Given that this is a public plan now I feel like I can share that during the acquisition process of Slicehost we debated heavily what the right thing was to do with the brand. It was honestly a very very tough call and at the time the thing we all (Matt, me and Rackspace) were most sensitive to was to not disrupt the Slicehost community. We succeeded in that goal but the net effect (unintentionally so) was to disconnect Slicehost from Rackspace's cloud roadmap and effectively strand the product.
Unfortunately I think in hindsight I can say that our decision was probably the wrong call. I honestly think everyone would have been better off if we just ripped the band-aid early and tried to figure out how to make the core Cloud Server community out of the incoming Slicehost customers.
Control panels, features, pricing, etc all change and in the long run are only ephemerally important. What matters is having your interests be in the direct path for the company that is serving you and as a Rackspace Cloud customer you'll have that, while as a Slicehost customer you did not (post acquisition).
I definitely can see this from all sides and for Slicehost customers, all I can say is that at no point along the way did anyone (at Slicehost or Rackspace) ever have less than the best intentions for the brand and the community. There are some serious champions of Slicehost at Rackspace both in the form of former Slicehost employees still there (Paul Tomes, article site creator) and long time Rackers. Trust me that these guys are really working for the best for you all.
You mention that "...control panels...are only ephemerally important..."
Many in this thread disagree. You can see it getting brought up over and over as a reason we loved Slicehost, and a reason we don't like other alternatives as much.
I won't get too deep into UX theory here, so I'll just say this: people who make apps: if you hear the same comments over and over from your customers, they might be on to something interesting.
I truly believe in the expression "listen to what your users do, not what they say." So i'm not gonna pretend every lil gripe we have about this is right. But there is definitely a pattern in this thread that Rackspace can learn from.
[update: so far i've counted 18 mentions of the phrase "control panel" in this thread. it, and all user experience, matters a lot.]
The core issue is that you're going to have your needs met if you are on the critical path, especially for a company growing at such a fast pace that has to deal with the reality of making choices on where to spend scarce resources.
By your own logic, having more people be squeaky wheels for the Rackspace control panel should be a good thing. It will be even more pressure for that to be addressed.
A common theme in this thread is that people inside Rackspace see Rackspace Cloud + Slicehost as very similar, because 1) they have a common corp ownership 2) they have an increasingly common (but hidden-to-users) back-end 3) they have common staff in some cases.
But the customers in this thread see Linode + Slicehost as much more similar because 1) their control panels are a bit more similar 2) they share DNS flexibility + other featuers 3) Price (Linode moving in more attractive direction for some use-cases).
Neither interpretation is necessarily right, but in this era it's definitely time to rethink what customer needs are (if you're outlining a migration path). I don't think RS is wrong for discontinuing Slicehost (I'd definitely wanna EOL older products), I just think they're wrong in how they parse features, experiences, and what user needs are.
I see since I posted earlier that user:polvi (hiya!) says the Cloudkick posse is working on the control panel for RSC. I've no doubt that all providers will greatly up their UX game (something something Dave McClure design dogma :D).
It just feels disingenuous for RS peeps to say over and over "Slicehost services are not dying, it's just the brand that's going away" when real features and real UX are going away, at least for a window of time. And it's totally normal for users to go where those features and UX can be found, even if it's a separate company. To users, that feels easier.
UX at rackspace now is ick at best, but since the Polvi virus is currently spreading in the place, I'd bet on significant improvement.
I just see that there are two independent issues. One is that Rackspace UX needs help, the other is that there are dual/competing brands, one of which is feature stranded. Clearly they are related when you consider collapsing the brand into one, but I still think it's worth thinking about them in terms of two distinct issues.
1. The control panel experience was incredibly bad. Incredibly. Large chunk of features simply didn't work and consistently gave errors (I opened numerous tickets, some were claimed to be fixed but later the errors came back). It's unacceptably slow, to the order of 2-5 seconds of loading anything. And the overall flow and experience is just bad (it took far too many clicks to do something like change a DNS entry).
For a while my startup persisted with using Rackspace just because we already had some infrastructure on it. Once an opportunity to move off came along, I jumped on it (it's now hosted on DotCloud).
2. I had some bizarre experience with getting questioned on the phone while signing up for a new account (setting up a personal account to toy around with vs my other business account). I was questioned about all sorts of things like what I will be using it for, what I'm working on, various private details that I was unwilling to give out. As long as I'm not doing anything illegal or against your ToS, it's none of Rackspace's business what I'm going to run on my VPS.
