dmc_ss Posted January 3, 2010 Share Posted January 3, 2010 Hello People, Happy New Year to you all, hope you had a good one. Just want to ask for some general thoughts on a RAID issue I'm having with my host. RAID 1 array, two disks blah blah blah. One disk failed so the host replaced it, around the same time they had un-explained power problems (!!) which casued the server to lose power. Since the disk placement, the RAID array has usually been in Verify, often casuing a slight disk performance issue as expected. On the odd occasion when the Verify had finished all seemed well until a reboot was needed. The array went into Verify again. This has happened every time the server has been rebooted. I have also been unable to schedule a chkdsk or use Volume Shadow Copy services since, both reporting disk issues. Nothing significant is recorded in the event logs. The issue now is that the disk they replaced has had bad sectors on it! So they changed it again. They Verify finished and all seemed well, still no chkdsk or VSS though. I rebooted the server and you guessed it, Verify again. So I'm kinda lost now. Is the array damaged, is there file damage, chkdsk\VSS etc. etc. They are telling me now they want to replace the RAID controller and both disks. What are peoples thoughts? Should the array always be verifying? Thanks in advance. 0 Quote Link to comment Share on other sites More sharing options...
niels Posted January 8, 2010 Share Posted January 8, 2010 Real hardware raid controllers don't need reboots. They should function completely transparently, using 'drivers' only to access special features. Many hosters put in cheap raid contollers without really knowing how they work though. My suggestion: have them pull out the new 2nd disk and put it in a new machine with a new raid controller. Once they're sure the new machine works: swap the entire machine. This minimizes your downtime. Especially if they're not entirely sure about the cause. It could also be the backplane to which the disks connect, or a power issue, or cabling, etc. 0 Quote Link to comment Share on other sites More sharing options...
Daniel Posted January 14, 2010 Share Posted January 14, 2010 How are they rebooting it? If they're hard rebooting it then it should take a while to initialize but a graceful reboot shouldn't cause this to happen. 0 Quote Link to comment Share on other sites More sharing options...
rmccny Posted January 15, 2010 Share Posted January 15, 2010 Did they install a real RAID controller, or something more along the lines of a "RAID-lite" that can only do RAID 0 and 1? I have personally had several bad experiences with some of the cheap controllers out there. I now follow a new rule of thumb for controllers, if it doesn't have on-board RAM and battery, keep looking. Anyhow back to your problem... Do you know what controller your server has in it? I am suspecting that there could be some sort of drive firmware incompatibility with the controller. Just a guess at this point. 0 Quote Link to comment Share on other sites More sharing options...
Roger Posted January 15, 2010 Share Posted January 15, 2010 If your data is important enough for you to think about using RAID and you are serious about having the redundancy that RAID 1, 5 (not 0) or whatever provide. You must have quality equipment. Controllers, drives, cables. Otherwise, well you end up with the problems you have now. End result is when you need it the most the expected redundancy fails. 0 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.