Daily.co.uk VPS Downtime

Portal Home > Knowledgebase > Industry Announcements > Web Hosting Main Forums > Providers and Network Outages and Updates > Daily.co.uk VPS Downtime

Posted by MichaelGMorgan, 12-26-2011, 03:17 AM
I'm quite frustrated and annoyed about this, but so you know all Daily.co.uk VPS have been down for three days straight now and still no signs of them coming back! They're claiming multiple core system failures.

http://www.dailystatus.co.uk
Posted by Juln, 12-26-2011, 06:30 AM
I just found this thread through Google and thought it would be appropriate to register.

I'm also in the same situation, and this incident has cost me close to £1,000 in business. One of my clients is threatening to leave me if it is not sorted today, other clients are complaining about not being able to access their emails, etc., etc.

I myself have several important system-generated emails which will have been returned as "undeliverable" at this stage; these emails will need to be sent out again and it's about 20 different sources that I will need to chase up to get them. To be honest, most of them are corporate and I don't even know where to start. It's so stressful and frustrating, and all this over the festive period. If I had known it would take this long when it went down I could have re-routed the emails at domain level.

I really don't think this is down to multiple core failures as they say, even if it was, there is no way it would have taken this long to fix, not unless they were using technology from the 1950s.

I am guessing it's something much more serious which they don't want to disclose.

I want to say it's a police raid, because that would make sense as to why this is taking so long. Multiple core failures could just be a cover-up.

In any case they are going to lose a lot of business from this.

Sadly it's going to cost me a hell of a lot more to buy another dedicated server for this part of my business, nevertheless, it needs to be done. I would recommend LCN as they have yet to let me down, and I have been through eUK, Heart, UK2, FastHost, 1and1 and now Daily. What an absolute shamble.
Posted by Flapadar, 12-26-2011, 09:27 AM
They are mentioning array rebuilds.. this is kind of worrying. It suggests that all of their servers were on the same physical server.
Posted by hostbunny, 12-26-2011, 10:38 AM
The downtime is absolutely shocking. I never expected this. I wonder if there will some form of compensation? I hope I can get my data back at least this week. I don't want to host with them any longer.

I will now look at a dedicated server from 1&1. I know a lot of people complain that their support isn't great but I am capable of controlling my own server as long as it is on and it can be connected to.

This Daily situation is borderline of insanity.
Their hosting slogan "Great hosting it's about time" should be taken off their website.

Madness.
Posted by Juln, 12-26-2011, 10:44 AM
I'm absolutely with you hostbunny, but 1&1 are really not much better, I find their control panel overly confusing as well. You might want to consider Heart Internet or LCN. I don't really like Heart, but that's a personal issue not a professional one, from a hosting perspective they actually never let me down and I had a dedicated server with them for over 3 years. I'm with LCN now on my dedicated server and I haven't had any trouble yet, it's only been 6 months though.

Really don't know what I'll do in terms of this VPS, the cost to use another dedicated server is almost 3x what I'm paying to Daily for the VPS.
Posted by hostbunny, 12-26-2011, 11:34 AM
Thanks but I think 1&1 do the only dedicated servers within my price bracket. I will think about it but moving to another VPS does not appeal to me right now because of the issue with Daily.

I really thought they would get things back up and running today but it is now creeping towards the end of the afternoon in the UK and am thinking a Monday fix is now unlikely. They have deleted all the comments on their status page and now posted a single summary. Presumably this makes things look less amateuristic. It's a total mess though and well beyond "inconvenient". Google analytics show the visitors to my sites flatline since Thursday and I am guessing the longer this goes on the worse for Google search rankings this is. They should realise this.
Posted by freethought, 12-26-2011, 11:45 AM

Quote:

Originally Posted by Flapadar

They are mentioning array rebuilds.. this is kind of worrying. It suggests that all of their servers were on the same physical server.

They say that the disk failure occurred in a SAN - i.e. a big shared storage array accessed by multiple servers.

This is the normal way of providing the ability for VPS to failover between different underlying physical servers.
Posted by Juln, 12-26-2011, 11:50 AM
I have everything up to 14:25 yesterday, there was another two updates this morning which I didn't save, but they basically covered the same thing, "we are not knowing what is the happen here because that, so we are ringing the Dell support who is helping us with the telephone, bla bla bla".

Quote:

Update 1425

The data migration is currently in the final stage still. The data (excess of 12 terrabytes) is performing integrity checks. This final 1%-2% of the migration is the longest period out of the whole process due to the integrity checks and subseqent removal of the old data.

We will update again as soon as there is a new development

---

Update 1142

The restore is still taking place currently. The moment we have specific details of exactly what is remaining and all final tasks are complete, it will be posted on this page.

The final 10% of the rebuild/migration took considerably longer than the previous stages, which has pushed the initial schedule out of having this completed in the early hours

---

Update 0851

We are awaiting the next report from our engineers who are finishing off the rebuild/restore. We shall update this page as soon as we have this as we will then be in a better position to advise on the remaining stages.

---

Update: Sunday 25th @ 0741

The last 10% of the rebuild/restore took longer to complete than the previous data. We are currently at 99% so awaiting for the final 1% before instigating the procedures to bring the VPS back online

We will update again by 0900

---

Update 1821

We are in the final phase of the rebuilded array and engineers are already on standby to take over the manual steps when the rebuild/migration has eventually finished.

We do not have an exact timescale still, but will update this page the moment we do.

At the current rate of data restore/transfer, we will be looking at the early hours of the morning, so unless we have new information sooner, we will update this page again no later than 0900 on Sunday 25th

We fully understand the frustration of customers affected by this outage but would stress we are doing everything humanly possible and working around the clock to restore their services.

We have updated the FAQ above based on some further customer questions

---

Update 1233

Migration/Rebuild is still continuing as normal and we are currently looking at late afternoon/early evening.

We will add a progress report by 1900

---

Update Saturday 24th December @ 0856

The Migration/Rebuild has been running throughout the night. Again, no errors have been detected and the process is over 50% completed now. The current estimated completion time is still today, and we are looking at the afternoon.

We apologise for the time this is taking. Unfortunately, rebuilding arrays after events such as this is a long process and the sheer volumes of data mean that the tasks running alongside with data migration is at a similar duration.

We will update this page again by 1300

---

Update 2040 The array rebuild and data migration has not encountered any errors so data looks to be fully intact, however the process is taking longer that expected and we suspect this will run overnight at the present rate.

We apologise for the extended delay. Please be assured we have our sys-admins working overnight to monitor the process and intervene when required.

We will update again this page again by 0900 tomorrow (Saturday), or overnight if there are any further developments.

---

Update 1700

The two tasks of rebuilding the array and copying the data over to a separate system are both still in progress and no errors have been detected so far. Additional discs have been inserted meaning that if the rebuild does encounter any problems, then having clean discs in will prevent a failover automatically instigating

We will continue to monitor and update this page again at 2100

---

Update 1452

The drive array rebuild and data migration are both still in progress now. Our engineers have removed any unnecessary files (e.g. old log files or previous versions of data snapshots) to reduce the data volume as much as possible to make this process as quick as possible.

The current estimate is that the full array rebuild will take several more hours to complete due to the amount of data and complexity of the array. At which point, replacement drives can be fully added to the array with up-to-date and live data that has been copied over to them.

As such, we shall next update this page at 2100 this evening, or sooner if we have any further developments or changes to this schedule.

Please be assured we have brought in our full compliment of engineers along with external support in an effort to restore service as fast as possible.

---

Update 1347

The main hardware fault in question is down to a drive raid array error where several entities encountered unrelated errors in a relatively short space of time.

Currently the drive array is rebuilding following repairs made by the engineers and we have new discs on standby to replace any discs at fault once the array is ready again.

In addition, the engineers are also migrating all the current data off the drives (the data is still recoverable despite the faults) as this can be used to form contingency plans in the event the array rebuild fails or encounters any other problems (i.e. the data can then be put onto a backup system).

Currently the fastest option to restore service is awaiting the rebuild of the current array to complete.

Our engineers will have a better indication of how long this will take within the hour, so we will update again by 1500 or sooner

---

Update 1247

Our engineers working alongside the Dell engineers and have identified some of the hardware aspects causing the faults to the network. They are in the process of bypassing the faults in question in an attempt to restore service.

At this stage it is unknown if this will fully restore service, partially restore service, or if the hardware will need to be replaced altogether (which is possible, however a last resort as this would take the longest).

The engineers will provide an interim update before 1400 as to progress, which will be posted directly on this page

---

Update 1011 The root cause for the loss of service is hardware related as well as network on some of the network controllers. We have multiple failover systems and the discs are all raided, though in this instance we have had an extremely rare occurrence of more than one failure within hours of each other, hence why the issue is not yet resolved.

The Dell Engineers along with our own network controllers and system administrators are either connected remotely or on-site at the Data Centre now, ascertaining the full extent of the issues and formulating the most effective way of restoring service.

The next update will be no later than 1300

---

Update 0930 Our engineers are working on restoring full service still. They are currently speaking to the engineers at Dell for the hardware related aspects of the outage. Once they have full details, we will be in a better place to provide further information.

We shall update again by noon at the latest, or sooner if we have any further information before then.

---

At approx. 2300 on Thursday 22nd December, we experienced a loss of network in our Data Centre to some of our core systems responsible for providing connectivity to our VPS platforms.

Our engineers were alerted to this straight away and have been working on restoring full service overnight.