Now I use Linode for all my personal VPS needs and I love it. The control panel is super snappy, the user experience is simple and efficient, and they don't ask me questions they shouldn't be asking. Also when I report a problem, they actually fix it and are very open and transparent about everything.
Since then, I've recommended Linode to at least a dozen people specifically against Rackspace. Yes, I may be a squeaky wheel, but I believe fixing the control panel (and customer privacy policies) should be Rackspace's #1 priority, especially given all the other feedback you're getting inline with this. Each person who posted here could represent a dozen customers down the line.
What's the draw for using CP on Linode? I'm completely happy with the Linode control panel experience.
AWS I don't have to worry about because I use DotCloud which handles all my network configuration and instance maintenance.
I was suggesting it initially as a way to not suffer any pain with the existing Rackspace Cloud CP.
or, people will just leave because they don't want to deal with having to downgrade to a smaller featureset and worse CP at a higher cost, and then have to wait around for it to perhaps get better. i'm already preparing my slices for migration to linode. sorry. been with slicehost for years and was perfectly happy up until this.
I think you'd be hard pressed to find someone at Rackspace who will tell you their existing CP is great. Definitely not a matter of if, but when on the improvement there.
not cheaper at all when converting slicehost to linode or some of the other competitors.
Tell me why I should stay on slicehost instead of moving to linode or somewhere else. How does this move benefit me? In what ways does this move make your product better than linode's?
>disconnect Slicehost from Rackspace's cloud roadmap
What is Rackspace's cloud roadmap? To be direct: I don't care, at all, about rackspace's "cloud roadmap" and it sounds like we're all going to be eating the cost of either a poor engineering decision decision on slicehost's part, or a poor business decision on rackspace's part.
As somebody that's been a loyal customer of yours for about a year and a half, and has recommended your service to countless people, that kindof hurts, man.
Maybe it's just the phrasing, but what it sounds like you're effectively saying is "We're shuting slicehost down. Please please please pretty please sign up for rackspace instead because that would be good for us."
(I'm sorry, this comment comes across as a bit rude. I think I might just be a little unclear how this will affect me.)
I wouldn't want conduct a point for point comparison of the cloud server offering to Linode. I also don't want to discourage you from trying out Linode. It's clear those guys have a solid product and a great following and if you think you fit better over there, give it a try. Rackspace wants to have the most engaged and happy customer base it possibly can, it is totally counter productive to try and 'convince' you to stay if you really are better off elsewhere. There are a handful of providers I truly would trust to use myself, and I would absolutely put Linode on that list.
Having said that, here are some things I think Rackspace can offer that Linode currently does not. I could very well be wrong about any of these, I'm not intimate with Linode's current service. In addition you may not be interested in any of these features.
- Cloud Files
- Cloud LB
- Hosted mail options
- Hosted calendar
- Windows Cloud Server instances
- Backup/image integration with Cloud Files (vps backups go into your cloud files account and can be downloaded as well as persisted when an instance is deleted)
- Managed Services
- Desktop backup and virtual drives to the cloud (jungle disk)
- Platform/application hosting
- All of the above on a single bill from a single provider and with private networking in between for free (for colocated services in the same DC)
As a slicehost customer you couldn't use any of those products, as a cloud customer you can. In addition you will have access to all of the new cloud products in development at Rackspace (whatever those may be, but use your imagination and I'm sure you can guess on likely candidates) as well as all the datacenter locations where Rackspace has active and growing footprints. To put it simply, you'll be switching into an active and growing product suite and I do believe that benefits you.
I personally would not consider it 'shutting down' Slicehost, but instead a merging of Slicehost into Cloud. It is true that the backend mechanics of the two products are essentially the same, and that's why you should not overly fear losing core functionality.
I don't think your comment comes across as rude at all and I completely understand your sentiment. All I can say (I and I think I'm repetitive to do so here again), is that I think Rackspace is doing the right thing here. From my perspective I'm breathing a sigh of relief because at least I know that all the still existing customers will finally start getting some TLC (via being on the actively developed product again).
I'm certain Rackspace wants the transition to be as painless as it can be and I'd expect quite a bit more detail to come out over the course of the year as this unfolds.
- Rackspace Cloud will offer the same technical capabilities,
- the transition will be as easy as Rackspace can make it,
- migrating to Rackspace will give users access to new features (Cloud LB/Files, Windows instances and so on...),
- Pricing will end up being close to lower for a good chunk of the customers,
Only thing is that, despite the fact that I spend a fair amount of time setting up my clients on various cloud services, I favored (and actively advertised for) Slicehost for some use cases because of two main features;
- Price; I could save a few bucks by setting up my AWS instance, but I like the simplicity of setting up your instance and be done with it, with a fixed price that you expect paying for at the end of the month.
