upgrading to larger hard drive. how to clone/ghost drive when host won't help?

[Jorel]

I'm trying to upgrade my primary drive from an 80 gig to a 200 gig. My host says they can't ghost it because they only do it for Windows (I'm running CentOS). I would definitely like to avoid having cPanel installed on the new drive and then copying all old accounts and settings over. That would really suck since this shouldn't be that difficult. I had to have a drive upgraded with EV1 (or was it ServerMatrix?) and they just cloned the drive no problem. I'm using LayeredTech now though. Any ideas for what I can do or what I can suggest to my host? Thanks.

[ntwaddel]

they should be able to clone the drive with dd

[Jorel]

what a great situation i'm in now. without asking, they just installed CentOS on the new 200gig and slaved the 80gig. they say i have 24 hours before they will unmount the 80gig. so my sites are offline and i'm stuck not knowing how to copy the drive. plus, wouldn't it make more sense to make the 80 primary so i could copy its contents to the 200, then make the 200 primary after that? all the instructions i've found using google suggest it that way and it makes more sense because then there isn't near as much site downtime.

i do appreciate the dd advice but more help would be great.

[chirpy]

Now that you have a new system disk, the following thread over at EV1 works wonders (I use it often for clients):
http://forum.ev1servers.net/showthre...threadid=38797

[Jorel]

Now that you have a new system disk, the following thread over at EV1 works wonders (I use it often for clients):
http://forum.ev1servers.net/showthre...threadid=38797

i would use that but the new drive is just plain CentOS without cPanel and that guide assumes you have cPanel installed on the new drive.

[mgsnnetworks]

Ok I am guessing your drive you installed is mounted and has a new partition made for it. Make a new home drive on that partition called home 2 (mkdir home2) then type cd /home
type mv *.* /path to home2. That should work

[Jorel]

Ok I am guessing your drive you installed is mounted and has a new partition made for it. Make a new home drive on that partition called home 2 (mkdir home2) then type cd /home
type mv *.* /path to home2. That should work

the new drive is lacking cPanel. it's a vanilla CentOS installation. so i don't think simply copying the home directory will suffice becuase a lot of essential cPanel elements would be left out.

[fmalekpour]

Hi,
This could be off-topic here but I'm asking it anyway, assume you have two hard drives, one has RH9 and CPanel installed, partitioned and fully functional, second hard drive has CentOS and CPanel. You need to boot the system with first HDD as primary and have the second hard drive mounted as a folder or some how accessible for copy proposes. How will you do that? My problem is when I put both of them on one server there will be two / two /boot two /home etc and system will not boot. I want all partitions of the first HDD be there but can also access partitions of the second hard drive in a folder or mount or ....

Any idea?

[Infopro]

the new drive is lacking cPanel. it's a vanilla CentOS installation. so i don't think simply copying the home directory will suffice becuase a lot of essential cPanel elements would be left out.

Why wouldn't they reload cPanel as well if they setup the new rig for you? If your account pays for it, get it installed. Seems the best way to go here to me. Then just setup the backups config and restore away.

[kris1351]

You can install Cpanel yourself also:

cd /home
wget http://layer1.cpanel.net/latest.sh
sh latest.sh

That will put the base Cpanel on your server, they should have restored the drive with Cpanel and other apps you paid for though. Which DC is it?

[Jorel]

You can install Cpanel yourself also:

cd /home
wget http://layer1.cpanel.net/latest.sh
sh latest.sh

That will put the base Cpanel on your server, they should have restored the drive with Cpanel and other apps you paid for though. Which DC is it?

yes, seemed off to me too. whichever DC layeredtech.com uses

[WCrooks]

Savvis datacenter. While at least one person over at LT (Cheryl in sales) is fantastic...I would gladly adopt her as my personal sales rep for the remainder of my time buying dedicated servers -- I find that they need to make some improvements support wise. If you weren't on a truncated timetable, I'd say find a new dc immediately. So in this case...my suggestion is that you get cpanel installed on the new drive, transfer your accounts immediately, let them do whatever it is they're doing as a result of this poor execution, and search for a new datacenter. Alternatively...you could pull a very tricky stunt and [how much data, btw?*] move all your site files to a reseller account somewhere...and keep them there until you've got your server issues sorted(eg: new dc or whatever) -- technically this would be possible via ssh using cpmove archives.

*I'm guessing since your upgrading your either close to using, or actually using, your 80gb fully?

edit: I just noticed it's been more than 24 hours...so I assume this has either been resolved, or the drive has been unmounted regardless?

[Jorel]

Sorry to not post a follow-up. My sites were down for about 10 hours or so in total. I complained about the fact that they switched the drives without even asking me if that's the route I wanted to go, and without asking if I knew how to copy everything myself. After this long period of downtime, a new support tech answered and switched the drives back for me and offered to do the drive clone during their normal work hours. I agreed to have this take place Thursday morning.

LayeredTech is both good and bad. Buyer beware, their prices are cheap because they don't offer much support and the support they do offer isn't always of the highest quality (they pretty much admit to this). But when I'm running a charity hosting service run off of donations, I couldn't/can't resist the call of a $205/month dual opteron machine with cPanel. In my opinion, the main problem with LayeredTech is not the response the time for a ticket you just opened, but the time it takes for further responses after the first. Most companies have this problem because they stress the importance of the first reply so much, but with LT it's to the extreme. Still, I find their pricing to be irresistible so I don't plan on moving anytime soon.

[WCrooks]

Glad to hear it worked out.

As far as LayeredTech...I agree -- they're good...until you need help. Basically. Thats one thing I miss about RS/EV1 -- the tech support people there are so good I could kiss them -- and if your not happy..you can complain to the CEO!I mean...thats classic. Having the CEO apologise for a support response having it's "ball dropped" initially and leading to a 15 minute response time goes down as my most amusing experience ever. Unfortunately their server offerings and prices leave a lot to be desired.

It's not such a problem with me...because I've submitted maybe 2 support tickets...one was a question -- rather than a problem. But I still think it'd be worth it to pay a bit more for decent support...but your right -- a charity funded hosting property needs to keep costs low...and LT is good for that...if you can avoid situations like this -- and their other shortcomings, LT can be an effective provider. But if it were me...and it took this long for the host to rectify a mistake they purposely made -- I'd be looking for alternatives(when a truncated timetable is imposed by them). I'm always quick to resolve issues involving my clients, and if the mistake is my fault -- I am very prompt in resolving the mistake -- and making up for any difficulty arising...thus, I expect no less from my DC.

Nevertheless...what matters is...things worked out for you -- glad to hear it.

Good Luck in the future, too (pardon the quote from somewhere I don't quite remember)

[Jorel]

The plots thickens.... the drive upgrade was supposed to happen this morning. Here's the ticket update:

"Your server is now online and responding to SSH login requests. The primary drive has refused all hardware and software attempts to replicate the data to a separate drive. In place of putting the server back online a 2nd time without actually accomplishing anything I have upgraded the secondary drive. The only remaining option to upgrade the primary drive in this server would be to do an OS reload on a 200GB drive to replace the 80GB primary with no data replication. Is this acceptable with you?"

No data replication? That doesn't sound very acceptable. I don't know what to do now. This is pathetic.