Re: synchronous metadata update

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hello,
On Mon, Dec 22, 2008 at 4:10 PM, Sandeep K Sinha
<sandeepksinha@xxxxxxxxx> wrote:
[snip]
>> Can you please describe any scenario, where this may happen?
>> One i can think of is when your /sbin is groked.
>>
> Well, there can be many scenario's which may lead to such situations.
> In commodity file systems liek ext2/ext3 we dont handle sector failures.
> If such in-consistency occurs at your fs level, then you will be in
> big problems.

yes indeed.
>
> Also, there can be possible situations where your superblock gets
> corrupted and the fsck wont have any information pertaining to the
> disk layout of the Fs.
> In such scenario, you won't even have access to the duplicate SB's.
> Duplicate SB's are helpful mostly when you have some kind of
> inconsistencies in the SB data but when the SB itself is corrupted,
> who will tell you where the alternative SB is ?

AFAIK , location of duplicate superblocks is decided by the block size
and thus is not explicitly stored in the superblock, though I am not
sure of this. Maybe one well versed with fs internals can shed some
light on this matter.

Cu,
   --Pradeep

>> I wonder if this is the case, then what happens when fsck itself
>> cannot be invoked. though your superblock is intact you still cannot
>> have a sane state for your fs because effectively fsck is not there.
>> How do we handle such situations?
>
> Obviously you will have to have something like a fsck.
> Look for some other similar tools.
> Also, if fsck fails for its own reason, it will probably be temporary, I assume.
>> Any hints?
>>
>
>
>> Thanks for the inputs Sandeep,
>>    ~Pradeep
>>>
>>> HTH,
>>>
>>>> Thanks,
>>>>       ~Pradeep
>>>>>
>>>>>>
>>>>>> Thanks
>>>>>> Shyam.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Regards,
>>>>> Sandeep.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> "To learn is to change. Education is a process that changes the learner."
>>>>>
>>>>> --
>>>>> To unsubscribe from this list: send an email with
>>>>> "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
>>>>> Please read the FAQ at http://kernelnewbies.org/FAQ
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Pradeep
>>>>
>>>
>>>
>>>
>>> --
>>> Regards,
>>> Sandeep.
>>>
>>>
>>>
>>>
>>>
>>>
>>> "To learn is to change. Education is a process that changes the learner."
>>>
>>
>>
>>
>> --
>> Pradeep
>>
>
>
>
> --
> Regards,
> Sandeep.
>
>
>
>
>
>
> "To learn is to change. Education is a process that changes the learner."
>



-- 
Pradeep

--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ


[Index of Archives]     [Newbies FAQ]     [Linux Kernel Mentors]     [Linux Kernel Development]     [IETF Annouce]     [Git]     [Networking]     [Security]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux SCSI]     [Linux ACPI]
  Powered by Linux