01-18-2011, 04:00 PM
|
#21
|
|
The new goggles also do nothing.
Join Date: Oct 2001
Location: Calgary
|
Oh yeah, iSCSI would work great too now that I've got a NAS that supports it.
Must say the software of the Synology is pretty great. Though I did end up getting the wrong one, I got the "home" version which has all the features, but the disk throughput is not much better than my DNS-323 was. Instead of a 411j I need a 411+
I'm using the Synology as an iSCSI target for a VMware ESXi server and running a virtual machine off the storage, works great!
|
|
|
01-18-2011, 04:06 PM
|
#22
|
|
First Line Centre
Join Date: Mar 2009
Location: Brisbane, Australia
|
Quote:
Originally Posted by mykalberta
Were there no tape backups being used?
|
Thanks for the reply, mykalberta. Tape backups were being used, but apparently taking days to do the full backup, so when the pickup service arrived on schedule, they weren't ready the last couple of weeks. I don't know where those tapes were, just that they weren't available for recovery. I haven't gotten the story on the incrementals yet, but my understanding is that they weren't being done regularly either. The net result that has been verified is that we would have lost two weeks work. We have a Double-Take server setup for immediate localised disaster recovery, but since it's in the same building, obviously something like a flood will render it useless.
It's a 200 person office with dedicated IT staff whose job it is to ensure these things don't happen. My thought is that getting rid of some of the human dependency would be a good thing, hence my wondering about using an online backup service for incremental stuff (in addition to tape).
|
|
|
01-18-2011, 05:08 PM
|
#23
|
|
Franchise Player
Join Date: Aug 2005
Location: Calgary
|
I am not surprised you lost 2 weeks worth of data. Things like that do happen and tape systems arent exactly something you keep more than one of them around. We had something similar go on for almost a month until we found out that the SCSI cable was bad. Not something you think of but after warrantying the tape drive 4 different times unplugging then plugging in the SCSI must have busted it.
I am a little surprised the data servers wernt moved offsite prior to the flood. I hadnt thought the flood down there was something unexpected ala New Orleans.
__________________
MYK - Supports Arizona to democtratically pass laws for the state of Arizona
Rudy was the only hope in 08
2011 Election: Cons 40% - Nanos 38% Ekos 34%
|
|
|
01-18-2011, 07:59 PM
|
#24
|
|
First Line Centre
Join Date: Mar 2009
Location: Brisbane, Australia
|
Sorry, I didn't make that very clear. We didn't actually lose any data, we would have if the flooding which nearly reached our building actually did. We moved the Double-Take server offsite when we found out the tape backup was non-existent.
My concern is more systematic than the particular flood instance. For example, if we have a fire... hooped. In addition to the physical systems, the checks need to be in place to verify that things are working.
|
|
|
01-18-2011, 08:38 PM
|
#25
|
|
#1 Goaltender
|
Quote:
Originally Posted by photon
Oh yeah, iSCSI would work great too now that I've got a NAS that supports it.
Must say the software of the Synology is pretty great. Though I did end up getting the wrong one, I got the "home" version which has all the features, but the disk throughput is not much better than my DNS-323 was. Instead of a 411j I need a 411+
I'm using the Synology as an iSCSI target for a VMware ESXi server and running a virtual machine off the storage, works great!
|
Nice. iSCSI will be the next big overhaul of my home network. No more local storage on any machines, other than maybe a small solid state boot drive.
__________________
-Scott
|
|
|
01-18-2011, 08:52 PM
|
#26
|
|
The new goggles also do nothing.
Join Date: Oct 2001
Location: Calgary
|
I'd want a pretty robust server to do that, but that'd be pretty cool. How would shared drives work then? Windows treats an iSCSI disk as a local hardware disk doesn't it? But if you had two machines accessing the same disk at the same time, would that mess things up?
|
|
|
01-18-2011, 09:11 PM
|
#27
|
|
#1 Goaltender
|
Quote:
Originally Posted by photon
I'd want a pretty robust server to do that, but that'd be pretty cool. How would shared drives work then? Windows treats an iSCSI disk as a local hardware disk doesn't it? But if you had two machines accessing the same disk at the same time, would that mess things up?
|
Probably I'd just network boot everything over iSCSI. gPXE will do iSCSI booting for Linux and Windows 7 apparently fairly easily - and I guess the Macs in the house would just use NFS or whatever.
I haven't really thought it through beyond just wanting a single big box in the utility room in the basement that runs the entire house, hosts my VM's, etc.
Well maybe two boxes. Redundancy and all that
__________________
-Scott
|
|
|
02-01-2011, 03:00 PM
|
#28
|
|
Dances with Wolves
Join Date: Jun 2006
Location: Section 304
|
|
|
|
02-01-2011, 04:01 PM
|
#29
|
|
The new goggles also do nothing.
Join Date: Oct 2001
Location: Calgary
|
10% of users using 50% of resources.. 70 petabytes!
There's no such thing as unlimited.
|
|
|
02-03-2011, 08:22 PM
|
#31
|
|
Franchise Player
Join Date: Nov 2006
Location: Supporting Urban Sprawl
|
Crashplan, uploading to another PC located remotely.
It is free, aside from purchasing a hard drive(s) for that location.
__________________
"Wake up, Luigi! The only time plumbers sleep on the job is when we're working by the hour."
|
|
|
02-03-2011, 09:04 PM
|
#32
|
|
Craig McTavish' Merkin
|
Quote:
Originally Posted by fredr123
|
I heartily endorse Backblaze. I have over 1 TB backed up and the service has been excellent.
|
|
|
02-03-2011, 09:05 PM
|
#33
|
|
#1 Goaltender
|
Quote:
Originally Posted by photon
10% of users using 50% of resources.. 70 petabytes!
There's no such thing as unlimited.
|
Mozy has 1 million users. 10% of those users use 90% of the space. Mozy stores 70 petabytes of data.
So, that means 100,000 users are consuming 63 petabytes of space. That's 630 gigs per user on average.
The other 90% of the users would be 900,000, using 7 petabytes of space, for an average of 62 gigs per user.
Neither number seems unrealistic or untenable to me. The fact that Mozy had to hike rates as dramatically as they did tells me that they really, really messed up their business model - I can buy a Drobo every year for the price Mozy wants to charge me now. Considering I buy my Mozy in 2 year increments, I could buy 2 Drobo's up front. That's sick.
Oh well, on to Backblaze. They seem to be doing a lot of their own R&D and engineering with their storage pods, which they seem to feel will allow them to continue to offer flat rate pricing. Time will tell. My initial upload to Backblaze will be done in a couple months, right in time for my Mozy subscription (at the fixed rate) to expire.
__________________
-Scott
|
|
|
02-05-2011, 07:42 PM
|
#34
|
|
#1 Goaltender
Join Date: Mar 2006
Location: Underground
|
I use Arq on my Mac as a front-end for S3. I have lots of metadata on lab files / images and Arq promised to keep these accurate during backup and it has.
Admittedly I've not tried other front ends for S3.
|
|
|
07-27-2011, 11:48 PM
|
#35
|
|
Franchise Player
Join Date: Aug 2004
Location: Calgary
|
I see that Mozy's plans only go to around 125GB ($10/mo), and then an overage of $2/mo per 20GB. I've got around 240+GB right now, so that would get pricey in a hurry.
Last edited by Ironhorse; 07-27-2011 at 11:53 PM.
|
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 05:41 PM.
|
|