Just FYI, your "big machine" really isn't consuming a lot of power over night. The old pentium 4s were actually pretty power hungry despite the fact that you look at 'em and think "small and weak." When your graphics card is cranking, the CPU is running full blast, and all of your memory/multiple drives are being accessed, it'll consume a lot of power but none of those happen during the night when you're just seeding. Set the hard drives to time out when they're not in use, allow whatever low-power CPU settings your system supports, and I bet it'll use substantially less power than the P4 machines will, since power management is actually a lot better on recent machines than the old ones.
One of our clients has a pair of HyperV servers set up at a colo with a bunch of virtual machines being used for various things. We've just setup a 6 TB NAS that we want to use as a backup. If one of the HyperV servers goes down, we want to be able to grab the virtual machines from that server and get it going on the secondary HyperV as a stopgap. My question: Is there a kind of software that will automatically take a snapshot of a HyperV VM and copy it over to the NAS? Thanks.
@Jimmy James - <a class="postlink" href="http://www.equallogic.com/partnerships/default.aspx?id=6467" onclick="window.open(this.href);return false;">http://www.equallogic.com/partnerships/ ... px?id=6467</a>
Backup/storage isn't my strong suit, but to clarify, when you say "snapshots" - are you using the vernacular to refer to an entire copy of your VM at the moment, or are you referring to the function of Hyper-V to create snapshots of your VMs? The latter is not a backup, it's the ability to roll back to a point in time. If you're referring to the former, VSS lets you create backups of running virtual machines with the built-in Windows Server Backup. MS has an article on it: <a class="postlink" href="http://support.microsoft.com/kb/958662" onclick="window.open(this.href);return false;">http://support.microsoft.com/kb/958662</a>
Is there a huge difference between the nVidia GeForce GTX 550 Ti and the GTX 560 Ti in terms of performance? I ask because Ars Technica's DIY guide has the 560, but it's double the price ($250 vs. $125): <a class="postlink" href="http://arstechnica.com/gadgets/guides/2011/03/ars-system-guide-march-2011-edition.ars/3" onclick="window.open(this.href);return false;">http://arstechnica.com/gadgets/guides/2 ... tion.ars/3</a> I'd like to cut down on the cost of this thing if I can, but I still do want to build a reasonably high-end rig. Edit: I see AnandTech's not too hot on it.
One of my clients currently has a hosted mail service and has decided to migrate over to a brand spanking new Exchange 2010 server. My question is this. Is it possible to setup Outlook to use Exchange 2010 and a POP setup concurrently or do I have to setup multiple profiles have the users switch between them? Thanks.
I have a Dell server, running Windows Server 2003, with raid scsi drives and they have both decided to go out. Is there any way to recover the data from them? Apparently it wasn't being backed up and they had everything stored on it. I've tried seating them in different slots but they just spin up, then "die". They are never even recognized by the system. I was quoted $800 for an evaluation and up to $5k for the actual recovery.
That's how you recover data from them. Are you sure it's the drives and not the RAID controller/disk controller/backplane? What kind of RAID? If it's RAID 1 (you said "both" so I assume it's a 2-disk RAID1), you could try plugging one of the drives into another system since both drives should contain the exact same set of data. The drive should be able to be recognized as an independent, non-RAID disk. That's assuming it spins up in the new system. If it spins up in the new system, the data may either be there normally, or it may be corrupted and you can try recovery tools. If it doesn't spin up in the new system, or if it's more than 2 disks and is actually RAID5, or if it's RAID0, you've got a nightmare on your hands and it's time to evaluate how much the data is worth, and if it's worth the $5k, pay it. And for fuck's sake, install a backup. After a lot of tinkering and multiple rebuild attempts, I rescued the data from a CPA who was a customer of mine a few years ago. When I initially told him the typical cost for a recovery if we had to send it out, he balked. I asked him how much the data was worth on the server, and what it would cost him to replace it. He said he had about 15 years worth of customer data on his server, and he couldn't replace it. So there you go, $5,000 is small potatoes.
