Re: 7.2 update instructions

2016-02-22 Thread William Shu
Yasha/Lamar + 1 for your views on these comparisons, and Lamar's university's 
conclusions (and justifications thereto) when done, on or off list.
William.


On Monday, February 22, 2016 6:54 PM, Lamar Owen  wrote:
 

 On 02/22/2016 11:50 AM, Yasha Karant wrote:


> would it be appropriate for me to post a response?  The differences 
> are deep and fundamental.
>

I can't answer that; a moderator would need to.  I would personally 
welcome a direct e-mail with the explanation myself, as my .edu is 
currently investigating 'CS' curricula (where 'CS' is the 
Google/Microsoft version and not the ACM version of 'CS').

And for the list, one of the more fascinating things you are likely to 
ever read is Yasha Karant's 'A Stick Figure Guide to the Advanced 
Encryption Standard (AES)' which is available at 
http://www.csci.csusb.edu/ykarant/cryptography/aes-cartoon.pdf ; I 
certainly found it interesting.

My paragraph was simply there to let you know that there are probably 
many more IT folk here than CS folk, and IT folk tend to have a very 
hands-on and practical 'here's the standard way to do it' answer and an 
eye towards maintainability, and all of that is just a part of the IT 
mindset.  Neither is the more correct mindset; the mindsets are just 
different.  A CIS-mindset is yet even more different, but that's not 
nearly as well represented here, nor are the CE or SE mindsets, but the 
IT mindset is very much predominant here.  As well, it was to serve to 
let the list as a whole know that there are different mindsets out there 
that are very different from the typical sysadmin IT-centric mindset.


  

Re: 7.2 update instructions

2016-02-22 Thread Lamar Owen

On 02/22/2016 11:50 AM, Yasha Karant wrote:
Actually, it has a great deal to do with the original post; however, 
your exposition of a workable methodology is reasonably clear and will 
be the mechanism for going forward and I thank you for your clarity. 


You're welcome.  My methods were developed through long experience and 
much trial and error.  Very scientific: hypothesize, experiment, verify 
or nullify.  Lather, rinse, repeat.


Presumably, to move existing to-save partitions from the older file 
system structure to the more current structure is not possible with an 
"imaging" method, such as dd, but will work with a full backup of an 
existing high level file system mounted upon a "physical" partition" 
(e. g., using tar perhaps with lossless compression) and then restore.


I actually use an rsync mirror (the copy is done with rsync -avHAX, 
although symlinks can be disturbed if not careful) to another disk; an 
rsync backup is restored quite simply, and using hardlinking can trim 
quite a bit of space used as well as giving you snapshotting. Backuppc I 
think uses a similar methodology, but it is overkill for my purposes.


Filesystem backup/restore tools such as dump and restore or using 
tarfiles are doable, but in my case the tarball would be too large to be 
manageable.  But using tar with stdout and stdin over an ssh pipe is 
very efficient.  The first book in the 'Linux Server Hacks' series has a 
lot of this and is a good 'cookbook' for these types of things and doing 
them the 'standard' way.




The second point you raise -- the difference between computer science 
and engineering versus information technology -- requires a response 
and clarification, as you posted your views to a public list (anyone 
may view/read). 


While this is off-topic for the list as a whole, it might serve as some 
clarification to many on the list who are very deep in the IT mindset 
and for whom the CS mindset seems alien.


However, as your comment is off the mission of this list (as I have 
discovered, engineering design issues are not for this list, but 
rather mostly technology), 


More to the point:  since SL is a rebuild of an already engineered 
system, the engineering being done isn't being done on this list. We're 
just putting together already engineered pieces in a creative way; there 
is plenty of creativity, but it's just a different sort of creativity.


would it be appropriate for me to post a response?  The differences 
are deep and fundamental.




I can't answer that; a moderator would need to.  I would personally 
welcome a direct e-mail with the explanation myself, as my .edu is 
currently investigating 'CS' curricula (where 'CS' is the 
Google/Microsoft version and not the ACM version of 'CS').


