Snap 4500 Configuration Recommendations?

[allgeier]

Our 4500 failed recently when two drives went down. We have purchased 4 new drives to start over with. (The old 4 drives will be used as paper weights while we decide if we want to pay for data recovery). I don't have admin access to the server (at least not yet) but I have been asked to make recommendations as to what can be done so we don't end up in this situation again. I don't know if two drives failed simultaneously or if 1 drive failed and went unnoticed until the second drive died, because no email notification was configured?

From reading documentation the only two recommendations I have come up with are:

1.) Designate one of the drives as a hot spare.
2.) Configuration email notification.

Questions I have are:

1.) Are there any other safeguards that can be put in place to prevent data loss?
2.) Can email configuration be set up to notify more than one address? (In case one person's spam filter grabs it).
3.) Is there a notification method other than email that can be implemented?

I haven't had to admin the server but from an end user perspective it seems to be a nice system. Any suggestions are appreciated.

Thank you,
-Shawn

[blue68f100]

Before you brick the HD's, you must have one to transfer the OS to the new drives. Snap-Tech can help you with raid recovery.

SNMP traps is what most larger networks use to monitor there equipment. You can send up to 4 emails, that do not req auth.

More than likely you had a drive fail that went un-noticed, then lost a second drive. The guardian OS has a couple of featues to help you backup and aid in recovery during a crash. It is very difficult to kill on of these if you setup SnapShots and diaster recovery.


All RAID Systems MUST BE BACKUP. Weather to another drive (NAS) or to some other source.

[allgeier]

Does it matter which HD I use? I recall someone saying the OS was on drive 1, which was one of the dead ones. I'll have to look into SnapShots and Disaster recovery. Thank you!

[blue68f100]

The OS in on all of the HD's with Guardian Units. You want to use a cold boot to dock the drive. And only put the drives in the corect position. Failure to do so could japordize your data. Make sure you keep track of the drive order and do not mix them up. If you need more info let me know, I will try to help yo out.

Just for info, were these Maxtor drives? And I hope you are installing something else, like interpise drives.

[allgeier]

<sigh> Yes they were Maxtor drives.... and we bought more.... our IT guys told us to buy the same thing.

I can reuse 1 drive to transfer the OS and then once that is completed I can replace that so that all 4 our new, correct? The cold boot means power off?

Much appreciated,
-Shawn

[blue68f100]

Personally I would send the Maxtor back, the odds are one of the 4 will fail in less than 90 days. And install WD Raid editon drives. The WD 2500SE RE drives are design for raid use and will give you increase performace and reliability. + they have a 5 yr warranty. Unfortunately the largest they make is 320gig in RE (EIDE).

Yes, I would put all of the orig drives back in, except 1 that failed, let it be new. Boot the server. Watch the drive leds on the new drive. It takes ~5 min to boot the server, + 1 min for the drive init and os transfer. Once the led of the new drive led has gone out ( will stay on for ~30sec), init a shut down. Remove all of the drives. Put the new drive in slot 1. Load the other drives on the sleds. I like hot swaping the new drives once the server is up. But you can dock them all and start, the results are the same. Then you will have to go into the storage section and setup you raid6 + snapshots space.

If your other units have maxtor drives it time to dump them for WD, before they fail. I have seen Maxtors die with in weeks of each other, so you may not have the time to get replacements.

Are you going to increase the capacity of the unts or install the same size drives? Now's the time to increase.

Good luck