.
>>>
>>> Signed-off-by: Fam Zheng <famz@redhat.com>
>>> ---
>>> v2: Improve error message.
>>>       Fall back to a warning if mode sense failed.
>>>       Check unmarshal return value.
>>> ---
>>>    block/iscsi.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>>>    1 file changed, 46 insertions(+)
>>>
>>> diff --git a/block/iscsi.c b/block/iscsi.c
>>> index 233f462..dcacbca 100644
>>> --- a/block/iscsi.c
>>> +++ b/block/iscsi.c
>>> @@ -1219,6 +1219,44 @@ static void
>>> iscsi_attach_aio_context(BlockDriverState *bs,
>>>                  qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + NOP_INTERVAL);
>>>    }
>>>    +static bool iscsi_is_write_protected(IscsiLun *iscsilun)
>>> +{
>>> +    struct scsi_task *task;
>>> +    struct scsi_mode_sense *ms = NULL;
>>> +
>>> +    task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
>>> +            1, SCSI_MODESENSE_PC_CURRENT,
>>> +            0x3F,
>>> +            0, 255);
>>> +
>>> +    if (task == NULL) {
>>> +        error_report("Failed to send MODE_SENSE6 command: %s",
>>> +                     iscsi_get_error(iscsilun->iscsi));
>>> +        goto out;
>>> +    }
>>> +
>>> +    if (task->status != SCSI_STATUS_GOOD) {
>>> +        error_report("MODE_SENSE6 failed: %s",
>>> +                     iscsi_get_error(iscsilun->iscsi));
>>> +        goto out;
>>> +    }
>>> +    ms = scsi_datain_unmarshall(task);
>>> +    if (!ms) {
>>> +        error_report("MODE_SENSE6 failed: %s",
>>> +                     iscsi_get_error(iscsilun->iscsi));
>>> +        goto out;
>>> +    }
>>> +out:
>>> +    if (task) {
>>> +        scsi_free_scsi_task(task);
>>> +    }
>>> +    if (!ms) {
>> ms points to freed memory after scsi_free_scsi_task.
>> furthermore the requests likely fails with task->status != SCSI_STATUS_GOOD
>> if the modesense implementation is broken etc.
> This is a mix of your and Fam's code.  Looks good?
>
> static bool iscsi_is_write_protected(IscsiLun *iscsilun)
> {
>      struct scsi_task *task;
>      struct scsi_mode_sense *ms = NULL;
>      bool wrprotected = false;
>
>      task = iscsi_modesense6_sync(iscsilun->iscsi, iscsilun->lun,
>                                   1, SCSI_MODESENSE_PC_CURRENT,
>                                   0x3F, 0, 255);
>      if (task == NULL) {
>          error_report("Failed to send MODE_SENSE(6) command: %s",
>                       iscsi_get_error(iscsilun->iscsi));
>          goto out;
>      }
>
>      if (task->status != SCSI_STATUS_GOOD) {
>          error_report("MODE_SENSE(6) failed: %s",
>                       iscsi_get_error(iscsilun->iscsi));
>          goto out;
>      }
>      ms = scsi_datain_unmarshall(task);
>      if (!ms) {
>          error_report("Failed to unmarshall MODE_SENSE(6) data: %s",
>                       iscsi_get_error(iscsilun->iscsi));
>          goto out;
>      }
>      wrprotected = ms->device_specific_parameter & 0x80;
>
> out:
>      if (task) {
>          scsi_free_scsi_task(task);
>      }
>      return wrprotected;
> }

i would add the prefix "iSCSI: " to the error_reports as we
have it for other outputs. (noticed this after writing my mail).

Otherwise looks good.

Reviewed-by: Peter Lieven <pl@kamp.de>

and actually also

Tested-by: Peter Lieven <pl@kamp.de>

Peter

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2014-10-31 15:52 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-10-30 11:23 [Qemu-devel] [PATCH v2] iscsi: Refuse to open as writable if the LUN is write protected Fam Zheng
2014-10-30 11:43 ` Peter Lieven
2014-10-30 12:09   ` Paolo Bonzini
2014-10-30 12:31     ` Fam Zheng
2014-10-30 12:55     ` Peter Lieven

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).