Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


ClubUptime Refugee Thread
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.

ClubUptime Refugee Thread

fanfan Veteran
edited July 2011 in General

Another provider which seemed to be good is getting out of business now. I guess it's time to open the 1st refugee thread here in LET v3.

Here's what I saw earlier:

CU Family,

>

We regret to announce that Club Uptime, LLC will be closing its doors, effective immediately.

>

With growing costs and a substantial number of nonpaying clients, we have been unable to continue our business with SoftLayer.

>

As a result, we will no longer be accepting new signups or renewals for our products or services.

>

We will update this page when backups are able to be made available to clients as we have them transferring to external backup servers right now.

>

Refunds will be issued to all customers who have already paid their invoice for the current month, and will be prorated if paid quarterly, semi-annually or annually. Please be patient as this process may take up to 4-6 weeks to complete.

>

BroHoster Customers please refer to http://www.brohoster.com for applicable instructions.

>

We've appreciated your business. Thank you.

>

-Management

«13

Comments

  • gianggiang Veteran

    I can't even access brohoster.com. (I'm not ClupUpTime nor BroHoster customer)

  • This is the link to the WHT regarding ClubUptime - http://www.webhostingtalk.com/showthread.php?t=1064528

    Now people are questioning whether the victims would be able get the backup and/or refunds as claimed.

  • KuJoeKuJoe Member, Host Rep

    Wow, sad to see so many diving into the Dead Pool lately. Did anybody here have any real knowledge about CU's operations? I'm assuming they managed their own infrastructure but I'm curious as to how after 6 years in business they still managed to go under? That's pretty shocking.

  • iKockaiKocka Member

    We all know that SoftLayer is not a budget provider and it is not possible not to have a loss when selling 512 MB VPS for $18/6 months. You can get loans from banks for some time, however if business won't be doing better, they will eventually cut you off.

  • KuJoeKuJoe Member, Host Rep

    Actually $18/6 months = $3/month and can be EXTREMELY profitable. It sounds like they were leasing their hardware though which is what kills your profit IMO.

  • Does softlayer even offer colo?

  • KuJoeKuJoe Member, Host Rep

    No idea but anything past 2 dedicated servers and it's time for colo IMO.

  • iKockaiKocka Member

    @KuJoe: Even if they were overselling and fit 80 VPSs per node, it would bring only $250 per node monthly. You cannot provide quality support and having servers leased at premium provider such is SoftLayer without making a loss.

    @Boltersdriveer: After the drama about BroHoster, clients started canceling their Minecraft VPSs. Matt needed to have a high number of servers at SoftLayer to get awesome discounts from them and that's where the problems started. A lot servers and less and less clients.

  • KuJoeKuJoe Member, Host Rep

    @iKocka Simple solution: DON'T LEASE. ;)

  • KuJoeKuJoe Member, Host Rep

    And keep in mind, $3/month for a non-oversold server is still a good profit. Maybe it's not as much of a profit as some companies would like but if you're not in it for greed it's a good profit none-the-less.

  • iKockaiKocka Member
    edited July 2011

    I know it can be profitable, but not in case you lease servers from SoftLayer.

  • KuJoeKuJoe Member, Host Rep
    edited July 2011

    Exactly why leasing is always a bad idea for a company that's looking at the long term sustainability. I can't believe how much money CU probably wasted in the 6 years they were in business. Just a shame.

  • If they just started out and were trying to acquire funds, perhaps leasing was the way to go. But 6 years? Wow. They should have already gotten their own hardware and co-located.

  • iKockaiKocka Member

    When Matt started the CU he was only 15 years old. At this age he couldn't have had much capital to start a company with, so leasing was the way to go. After 6 years and 6000 customers (according to Matt's posts) they should really have 1 or 2 racks filled with servers as it is much more profitable on long run. Just another kiddie host.

  • drmikedrmike Member

    There's another recent thread here:

    http://www.lowendtalk.com/questions/14338/clubuptime-dead-pool

    That WHT one already has 30 pages on it.

    Does softlayer even offer colo?

    ev1, which is now a part of softlayer, used to. ev1 used to be considered a budget host when it came to servers. $1 for the first month and all that for their servers. Attracted a lot of abusers that they couldn't/wouldn't deal with.

  • tsctsc Member

    There was another thread on Facepunch about the final demise of BroHoster, their game server division.

    Allegations are that Matt took $93,000 from company accounts and disappeared. Granted, this is as of yet unverified (except for the testimony of a former staff member).

  • It has already been verified. The previous owner has chimed in, and the thread has been closed. If you look throughout the thread, there's a youtube video confirming this.

  • tsctsc Member

    Really? I never found any mention of that in the new thread, which isn't closed at the moment.

    You may be thinking about the first thread from a few months ago when BroHoster was hijacked from its owner, which is closed currently. That incident seems to be separate from this one.

