New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
There may very well be something else going on where they'll need to intervene. I didn't try to format your hard drive for obvious reasons; I just looked at what it said since it booted into rescue and made sure the tables were OK. I, however, can't tell you why it "sticks". Go sleep for a few hours and see if it moves.
Ok, thank you for your time. I will waiting Online.net support.
Good luck with it. If you let @willie do a hard test (format HD) he can tell you if you actually have a hardware issue. It wasn't throwing anything at me, and the tables were sane.
WSS, I'm logged in and (nm edited part) don't see obvious probs. My suggestion was going to be reinstall, follow all steps carefully and document results before continuing ticket. I suspect something is wrong with the mdadm config but I'm not clueful enough to understand or want to mess with it.
That's pretty much where I ended up, too.
Who are using onlinet.net will know install new OS is very simple.
Just choise OS (Ubuntu, CentOS ..) and config partitioning (or simple click Next as default, i do that)
And done.
So what i need carefully ?
Just do the steps, write everything down as you do it or take a screencast, then document afterwards exactly what happened ("I clicked the install button and saw a screen showing XYZ...") and post the info here, and into your ticket if it's not clear here what's wrong.
Unfortunately the syslog in rescue mode only shows what happened in rescue mode. But there are some suspicious looking messages in the raid setup part:
Maybe someone here can say whether those are normal.
Basically, md doesn't recognize it.
Being that he's aborted 4096 installations, who knows what the hell they were beforehand, but generally installing FreeBSD with automated UFS or ZFS would take disks out of the pool, should there be anything left that md thinks is/was once it's- hell, it could even be from the LAST rental.
@son87c How much are you paying for this, anyhow?
It's an LT1701.1 which I think is the same thing I have, 19.99 euro/month with E3-1230v3 and 2x 1TB disks advertised. This particular unit appears to have E3-1231v3 and 2x 2TB disks (drool).
Title should be: "LET members ridicule OP for an hour before realizing there might actually be a hardware problem"
That's what I thought.. 2x2 isn't bad.
I didn't see you offering free admin services or constructive advice. Wait- I still don't. Also, lol at software raid issues being hardware related because disks have been formatted.
But I'd hope starting a new installation would wipe all that out. And cripes, they better wipe the disks after a customer cancels a server, before handing it over to the next customer :O
If they're like OVH- oddly, no. I've had complete LVM names and data on mine. So, I make a habit of doing a few DDs from a ramdisk boot just to ensure the bastard is wiped when it's time to go away.
OMG. I asked OVH about that and they told me they definitely completely wipe all disks before the next person gets them. I don't think I had anything really sensitive on my OVH dedis but this makes me want to use LUKS on everything. Wow.
Yeah, I guess dd'ing from ramdisk (not sure how to launch that) or rescue mode would handle it. Shouldn't be the customer's responsibility though. Sigh.
Rescue mode, dd piped to ssh, ez pz
Well, I didn't actually dig for data (it sure seemed to be there), but the LVM names and so forth were intact. I still wouldn't turn over a server with anything I cared about without nuking it first..to ANYONE.
In the case of a vps there's not much you can do. Hmm. I've just kept sensitive stuff entirely off of hosted servers. Not that I have much that I'd want to put on them anyway, but I think of colo-ing a box somewhere someday, and/or running a tiny home server on my crappy DSL connection.
You always have 'dd', but there is a program called 'wipe' that will (slowly) delete everything you allow it to.
With a vps, you can't count on wiping utilities. You don't know what the underlying file system does with rewritten blocks, the host could be doing occasional backups, etc. I have an application with some sensitive info (authentication credentials) and am thinking of running it on a Scaleway dedi keeping the sensitive stuff entirely in RAM. But that's just a few bytes, nothing like a whole disk.
God damn it....
While technically true, I doubt many hosts are going to take too many backups of your $3/mo host. Without having physical hardware, there is no guarantee for your container. Period. Neither for hardware, or otherwise. But, we're just wasting bandwidth now.
I am having the same issues with my 1701.1. The server simply refused to bootup from the HDDs, even though the support people tried to fix (as they did in the screenshot in the OP, same guy Julien) and said it was fixed, it wasnt. Now they asked me to replace the server(as they confirmed it was a hardware issue) but there is no more stock available. So I am still waiting for a resolution. Hoping they will provide something soon.
Had a HDD on LT 1701.1 throw 9 reallocated sectors in a day. Was nice to support (and helpful) and they replaced the hdd with an identical one within two days (note that it was not urgent and started on a saturday)
So, I guess "Online" is more o a serving suggestion for you guys..
If there are a handful of bad sectors- you should be able to see this by browsing the re-installation log. Also, unless it's a completely different setup, you should also have that already marked, so.. it doesn't hang up.
Shit, I remember when 9 reallocations just meant your disk was formatted. Oh god, that was almost 25 years ago.
Srm and sfill do a good enough job to delete stuff.
We got same issue on one of our XC DEALS 1701.2 we just reported it to online.net We have few pages of this servers so no hurry for us.