I went way off-topic in asking about the pool and would like a moderator to
move the last few posts since mine into another topic if possible please.
Thanks Les for answering those questions. I guess it makes more sense, but it's
not completely apparently what all is going on from what you said or what was
written by Jeffery and Craig. Couldn't there be a way to do block-level and
file-level de-dupe based on how often the file changes or if it's specifically
a database file? The Indexing service in Windows can create a .edb file in
upwards of 6GB. Definitely best to do block-level de-dupe on that rather than
file-level. I think BackupPC could be far more functional if that sort of
backup was available in the system itself. Maybe there could be a way to change
it on a per-client basis or have a client and then have a special area of the
client for that file.
I actually do have DDNS setup, but it's based on my DHCP server and that
doesn't tell it, if the machine connected on Wi-Fi, to prioritize the DNS name
hostname.local to 192.168.0.12 instead of 192.168.0.13. The easiest way to do
any of this, especially with a machine that isn't allowing the pinging of
machines by windows hostnames, is to do it by IP address. The DHCP method in
BackupPC counts from the bottom up. If the Wi-Fi card is setup as a lower IP,
then it will be pulled first also not solving the issue.
The main thing here is, there should be some way of solving this problem in
BackupPC, not creating other problems with having to setup local DDNS servers
on in upwards of 20 machines or more and then having to tell people to turn off
the wireless cards when they plug in. It'd be so much easier just to say
"ClientAlias {} take an array" just like how file/directory excludes work.
Most of what you said isn't a solution, it's yet another problem. If you
thought it'd be better to talk to the Bacula client, then why is SSH-port
forwarding the best option? If BackupPC already knows how to talk through SSH
pub/priv key, then why doesn't it just allow that for Rsyncd as well?
One of the biggest things on the configuration side of things is having a
global config and then having to modify that on a per-machine basis rather than
having configuration profiles or a way to have a global config and then have a
machine config underneath it which allows even more modification. I'd rather
see a method of adding functionality for easier management of multiple machines
which are similar Linux vs Windows rather than seeing "make global for Linux
and go by hand and change all the Windows machines, but if you make a change to
the global config and want that reflected in the Windows machines too, there's
no way to do that other than doing it by hand". See what I mean? One solves the
problem the other does nothing but create more headache and doesn't solve the
problem.
+----------------------------------------------------------------------
|This was sent by Saturn2888 AT gmail DOT com via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
BackupPC-users mailing list
BackupPC-users AT lists.sourceforge DOT net
List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki: http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/
|