- UX and UI; it just worked. Damn simple and straightforward.
- Feature set; only what I needed, no crazy catalog of 20 different ways of putting things together.
I remember signing up for my first instance about 18 months ago; I was convinced by the straightforward and simple website and the pitch line on the front page: "Sick of oversold, underperforming, ancient hosting companies. We took matters into our own hands. We built a hosting company for people who know their stuff. Give us a box, give us bandwidth, give us performance and we get to work.". Indeed, Slicehost would set up my box, give me access to it and then get out of my way. By experience, migrating on Rackspace Cloud will not provide me with the same value, and IMHO that was the core value of this product. That's a deal breaker in my book.
I also stated that you might not care about any of the other cloud products or extra bells and whistles. The fundamental building block (a cloud server) that Rackspace sells will be and is identical to a Slicehost slice, you can just ignore everything else and pay the lower (likely) rate.
I can't deny that simplicity is gone from the equation given all the extra products/services, but the core Cloud Server is the same simple VPS you once knew and loved.
I've had a SliceHost account since 2008, and a RackSpace account since 2009. I've run servers on both platforms, I've taken a look at the different features and I decided to choose SliceHost.
RackSpace might indeed be cheaper and offer more features, but I chose SliceHost because it is quick, easy to use and gets the job done.
One of the things that sold me on using SliceHost in the first place was your non-SLA SLA, http://www.slicehost.com/questions/#sla
SliceHost has a lot of http://en.wikipedia.org/wiki/Whuffie, the RackSpace site has a lot of http://en.wikipedia.org/wiki/Astroturfing
Now, I freely admit that Rackspace may just not be terribly worried about small-time single VPS accounts (like mine). I don't have access to their customer list, and perhaps the number of customers in situations similar to mine is a much smaller percentage than I think it is (so perhaps I am the one who is clueless about Slicehost's customers). Maybe those of us in that category simply make the most amount of noise about this, and at the same time mean the least to Rackspace's bottom line. I'm not saying that is the case, I'm just saying that I don't know, but it could be a possibility. Mark's email certainly seemed to be targeted at corporate business types, who make decisions on where to spend technical money, but actually know very little about how technology works. That email contained the exact kind of jargon-filled spin that rubs Slicehost customers like me the wrong way. It certainly leads me to question whether Rackspace has just outgrown customers like me and is not particularly concerned if many of us switch providers.
You said "Unfortunately I think in hindsight I can say that our decision was probably the wrong call.". Well, know this -- I was a fairly new customer to Slicehost when the news came that Rackspace had bought Slicehost out. I was pretty disappointed, because I specifically did not want to use Rackspace based on the experience of others I knew who had used them. If Slicehost had not led me to believe that it was going to continue to exist as it was after the acquisition, I would not have remained a customer. Like many others, I was a bit skeptical, but was impressed enough by Slicehost that I wanted to stick around. And I've been 100% satisfied from then till now with Slicehost's service. Judging from the response I've seen here and elsewhere, I think a lot of other people would have switched away from Slicehost right after the acquisition, had you not made the call you did make then. So maybe you are too quick to question the decision. It could have meant a lot less revenue for the company since the acquisition.
I know that the Slicehost folks made a really big deal about how great they thought Rackspace was, and how the deal was really good for both companies, and that Rackspace was an equally great company. When you say "Trust me that these guys are really working for the best for you all.", I have to say that I'm at best much more skeptical than I was immediately after the acquisition.
I'm definitely going to be shopping around.
The title got edited. The original title was something to the effect of "Rackspace kills slicehost... ", which the author defended as a 'killing of the brand'. The current title is not misleading at all.
I have a ton of apps running on my old slices. A client pays monthly for the one slice and it really doesn't look like it will cover costs with the shift to their 'cloud' servers. I really don't want to have to go and reconfigure several versions of python and django and several apps. I don't want to migrate databases. I don't want to move various processes that have been running without being touched for 3-5 years.
The email doesn't give us any indication of the migration path. I don't care about your IPv6 problems or your OpenStack announcements.
Even something like domains hasn't been addressed. I use the slicehost interface to manage a dozen or so domains and their subdomains. Am I going to have to shift nameservers and reconfigure these domains?
Read their initial announcement to gain some perspective on how the acquisition was described.
"Our number one goal is to continue delivering the Slicehost experience. In fact, we hope to make it even better." http://www.slicehost.com/rackspace
It failed to say the only thing I wanted to hear: "this migration will happen automatically and you will not need to do anything".
