Club Cobra is back online. Another HD down.

[Brent Mills]

Well, it turns out there was another HD failure. This time it was one of the boot disks. Used the spare drive I bought last time around, reinstalled the OS and all of the necessary software to get the site back up and running, and here we are.

Lesson. I think I'm done purchasing 15k rpm HD's. Fast, but they don't seem to be so reliable.

Anyways, it's been a lot of late nights lately, and I hope we're through the storm. Next plan is to buy replacement drives all around to put off some of the troubles we've been seeing.

Sorry for the inconvenience folks.

[Jamo]

Doggone it, and I was starting to enjoy the porn spammers.

Thanks buddy...

[Rick Parker]

Jamo: And the complainers?

[Gerry K]

Drives were probably made in China!!!!!!!!!!!!

Gerry

[Silverback51]

Don't be sorry. Even as a new comer I really appreciate your efforts.

Thank you.

[decooney]

Hey Brent,
Maybe you can do some background on the product-supplier and failure rates of the 15k drives you are using before ruling them out by speed in general. We run 10+ Terabyte sized databases in my line of systems work, approching 50TB over the next few years, all on 15K mirrored drives with low failure rates. They do fail but no more than 10k, and we did find a much higher failure rate on lower grade (cheaper) Tier2 disk subystems vs. Tier1 grade stuff. Good Luck on it, and thanks again for all of the hard work you guys all put into the ClubCobra site. DC

[Brent Mills]

Seagate drives. There's 8 drives stuffed in there, so it could be heat related....though there is good airflow I thought.

[xracerbob]

Brent- Thanks for all your hard work. The silent majority out here really appreciates all the work that goes on in the background.

Bob

[Velox]

Brent - thank you for all your hard work. Club Cobra means alot to a great many people, and your efforts to keep everything going are greatly appreciated

[Ron61]

Thanks Brent for your hard work. There was an article in PC world some time back about hard drives and their failure rates. Some seemed to have high ones and others went for long times, but they will eventually fail. That is why everything in the ESS computers for the telephone companies is backed up with two processors and two complete sets of memory and storage. If one crashes the other comes on line and runs everything until the bad one can be repaired. But that is a very expensive option and one that only major companies would likely use.

Ron

[PDHse]

Good work Brent

[Steve Riley]

As in the past, thank you for all the work that is done to keep this site available to us all!
Steve riley

[decooney]

Originally Posted by Brent Mills

Seagate drives. There's 8 drives stuffed in there, so it could be heat related....though there is good airflow I thought.

B, Are the drives in a seperate storage unit, or part of the main CPU case/cabinet? Also, if this system is at an ISP type data center faciltiy, in a stack-rack cabinet, you might have them flow check the cabinet with the door open and closed to see what sort of airflow you are getting. Sometimes customers would leave the door off the cabinet and air was not being circulated in and out properly. In my prior life at ISPs and former DCs, there were customers dense-packing and stacking various units of different types right on top of each other, so close, with no real air gap in between each unit on some of the older racks, and if the data center chillers were not blowing cool and flowing perfectly in and out of the cabinets, overheating and internal component failure resulted more often. If it is at an ISP, you might ask one of their raised floor techs how many high-temp alarms they get from their chillers, and if other customers are experiencing component failures. It could be their are maxed out on their chillers... lots of things to check if this is at an ISP. PM me if you want to brainstorm some more..

[COBRA36]

Thank you for continuous effort to keep this great site up and running.
My day isn't complete without looking at the CLUB COBRA site
Again thank you. John

[tcrist]

Brent,
Thanks for getting CC back up. I know it is a lot of work taking care of this site and the server. You do one he!! of a job, we all enjoy it and are pretty much lost when it is not running (though that is not very often).

Terry

[FST FOX]

Thanks again for your efforts

[mikiec]

Brent..

Again thanks for your hard work.

Any possibility you can run a raid or mirror configuration?

Mike

[RedBarchetta]

Brent, you da man!

And to think all this time we thought CC just magically appeared out of the nebulous cloud that is the Internet?!? Like anything else in life, it takes REAL WORK to get results.

-Dean

[Brent Mills]

Originally Posted by mikiec

Any possibility you can run a raid or mirror configuration?

Mike

I had the boot drives mirrored, so I don't know why there was a failure, but during boot the second drive just wasn't recognized as a boot drive. Tried fixmbr with no luck....Tried repair and no luck....along with numerous other things. In the end, a new drive went in and reinstalled everything.

Two weeks ago, one of the 6 other drives failed intermittently (raid 5), which just caused the array to fall over and then I would have to reactivate the array, which doesn't work for keeping a site up without babysitting it 24/7. Just bad luck I guess.

[BMEP]

Since you said "fixmbr", I'll assume this is a Windows box. If you have the boot drives mirrored in Windows (vs. using the disk controller), there are a couple little "gotchas". The best thing to do is test it right after you load the OS and get it mirrored (by unplugging the primary drive and trying to boot).

Here's a Microsoft KB article that might help.

http://support.microsoft.com/kb/167045

Also, you need a boot.ini that points to the secondary drive. This can be done with a floppy boot disk.

http://support.microsoft.com/kb/311073/en-us

HTH, Rob