Currently some ranges of VPS are unavailable (via normal website, ping, RDC, SSH etc.)

We will update this page again by 0930 when we have an interim report from our system administrators, explaining the exact details of the issue and the estimated duration.

We sincerely apologise for any problems this outage will have caused if you are affected.

Posted by hostbunny, 12-26-2011, 04:07 PM
Time seems to run at a different speed at the the Daily headquarters. They posted that at 5 O'Clock they would have the data copied across in about 4 hours. At 8pm they posted they are still on track and be done in 2 hours whilst being 30% of the way there...

Even so, what are the chances of anything being up by 10pm tonight, by 11pm, by 12 midnight? I'd say zilch.
Posted by RONIS, 12-26-2011, 04:43 PM
For critical data, i suggest always use cloud servers, they have some redundancy at least.
Keep us posted, guys !
Posted by freethought, 12-26-2011, 05:19 PM

Quote:

Originally Posted by RONIS

For critical data, i suggest always use cloud servers, they have some redundancy at least.
Keep us posted, guys !

This is exactly the kind of issue that cloud servers are (generally) vulnerable to - SAN failures!
Posted by hostbunny, 12-26-2011, 05:25 PM

Quote:

Originally Posted by freethought

This is exactly the kind of issue that cloud servers are (generally) vulnerable to - SAN failures!

It seems to me that a "shared hardware" outage is more serious than a network outage. Ie if I had a dedicated server it would have been back online by now I would have thought. Correct me if I'm wrong. Think I want my own dedicated server even if I can only afford a crap company one.
Posted by RONIS, 12-26-2011, 05:28 PM
Cloud usually has 2 SAN servers or even more. Also some clouds use distributed file system instead of SAN, but this is not the case.
For me, it looks like they messed up something in their storage server.
I hope that they have a fresh backup, if not, i feel sorry for you, guys.
Posted by freethought, 12-26-2011, 05:30 PM

Quote:

Originally Posted by hostbunny

It seems to me that a "shared hardware" outage is more serious than a network outage. Ie if I had a dedicated server it would have been back online by now I would have thought. Correct me if I'm wrong. Think I want my own dedicated server even if I can only afford a crap company one.

In theory, a well designed and configured SAN from a decent vendor that has a lot of experience in the SAN market (such as Dell, HP, EMC, NetApp etc.) should be very reliable, however it does add complexity (particularly if replication is involved) and means that if it does break it is often more serious and complicated to fix.

The best thing of course is to have the data physically held on two separate servers, preferably in two locations. However, this does mean that you then have the fun of a) keeping them in sync and b) handling the failover.
Posted by freethought, 12-26-2011, 05:39 PM

Quote:

Originally Posted by RONIS

Cloud usually has 2 SAN servers or even more. Also some clouds use distributed file system instead of SAN, but this is not the case.
For me, it looks like they messed up something in their storage server.
I hope that they have a fresh backup, if not, i feel sorry for you, guys.

Actually, according to Daily they have some form of replication in place:

Quote:

Originally Posted by http://www.dailystatus.co.uk/

What we then encountered was the extremely rare event where we experienced both another disc failure on our other live SAN and the replication cluster at the same time (approximately 2300)

It sounds like they are using Dell Equallogic SANs, which support asynchronous replication at scheduled intervals.

This means that the replicated data will be slightly stale - for example if they replicate the changes every quarter of an hour then in theory you can potentially have around 15 minutes of data loss (depending on how long replication takes, which in turn depends on the volume of data changed) in a scenario where you failover to the second SAN.

Synchronous replication solves this but is much more complicated and has a performance impact.

Quote:

Originally Posted by RONIS

Also some clouds use distributed file system instead of SAN, but this is not the case.

You always have to have some form of centralised control in any distributed storage system. For distributed file systems this is normally the metadata and file locking part. Of course you can usually have some form of high availability replication and failover setup, however as with a SAN replication system, this adds complexity and another potential point of failure.

Unfortunately, none of the various systems available in this area is 100% perfect, or everybody would be using it
Posted by RONIS, 12-26-2011, 05:48 PM
For mission critical data, async replication is not an option and I hope, they realize that.
Posted by freethought, 12-26-2011, 05:51 PM

Quote:

Originally Posted by RONIS

For mission critical data, async replication is not an option and I hope, they realize that.

For truly mission critical data, the client should do the appropriate due diligence with their server provider and make their own arrangements where necessary to fill in the gaps.

At least asynchronous replication at the SAN level is better than what you would get from a dedicated server - loss of the RAID array or controller either leads to data loss or at least renders all of the data temporarily inaccessible until the issue is resolved.
Posted by hostbunny, 12-26-2011, 08:53 PM
Well as I suspected no update in the promised time frime, nor has anything changed. The downtime for sites, email and everything is now reaching the 100 hours mark. Astonishing.
Posted by ezbnc, 12-26-2011, 09:20 PM
100+ hours you have got to be kidding me.
Posted by Juln, 12-27-2011, 05:22 AM
I'm telling you, something else entirely is going on. No technical problem would have taken this long to sort out, not with a business that size.