Having failed to do that, it then failed to give me any resolution at all, just leaving a cloud of doubt and ominous phrasing about months of "dialogue". The only reason you need months of dialogue with your customer is if you are about to piss them off enormously.
I was pretty much already planning to migrate to Linode for other reasons (more RAM on the smallest server is huge), but this gives me a good reason to do it right now: months of dialogue sounds more painful than an hour or two re-setting up my server.
If you're worried about costs, Linode's price is half Slicehost's. That should make the migration time worth it.
Lesser features -- Slicehost has several kernel versions available. Rackspace Cloud does not. Slicehost control panel's UX is just much better.
In spite of all this, I've been moving as many of my customer's servers over to Rackspace for a while now. Most of them don't need the extra bandwidth you get for the slice, making Rackspace Cloud much cheaper when you add in backup images.
IPv6 is a big deal though. Everyone's going to have to touch their long-running servers sometime before the end of this year. With IPv6 Day closing in this June and the eventual migration of residential broadband to IPv6, this is one of those inevitable things that isn't going to magically go away.
I think you're confused about the implications of the IPv6 transition -- IPv4 is a subset of that address space, so existing servers will "just work" via Carrier-Grade NAT that ISPs are providing.
Some tiny percentage of the Internet is IPv6 reachable today, and it ain't all gonna change this year. The IPv4 space can't turn into a parallel ghetto Internet after all...
Dual-stack isn't the answer when you've got no IPv4 addresses left to assign.
Edit: To be fair to Slicehost, they provided me with wonderful service for over a year and I never had a single issue while I was with them. I'm sad to see that they'll be shutting down.
i.e. "Moving from slicehost? Provide your credentials and we'll bring everything over automatically. Just click this button!"
Now I've set up a Linode account and want to migrate my config from Slicehost, and it looks like this might be a useful opportunity to test out that tool. But for the life of me I can't remember it's name.
Does anyone have a link?
Also laughing hard at the irony of the title.
I have no idea why they don't do it, they'd have a billion more customers in a few days. I know I wanted to switch off Slicehost for years, their price is DOUBLE linode's and the performance is worse.
I ran some quick checks on Rackpace a few months ago and with the little data I gathered Linode still seemed quite more performant in synthetic and application benchmarks.
Maybe it will just work out...
Anyone know if linode's irc channel is equally staffed by admins/support?
I have to deal with tech companies all the time. I hate most of them. A lot. They always seem to blow it on customer service, core technology, or pricing.
But not Linode. The prices are really good, their technology is really really good (and they're constantly working on improving it), and their support...? Well, I've submitted a few support tickets (once was even along the lines of, "My backup server is acting strange, the disk isn't full and it can write to existing files but it can't create any new ones. I've been staring at it for about an hour now and I have no clue what's going on.") I've never had a response take more than 7 minutes. They've always been fast, friendly, and knowledgeable.
I really can't say enough good things about them.
They've occasionally had outages, including a couple of ugly ones in Fremont (not directly their fault). How many companies can you think of that are so awesome that, when that happened, it still doesn't tarnish their reputation in your mind?
One of my servers went from using only a little bandwidth to using 25Mb/s almost overnight (customer's website got featured on a radio show). The server never stumbled; I never got a notice along the lines of "hey, you're using a lot of bandwidth there, buddy", I never got throttled, it just worked.
If another VPS provider showed up tomorrow and offered almost everything Linode's offering, feature-wise, but for half the price, I'd stick with Linode.
I could bore you to death with crap like this for probably an hour. All it comes down to is, Linode is one of the few Really Great Companies out there right now, doing business with them is a joy, and I'd love to see them just completely own the VPS market.
edit: oh, and I've been with them since October 2009, which has given them plenty of time to annoy me. They haven't, not even once.
In the end I stayed with them because the inertia was big enough that I was lazy to migrate from them. They certainly did no wrong to reject my request. But I shall remember having to spend the last few hours of the day before evening frowning on the train and working immediately after checking in at the hotel. I could have been made a more loyal customer.
This event is enough to push me over the edge. I shall look into migrating to Linode this weekend. (I have already given them a go several times the past year).
For this reason I am using 32 bit kernels with linode for small instances.
If they could automate this, or just go all out on hourly billing, then my current dilemma over which infrastructure to move our slightly tired dedicated app servers onto would be instantly solved.
I really want to go completely linode, but I also really want to be able to instantly spin up loads of staging/testing environments without it becoming expensive.
Unfortunately I have a feeling that the thing that makes linodes reliable, cheap and fast is the same thing that means you cant spin them up and down on demand.
I guess I could just rsync my entire linode environment onto a blank ec2 machine, but its kind of a 'fake' staging environment if its on a different network.
