Re: A question about deleting a big file structure from a big disk in Jessie: Why does this work? I'm really worried.
On 20150402_1746-0700, David Christensen wrote:
> On 04/02/2015 04:21 PM, Paul E Condon wrote:
> >For several years I have been making daily backups of my four Debian
> >computers using Rsync and a small script of my own devising. The data
> >has been accumulating on an external USB drive in a partition with the
> >label, gfx5. Some time ago I decided to a make a copy of these data,
> >so I would have more than one copy. I had to use Rsync to do this
> >because it I were to use cp the copies of files labeled by different
> >dates and hard-link together on gfx5 would exceed the capacity on the
> >target disk (which was/is labeled gfx2). This is a simple one line
> >command to Rsync. When I tried, the job would always crash well
> >before completion.
>
> You're using a "Testing" operating system distribution (Jessie), not a
> "Stable" operating system distribution (Wheezy):
>
> 1. If you want to help debug Jessie, then you should create a script that
> demonstrates the undesired behavior on a fresh install of a specific
> snapshot of Jessie and post your script and console session. (E.g. the
> script should not depend upon your data, systems, or networks; it should
> produce similar results on "equivalent" machines.)
Until I discovered a pattern in failures, my default assumption was that
the problems were inattention to detail on my part and frequent upgrades
of Jessie, which can happen almost daily. Does someone have a stable of
i386 computers each one with a particular weekly build on it? I don't.
>
> 2. If you want reliable operations, then you should use Wheezy. If that
> doesn't do what you want, post your console session.
I got into this thinking I want to follow Jessie development, and
tinker with a pet project while exercising Jessie to see if I could
notice any bugs. I don't want to be running Wheezy when Jessie is
released, and I don't want to explain openly why. The evidence that
made me write was gathered from nine windows of ssh sessions on 'gq'
being displayed on 'big'. I would have to make a video 'movie' of the
nine together to show what I saw. I can't do that. I don't have the
knowledge, skill, or equipment. It was gleaned for which windows
changed in coincidence, and which changed singly. If anyone does have
several computers that can be dedicated for a short while, and the
curiosity to see if my observations can be reproduced, I think it
would be nice if they would contact me for more details. I don't want
to write a great treatise that nobody will read or understand. I've
been criticized for the way I write.
The basic list is two computers, two USB hard disks 3TB or maybe
larger. The computer to which the 3TB disks are plugged in should have
12GB or more of swap space to accomodate Rsync of this big tranfer as
one go. (the Rsync options are -aHvv ). And a copious source of
structured data (a local GIT server, perhaps) Maybe someone who knows
about Clouds could do the whold test with Clouds and virtualization,
about which I also know nothing practical.
>
>
> In either case, you will want to check the source and destination file
> systems before invoking 'rsync':
>
> $ man fsck
I know about man fsck. I have read it many times. I have already
eliminated many hypotheses thru careful readings of man fsck. And
careful examination of the source and destination file systems.
I know how little chance I have of convincing this audience that I have
seen what I believe I have seen. Offering to help qualified people to
develop their own independent tests of my hypothesis is the way I want
to go. At some point this will be decided. If I'm right, I really won't
be happy, because it will be very bad for Debian.
Best regards,
--
Paul E Condon
pecondon@mesanetworks.net
Reply to: