Tag Info

Hot answers tagged

14

It doesn't power the disks, it just keeps the data in the cache for (in this case) up to 72 hours until you bring the machine back on line. When you power the machine back up it will write the contents of the cache back out to the disks. All it does is protect against a power failure. If (for some reason) the machine loses power without cleanly flushing ...


7

It works like this: Most operating systems have a system call that allows a so-called "synchronous write". This means that during a write operation, if a write has completed then it's guaranteed that it was committed to disk. Synchronous write is therefore non-cached. It blocks the application until it has completed. This kind of operation is obviously ...


7

"I know that SATA drives can operate on a SAS controller" - you may know many popular SAS controllers than can run SATA disks but don't make this assumption, there's nothing in the SAS specs that state they should also run SATA disks, they just share a lot of common features (connectors/cabling etc.). Manufacturers can pick and choose to do as they wish. ...


7

The controller as a performance bottleneck is quite true, and it can represent a single-point-of-failure as well in some architectures. This has been known for quite some time. For a while there were vendor-specific techniques for working around this, but since then the industry as a whole has converged upon something called MPIO, or Multi-Path I/O. With ...


5

To give you a short, precise and correct anwer: no. While I have not personally swapped any of them, I have watched our hardware techs do it dozens of times. It requires the server to be shut down, then the battery can be swapped and the server can be booted again. From the P400 manual:


5

It sounded like your basic description would work. I currently have a Norco 10bay 3U with built in 2x SFF-8087. I've been looking to expand to case similar to the 24bay one you found, as I've already got 2x arrays of 5 drives. The PC is a straightforward Core i3 setup with a Supermicro AOC-USAS2-L8E as the controller. I use Solaris (or you could use FreeBSD ...


5

The Supermicro 846E16-R1200B http://www.supermicro.com/products/chassis/4U/846/SC846E16-R1200.cfm may be a better choice as it has an lsi sas expander on the backplane and will allow you to deploy a 4 port SAS RAID card like an LSI 9260-4i or LSI/3Ware 9750-4i instead of a high port count card that costs a lot more. You can still go with SATA disks if that ...


4

Setup DHCP on the second server. Set it up to handle the same DHCP scope. On the first server, block it from issuing IPs 1-128. On the new server, block it from issuing IPs 129-255. (adjust ranges as needed). That or setup the second server to issue private IPs on a different subnet (you'll need to make sure that all your network equipment works on this ...


4

DNS doesn't facilitate network browsing. NetBIOS facilitates network browsing. Network browsing has never been "reliable" in the sense that the list of computers is always accurate and complete. If you want to have an accurate and complete list of computers while browsing the network then you'll have to implement a WINS server. Personally, I don't ...


4

It essentially doesn't matter where you put the FSMO roles : although it's a good idea to have the PDCe in the office with the most people, since that's the one that handles notifications for password changes, lockouts, etc. With a HQ and a single branch office, you might as well leave all the FSMO roles on the HQ, especially if that's the office with better ...


4

I'm sorry. I'm really really sorry, but if you have the AIR-CAP2602-I.. then it's only suitable for lightweight AP mode, and requires a controller. You have two options.. you can go back to your reseller, chew them out for not spotting that you'd need a controller when you bought it.. and get them to exchange them for AIR-SAP2602-I versions you can go ...


3

Yeah, you're doing it wrong. You replace the failed disk, then you rebuild the array. Of course it's not working now. You're trying to rebuild data onto a bad disk. That's not gonna work. I would also suggest that RAID5 (in this day and age), with 8 disks is a bad idea. Use RAID6, or at the very least, have a hot spare. The disks aren't large, so you ...


3

HP provides support and warranty services for their own branded hardware. The same for Intel. That's it. So if you have an HP server, you use HP components to keep your warranty and support contracts valid. If you stick third party hardware in it, even if it works, you can't get support for it from HP.


3

There are a couple of glaring problems here: Your Domain Controller that you've screen shotted is multi-homed. This is a problem. There is usually not a good reason to do this and there are special consideration to take in regards to DNS registration, etc. Have you followed these? I'm assuming not. Really consider redesigning your network to not multihome ...


3

Looks like the drive is dropping and then reconnecting. This indicates one of three things: Most likely a bad drive, I would start checking the SMART logs and see what you get there. A bad cable/SCSI controller (normally RAID cards) ... if SMART checks out and this continues, swap the cable first then the card. You are doing so much sustained disk I/O that ...


3

It depends on your write patterns. I typically bias my servers towards a 75:25 write:read ratio... But 512MB and 1GB have been good enough to buffer write activity for my applications. You'll be flushing to disk often enough that size may not matter for most apps. Having the extra cache may be useful for situations where you could benefit from allocating ...


3

There is definitely a diminishing-returns thing going on with higher capacities. The read-cache will contain the hottest blocks being accessed, which in many cases is probably the blocks associated with your filesystem metadata. If your entire metadata can fit into the read cache, overall filesystem performance will be noticeably snappier. The size of ...


3

Just create your new DC and demote the old one. You can do either, if you're really set on having a DC with the old name, you can rename a DC, but it's more of a pain than it's worth, in my opinion. Ask yourself what function keeping the old name serves and if it's really worth the effort. In my experience, it's not. (Oh, and since I forgot to mention ...


3

NL-SAS drives have the mechanical characteristics of 'enterprise' SATA drives, with SAS controllers - this is mostly in terms of reliability/endurance/error rates/etc. They're not electronically sata on the controller side of things, with a SAS bridge - they're SAS all the way. But, to answer your question specifically, no you cannot connect a NL-SAS disk ...


3

Think of that BBU cache as adding a similar level of protection to that afforded by a journaled file system. It's there in order to allow transactions, simple writes in this case, to be completed if they are interrupted by a power failure. Once power drops the controller cannot continue to write, as that would result in completely unpredictable results. ...


3

Have you considered to use cheap 1U or 2U servers ? As this kind of server is very common and as you will buy lots of them, you may be able to get them at a very good price. There will be less data on each server, the loss of one will be less important and putting a new machine into the cluster will be quicker. It seems unlikely that one 1Gb/s network ...


3

Having come across the same issue on an identical platform again, with no valuable data, I decided to see what would happen and just pressed "Commit Changes" after being given that warning. It turned out that this does did not lose the array, and added the new disk to the existing raid 1 array, as expected. As I suspected, the error message was just too ...


3

If the program is running as a service then it would keep running; interactive sessions have no bearing on services (other than being able to start/stop them or whatever). If the program is running in the background of the session (like the system tray, or without user interaction entirely), when the session ends, so does the program. Logging off ends the ...


2

If the RAID stripe data is stored on the controller, which it usually is, then you should be able to just move the controller and drives as-is to another server and it should show the logical drive within Windows after the card is detected and installed. To make things easier, make sure the RAID controller doesn't show up in the list of bootable devices in ...


2

The VMWare forums suggest it can be beat into submission, but it is officially not supported, and if you have any additional problems with it VMWare will not support that configuration. Always check that hardware is supported on ESXi before buying it: ESXi Compatibility List


2

The domain controller a user uses to login against has to do with the Site config. If you haven't done anything about Sites, you have one big one. Users and computers will (within some bounds) pick a random DC in that site to login against. If you do have separate Sites declared in AD, the login process will respect Site boundaries; it will only cross to ...


2

Personally, I'd get a new Smart Array 532 and attach the drives to it. The chances of the existing SmartArray having gotten shorted in the dunking event are non-zero and you don't want that further corrupting your data. Best to take the drives as they are, but them in an entirely new enclosure, and hope for the best. The Smart Arrays are pretty clever about ...


2

Based on your application-layer redundancy, I'd make a strong recommendation for GlusterFS. The whole point of Gluster is to create redundancy through lots of cheap storage, even without RAID. Putting together many Norco boxes with individual drives configured properly (with redundancy across servers) is a reasonable way to go. Gluster scales very nicely ...


2

Short answer first: No, Mainboards dont have a standard port for disk controllers. Nor do cases have a standard slot for disks. Long answer: In case your server actually accepts SATA or SAS disks in the slots, you have to check what kind of slots are on the mainboard (PCI, PCI-X, PCI-E). Then you can choose a extension card for your server. You might go ...


2

If you mean an active directory domain controller, there are articles describing how to authenticate using LDAP and extensions to the AD servers. The answer may be more involved that can be easily summed up here, so you might want to google for "fedora authenticate active directory" and find a tutorial on the webbertubes. Make sure you have decent backups ...



Only top voted, non community-wiki answers of a minimum length are eligible