  • I'd like to clarify a few points. Brohoster was never "hijacked" from the previous owner. He sold it to me last year, before it grew to any substantial size. In fact, had I not bought it, they would have closed their doors. The sole purpose I bought the company was to help Jacob since he came to me looking for support since he couldn't keep it running at the time. We managed to drop the prices as much as we did because of our deals at Softlayer ($294/month for Dual 5620 servers with 24GB Ram and Quad 500GB Drives in Raid 10). Once myself and the rest of the staff felt Jacob was no longer doing his job, we fired him. No company was stolen: He had an idea, he sold that idea to us, and he stopped working. Simple as that.

    The second note is that I did NOT take money from the company. I made less than the rest of the staff on a month by month basis. The company itself was running in the red on a month-by-month basis for the past few months. We had attempted to sell the company in exchange for nothing but the cost of all debts being covered. The buyers interested decided to attempt to "default" on the Softlayer payments last moment, which is what lead to the downtime with no notification. Had we known the buyers were going to simply turn around and say "Hey, we won't pay Softlayer. Instead, we'll just tell your clients to pack up and move to us", we wouldn't have let things go on as long as we did and we would have notified the clients days earlier. Also note that we did not sell to them in the end and we told them to shove off.

    I do greatly apologize for the inconveniences caused all around between downtime, the inability to provide backups at first, and all of this happening while I'm in the midst of traveling with little internet access (However I've been working very closely with Brian and our providers in the background to help ensure everything was handled properly). We're doing what we can to get everyone refunded in as timely a manner possible, to get everyone their data, and to make sure there's as smooth a transition as possible to their next provider.

  • KuJoeKuJoe Member, Host Rep

    It sounds like there was some bad blood between CU and BH. On WHT the BH people (Teri and Larry?) are saying you (Matt?) were hands off at the end while the BH folks and Brian picked up the pieces. Of course this is just what they're telling everybody on WHT so who knows what's really happening behind the scenes (except the people leaking internal screenshots of your SL console I guess).

  • @Kujoe: It's not that I've been hands-off, but I have been working in the background. This all fell during a time when I've been traveling with limited internet access. I have worked closely with Brian, although Brian and Larry have done most negotiating throughout this process.

    With regards to the screen shot, that's completely unrelated from unrelated issues (The VP of Networking was causing us a lot of trouble over IP address space).

  • KuJoeKuJoe Member, Host Rep

    Ok... well I wish you luck in your new venture. It just looks like a lot of misinformation and drama from an outsider's view.

  • @clubuptime: As you stated Dual 5620 servers with 24GB Ram and Quad 500GB Drives in Raid 10 cost $294 par month. Even if you oversold server resources twice you would only make $288 inocme per month with your $18/6 months 512 MB offer. Not to mention that there are IP space and support costs. So why the hell were you offering this if it doesn't cover even server cost? It was only matter of time before you run out of money.

  • @iKocka OpenVZ can be oversold far more then that on gear of that quality before it is even noticed, more like a magnitude of 500% is very easy to get away with on those small accounts. Keeping IO over 50MB/sec you can oversell that box quite a bit. That is actually too much box for the array.

  • fanfan Veteran

    @miTgiB What about Xen and KVM? Just wonder how those not-always-oversold platforms perform.

  • Xen on SolusVM cannot be oversold for disk space or memory allocation, but can still be oversold for IO and CPU, and a dual 5620 with only 4 drives is going to get IO overloaded first. There is a balloon driver that allows memory to be oversold in Xen, and as for KVM, CU didn't offer it, did they?

  • iKockaiKocka Member
    edited August 2011

    @miTgiB: Isn't 120+ VPS an overkill for OpenVZ node. As I know machines are not isolated, so basically any abuser can bring entire node down. Also 4 RE4 disks in RAiD-10 can write/read with about 250 MB/s (that's 1-2 MB/s per VPS). Since they oversell and offer burst RAM too, node is likely to run into SWAP at peaks, so I/O speed additionally decreases at that time. Doesn't sound good.

  • @iKocka You use too much logic and not enough real world experience, 4 disk raid10 is going to yield 150-200MB/sec, but ram you can oversell the bagezzus out of. Real world memory usage in OpenVZ, on a node with 32gb of ram, and 32gb sold, has 6gb in use. Unless everyone is running minecraft on the node, you'll never see what you've described.

  • @MitGib: We only used the 5620 servers for our Xen platform as we did dedicated CPU time. We used Xeon x3470 servers with OpenVZ as it helped buffer us mentally with the amount of accounts we could fit. I/O and CPU issues were very rarely an issue on our OpenVZ platforms based on the x3470. The two servers we had on 5620's had issues until we cut the client count down.

    The platform was actually very well managed, but we ran too many sales on OpenVZ to get our name recognized which lead to the downfall of that platform.

    Our Xen platform was doing very, very well on the other hand (minus a few issues as a result of kernel bugs and the Adaptec drivers)

Sign In or Register to comment.