And for the list, one of the more fascinating things you are likely to 
ever read is Yasha Karant's 'A Stick Figure Guide to the Advanced 
Encryption Standard (AES)' which is available at 
http://www.csci.csusb.edu/ykarant/cryptography/aes-cartoon.pdf ; I 
certainly found it interesting.


My paragraph was simply there to let you know that there are probably 
many more IT folk here than CS folk, and IT folk tend to have a very 
hands-on and practical 'here's the standard way to do it' answer and an 
eye towards maintainability, and all of that is just a part of the IT 
mindset.  Neither is the more correct mindset; the mindsets are just 
different.  A CIS-mindset is yet even more different, but that's not 
nearly as well represented here, nor are the CE or SE mindsets, but the 
IT mindset is very much predominant here.  As well, it was to serve to 
let the list as a whole know that there are different mindsets out there 
that are very different from the typical sysadmin IT-centric mindset.


Re: 7.2 update instructions

2016-02-22 Thread Yasha Karant

On 02/19/2016 08:09 AM, Lamar Owen wrote:

On 02/18/2016 03:16 PM, Yasha Karant wrote:


As the "old" partition scheme is increasingly considered "obsolete", 
for the new layout scheme, how does one not overwrite the entire file 
system other than having two separate hard drives, a "system" one and 
and "non-system" (e.g., /home ...) one (for which the "hard drives" 
could be multiple drives in a RAID configuration, etc., but not 
"system")?


While this has nothing to do with what you originally posted, I'll 
bite.  In my case I have set up a separate logical volume for /home 
from the one where / is mounted.  Whether this LV is on the same 
volume group as the LV for / is irrelevant; in my case they are on the 
same VG, and I tell (told, in the case of one 'upgrade') the installer 
to use a particular existing LV for /, a particular partition for 
/boot, another LV for swap, and the last LV for /home.  All are set to 
format *except* the one for /home. It took a bit of time to get used 
to the EL7 installer's way of doing mount points, but now that I've 
used it a few times I really prefer it to the old way for many (but 
not all) use cases.


But my question is 'why do you always seem to pick the hard way?' to 
do things.  (I already have a good idea why, actually, as it has to do 
with a basic difference between 'Computer Science' and 'Information 
Technology' (as defined by the ACM's 2008 Computing Curricula 
Standards) and a basic difference between the CS mindset and the IT 
mindset.)  Just understand that most of the advice you're going to get 
here is squarely in the IT (as defined by the ACM) mindset, including 
from me.
Actually, it has a great deal to do with the original post; however, 
your exposition of a workable methodology is reasonably clear and will 
be the mechanism for going forward and I thank you for your clarity.  
Presumably, to move existing to-save partitions from the older file 
system structure to the more current structure is not possible with an 
"imaging" method, such as dd, but will work with a full backup of an 
existing high level file system mounted upon a "physical" partition" (e. 
g., using tar perhaps with lossless compression) and then restore.


The second point you raise -- the difference between computer science 
and engineering versus information technology -- requires a response and 
clarification, as you posted your views to a public list (anyone may 
view/read).  However, as your comment is off the mission of this list 
(as I have discovered, engineering design issues are not for this list, 
but rather mostly technology), would it be appropriate for me to post a 
response?  The differences are deep and fundamental.


Yasha Karant


Re: vlc no audio

2016-02-22 Thread John Pilkington

On 20/02/16 10:35, John Pilkington wrote:


Good.  After being reminded of my problem with the DVD .iso playback I
just reverted to vlc-2.1.6-2.el7.nux.x86_64 (I uninstalled first)

That fixed it, although the opening aspect ratios are wrong.


... and I've been bitten by this before:  don't leave the _local repo 
enabled after a downgrade or yum-cron and auto-upgrades will take you 
right back :-)