Anyone got any ideas?
So I don't know if that quite meets your needs (since it's just a credit and not a full refund), although the other post suggests a refund is possible if you ask for one.
I respect the desire to EOL an old product line holding up dev schedules, but this email has a lot of things which scare me (multi month dialogue about migration? Translation: "hit a button to reboot server" is not in the cards.) It also doesn't identify any benefit for me to stay: Rackspace's dev problems are not my problems because, prior to this email, my servers worked.
If you have any questions, concerns, or would like to discuss this with anyone at Rackspace - please do not hesitate to reach out directly to me: email@example.com
The biggest reason I love linode now is the control panel. Slicehost also had a good control panel, but the Rackspace Cloud CP is severely lacking. It's extremely slow and the DNS is in a really awkward spot (click a server and then click DNS to see a global DNS...it doesn't make sense).
I was a fairly happy Rackspace Cloud customer but Linode is just a great experience. What is Rackspace doing to improve the cloud offering?
That being said, I've raised the issue of the location of the DNS manager many times and have been told that it is by design and won't change. Go to the Linode or Slicehost CP and a link to the DNS manager is on the dashboard, where it should be. I just don't understand the reasoning for hiding it behind a specific server instance despite it being a globally applicable system. This is probably more annoying to me because of the slow speed of the control panel, meaning it takes a long time to get to the DNS manager.
Best of luck with the improvements!
If the major change is going to be a new control panel, and a different way of accessing support, then just say so.
I recommended them to several friends and colleagues, but from what I've seen of Rackspaces offering I wouldn't be able to do that with them.
RIP Slicehost :(
Also the fact that I haven't received the email myself is not reassuring at all.
I am not going to just sit around and remain without a predictable migration path, be it a migration to Rackspace or elsewhere.
I just have a little hobby server running on Slicehost. I feel sorry for people who are running businesses there. :/
Edit: Just to note. The only issue I've had with rackspace was the control panel. The support was excellent. The people working there are excellent. I want to use their product. The control panel not working just made me feel dependent on support for everything I'd rather manage myself.
That's all of the slicehost product that's left for now. If that's not killing the product, I'm not sure what is.
For me, this isn't a huge problem. But, if I was running a email-based service, or managed VPN connections to/from clients from my slices, I'd be seething.
My biggest concern is that the Rackspace Cloud product doesn't seem as polished as Slicehost. Slicehost was a great service for newer users, with great documentation and help resources.
You are free to disagree with my title but don't accuse me of things I do not do.
I'm hoping someone with Rackspace Cloud experience can pipe in, but at a glance, it looks like a more bare-metal offering, like EC2, where you have to worry about more of the low-level concerns. Thoughts?
So, the real point is that even if this wasn't apparent to you the customer, the substantial differences between Cloud Servers and Slices was branding and different pricing models; the support teams were also separate (and the two front-line support teams have different "styles"). Otherwise, behind the scenes, they're literally the same thing -- in terms of data centers, hardware platforms and hypervisors, etc., etc.
The email left me concerned of the transition and what it would mean for uptime of websites, but if it's literally just a change of billing, branding, and control panel, why are they waiting months and months to change this? If it doesn't affect my websites' uptimes or IPs, they should just rip off the bandaid now.
That said, my needs are fairly simple. The most complex thing I've had to do is provision additional IPv4 addresses on one of the servers (for SSL hosting), and a support person did that for me in under 10 minutes via their web-based chat support.
Unless you have a site that will actually use the gobs of bandwidth that Slicehost or Linode includes (and I don't), then Rackspace Cloud will end up being cheaper.. My single-IP servers cost me between $11 and $15 per month on average, based on bandwidth used.
The chat was awesome. You could jump in there at pretty much any time and start talking to somebody who could help you immediately.
Have really enjoyed the uptime, stability, and performance Slicehost has given us over the last few years. I've got our web site configured to be brought up on AWS in a flash if needed--but we've never needed...uptime of 99.99% in the last year, 100% in the last 6 months. And crazy affordable. Their support/incidence response rocks, too.
Vaya con Dios, Slicehost, I loved you much.
Here is my take: The longer Rackspace waits on transition details, the more customers will leave. I will if I don't see a concrete plan of migration, and I realize that these things are not easy.
I must say, Slicehost articles are great though. They are exceptionally well written. Not that Linode's articles are bad, but Slicehost articles are really good guides.
You do not make a compelling case for continuing with you through the transition.
Not very happy with this whole thing.
Edit: 1.5 cents is actually $10.8 a month. Nevermind :\.