bug-ddrescue
[Top][All Lists]
Advanced

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

Re: [Bug-ddrescue] Feature Request: Introduce parameter to skip "slow" (


From: Kim Pedersen
Subject: Re: [Bug-ddrescue] Feature Request: Introduce parameter to skip "slow" (still readable) sectors during first pass, if reading speed is below x kByte/s
Date: Tue, 27 Sep 2011 07:47:26 -0400
User-agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0.2) Gecko/20110902 Thunderbird/6.0.2


Hi Felix,

My interpretation of what happens with read-ahead is the device will be told to fetch say, blocks 100-109, if 103-106 has errors, you'll need to wait for the retry-timeout to take place on all of those blocks. With direct access (-d) you bypass the read-ahead and ddrescue will complete it's first pass a lot quicker as it will skip ahead on encountering the first errors at block 103.

Anyone else, please correct me if I am wrong. Performance wise I've never seen much difference between using -d or not on a good disk, so I always use it. YMMV


Kim


On 2011-09-26 16:43, Felix Ehlermann wrote:
Hi Kim,

thank you for your suggestion.

I have not tried the -d switch yet, but I assume that this setting would affect transfers over the entire disk, not just specific sectors. At the beginning of the drive I get a continuous "current rate" of ~105 MByte/s - not that slow for a magnetic disk IMO. So I don't think that there's a general perfomance issue because of the read ahead being enabled.

But I do understand that the read ahead might slow down transfers on damaged areas - yet the HD did not return IO errors on those slow sectors. It was probably able to relocate them successfully - I have started a second run over the HD which is right now at an average speed of ~37 MByte/s. In the previous run I got ~10 MByte/s average speed. (I started with an empty logfile in order to re-read the entire disk into a second copy)

Actually this rises another interesting question to me:
Is there a way to read data from the drive without having the drive relocate weak / bad sectors upon reading? I would be worried that a weak sektor was read with invalid data but correct (drive internal) checksum at the first attempt - in this case relocation would prevent another attempt at this sector as the (wrong) data read from it will be stored in the spare sector the drives firmware uses as a replacement.

Yours
Felix

On 26.09.2011 19:16, Kim Pedersen wrote:

Hi Felix,

Glad that you are happy with ddrescue - it is very good and very useful indeed.

What command are you using? Have you tried to use -d to disable read-ahead?

Many many years ago I came across the same issue as you, and find recoveries go better (faster) when you disable the OS tendency to read ahead (Sometimes many MBs)


Kim

_______________________________________________
Bug-ddrescue mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/bug-ddrescue




reply via email to

[Prev in Thread] Current Thread [Next in Thread]