Evan Layton wrote:
> Ethan Quach wrote:
>>
>>
>> Evan Layton wrote:
>>> Ethan Quach wrote:
>>>> Evan,
>>>>
>>>> Why not just move lines 272-274 to 301, instead of adding 266-271 ?
>>>
>>> We need to check for a valid name before calling 
>>> getActiveBEAndActiveOnBootBE(be.trgtBeNameOrSnapshot[0]) don't we?
>>>
>>> If I moved lines 272-274 down to 301 we're could be calling into 
>>> getActiveBEAndActiveOnBootBE with an invalid name in
>>> be.trgtBeNameOrSnapshot[0].
>>
>> Interesting.. I looked at getActiveBEAndActiveOnBootBE(), and there's
>> no reason why that function should even be taking an argument.  It
>> uses the argument in an error message, and it doesn't even make sense
>> to use the passed in argument there.
> 
> Hmm you're right that doesn't make any sense to have that there. This 
> argument could be removed and the error message changed so it doesn't 
> contain the argument.

Changing the error to be BEADM_ERR_NO_BES_EXIST seems appropriate.

> 
>>
>> ...anyway, the main issue I have is that I don't want to see the
>> .split() being called twice at 267 and 298.  Could you set an
>> is_snapshot flag at 267, and then use that flag at 282 and 294
>> instead?  You wouldn't need to resplit at 298 either.
> 
> Good point, it really doesn't make sense to be doing the split twice...
> 
> I've added the is_snapshot flag and removed the split at 298. The webrev 
> is updated.

Just one more nit.  284, no need to compare the string if its
a snapshot, so could you switch the order of these two conditions?


thanks,
-ethan


> 
> -evan
> 
>>
>>
>> thanks,
>> -ethan
>>
>>>
>>> -evan
>>>
>>>>
>>>>
>>>> -ethan
>>>>
>>>>
>>>> Evan Layton wrote:
>>>>> Hi,
>>>>>
>>>>> I need to get a review the following simple fix.
>>>>>
>>>>> Some background: This bug was introduced with the fix for 5749. In the
>>>>> case of beadm destroy we validate the name of the BE however if we're
>>>>> destroying a snapshot of a BE we were not splitting out the name of 
>>>>> the
>>>>> BE from the snapshot name before doing the name validation. I checked
>>>>> through the rest of beadm and didn't find any other areas where we 
>>>>> deal
>>>>> with both BE names and snapshots that we were not checking for a 
>>>>> snapshot
>>>>> before validating the BE name.
>>>>>
>>>>> 7071 beadm can fail to destroy snapshot
>>>>> http://defect.opensolaris.org/bz/show_bug.cgi?id=7071
>>>>>
>>>>> Webrev:
>>>>> http://cr.opensolaris.org/~evanl/7071/
>>>>>
>>>>> Thanks!
>>>>> -evan
>>>>> _______________________________________________
>>>>> caiman-discuss mailing list
>>>>> caiman-discuss at opensolaris.org
>>>>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
>>>
> 

Reply via email to