No idea what the RAID is. I'm not that familiar with servers. Right now it's at a local shop and they are trying to recover the data but are having the same luck that I did. And there was a backup installed. It was supposed to be monitored by the office manager and he apparently failed to keep a check on it.
Seriously. Working for an MSP, I see multiple different types of backup solutions. My favorite, while expensive, is our clients who have our BDR solution. If a server completely shits the bed, we can virtualize the entire server right on the BDR unit. Complete downtime is in the range of 1-2 hours, max. From there, once we get a new box, we can then do a bare metal restore and have the server back and operational on its own hardware in another few hours. We've had clients that have hundreds of thousands of dollars at stake that have had their asses saved because of this. PM me if you want more details.
You need to pull those drives out of that server and use another server to try and recover it. I have no idea if your local shop is competent or not, but if they're not doing that, they need to be trying that. If it's a pair of drives, it's probably a RAID1 so each drive should have a full copy of the data. An unmonitored backup is no backup at all... so, yeah. If the manager won't check on it, then set it up so he gets an email if it fails.
Drive recovery is one of those things where you can't do a whole lot without specialized equipment. Sorry. If they've plugged them into a known good backplane/server and still can't get either of them to spin up, it's time to ship them off.
Backups aren't worth shit until a successful test recovery is done. Regular test-recoveries using those backups have to happen, or you don't have backups. I've seen corporations spend tens of thousands of dollars setting up backups, as in stupidly complicated backups in Oracle cluster environments, that aren't worth shit because they were never tested. Once shit hit the fan, and they attempted to do the recovery, THEN they realized that the backups weren't complete, and didn't provide everything that was required. So, after two years of running backups, they finally realized that they lost all of their data. I've seen a company die because of such data loss. Morons.
I have a win/heur32 virus on my computer that has been absolute bitch to get rid of. As far as I can tell the virus hasn't done any significant damage yet, but I keep getting messages that tab away from what I'm doing to alert me to the threat and ask me if I want to quarantine it, which does nothing as it will pop up again within a couple hours. Google searches have offered some really bad and/or useless advice, but I've read that the quarantine notification may be coming from the virus itself. I have an Asus with Windows 7 and I use AVG, if that helps. I've also read that this is a bug with AVG, but I'm not so sure because this is the only time AVG has frozen my computer and I have a 6 core processor. Any advice would be very much appreciated.
I have been doing that with the other two servers. I restore a few files, to a test directory, once a week. Like I said earlier, someone else was supposed to be monitoring the third server. Hell... I just got out of him, yesterday afternoon, that an orange light had been blinking, on one of the hard drives, for a few months. Guess that would explain why both shit the bed at one time. And also, that they had taken the backup device off of the server and attached it to a desktop... which is why it wasn't backing up. I may or may not have mentioned it before but I have zero education in networking/servers/etc. What little I know, I have learned on the job. I was originally hired as an engineering technician. I got an extra title once they figured out that I am able to solve most of our computer problems. I guess, in their minds, computer repair = IT guy.
Just picked up a new QNAP 8-bay storage device last week, and have to say it rocks in a big way. It's not stupidly expensive or anything for home use ($1,200 empty), and it does a lot of very cool things out of the box. I figured it was time to actually invest a bit into secure, consolidated storage rather than just fuck around with a bunch of internal and external drives. I've not got all my torrents, music, etc., stored on this networked (and iSCSI capable) box. I've got 4x3TB drives in it in a RAID 5, and it's an awesome remote storage that supports my Windows (SMB), Mac (AFP), and streaming media devices. It can also act as it's own web server, or webdav device. I was originally looking at getting one of those Drobo boxes instead, but have heard too many horror stories of really, really reduced throughput once they get some data onto them, so went with the QNAP instead. Glad I did. It's administrated through an on-board web interface, and doesn't require any specific OS-based admin tools. It also has the ability to email messages directly to you if/when shit happens that has to be looked into. After a week of thrashing on it from a mixed Linux, OSX, Windows, media player environment, I can say it's highly recommended.
I just want to say +1 on QNAP in general. We've been using them as NAS devices for our clients and they've been top notch. They recently updated their firmware to allow you to do LUN backups, which is awesome since this allows for offsite backups with an external drive or whatever.