And if this last update is not a sick joke, then they're just trying to buy more time.

Quote:

Update: Tuesday 27th December @ 0747
We are currently requesting a new update from our engineers as to exactly what the current situation is now that we are into the following morning.

We will add this to this page as soon as we have it.

EDIT:
Oh wow, they changed the date too??!

Quote:

First Reported: 15:10:06 on 2011-12-26

Posted by hostbunny, 12-27-2011, 05:41 AM

Quote:

Originally Posted by Juln

I'm telling you, something else entirely is going on. No technical problem would have taken this long to sort out, not with a business that size.

I'm curious knowing how many people are affected. The company seems a lot smaller than I realised; ie. I never thought they could get away with a downtime like this and not that much going around on the internet about it all.

Also, you got to wonder what happens (or what does not happen) if you buy a VPS right now. Surely they should take their whole VPS ordering off line until this is properly fixed.
Posted by Juln, 12-27-2011, 06:22 AM
You can see their size and what they're making through Companies House. I'm curious to look into their accounts myself now.

Company No. 05743110

I agree that it seems odd nobody is mentioning it, although not everybody's reaction is to contact the other parties involved, the majority will likely keep emailing the company and may never even know that there is a whole discussion on the matter elsewhere...

Either way, they've lost at least two customers from all this, and I'm out of pocket over £1,000 and I'm pretty-much begging for my clients forgiveness. I guess I won't be making this mistake again.

Great hosting... my arse.
Posted by freethought, 12-27-2011, 06:29 AM

Quote:

Originally Posted by Juln

You can see their size and what they're making through Companies House. I'm curious to look into their accounts myself now.

Company No. 05743110

Daily Internet Services Ltd reported revenue for the year to March 2011 (latest accounts) of £1,267,978 with a net loss of £228,833
Posted by freethought, 12-27-2011, 06:32 AM

Quote:

Originally Posted by hostbunny

I'm curious knowing how many people are affected. The company seems a lot smaller than I realised; ie. I never thought they could get away with a downtime like this and not that much going around on the internet about it all.

Also, you got to wonder what happens (or what does not happen) if you buy a VPS right now. Surely they should take their whole VPS ordering off line until this is properly fixed.

I guess it depends how many SAN pairs they have serving their VPS platform. If this is just one pair from several that has failed then not all customers will be affected and VPS provisioning will still be working.
Posted by hostbunny, 12-27-2011, 08:46 AM
What a joke.

Quote:

Update: Tuesday 27th December @ 1130
Verify was finished successfully. Remaining rebuild process in progress.

We will update as soon as we have a progress report to see if we have an estimate of remaining time and stages

How many stages are we talking here?

First they were talking about rebuilding the SAN, then copying a partition and now we have to continue with a rebuild again before we can wait on remaining stages to complete. This is the worst experience with hosting I have ever seen and had.
Posted by hostbunny, 12-27-2011, 09:05 AM
My VPS is finally back up and running.
Posted by MichaelGMorgan, 12-27-2011, 09:34 AM
Mine is still down - they have said they're bringing them back up today though.
Posted by tonkatoy, 12-27-2011, 12:34 PM
Ours is one of those that is still down. Their system is clearly not fit for purpose - they must have had a physical drive failure with all the VPS's sitting on that went down too. They are only good at hosting little websites....I think we should all get together and launch a group legal action against them for this. If they had dealt with it in 24 hours at the most, it would have been bad but nearly a week says to me that they should not be providing this service and they have cost us thousands of pounds in lost business.
Posted by MichaelGMorgan, 12-27-2011, 01:23 PM
Mine is also still down, so I guess I must be the other one.
I've opened a support ticket to see if I am one of those two that are still down just to be sure. Will most likely have to wait 6 hours or so for a response if the last couple of days are anything to go by.
Posted by tonkatoy, 12-27-2011, 01:32 PM
I have another server which is still down with them as well. I think they may mean 2 of 'their' hosts = a number of their VPS's.....

I cant bear another day down. The impact of this on our income and reputation makes me feel sick.

Ive just tried to route emails through Google Apps but without access to the server I cant even do that as you have to verify the domain on the server.

I actually feel ill......
Posted by MichaelGMorgan, 12-27-2011, 01:38 PM
During the downtime I've been thinking up some sort of solution for if this ever happens again. I used a couple of EC2 instances for a few priority client sites as soon as I knew my VPS was down. I'm also considering moving email for all domains to either Google or to an EC2 hosted mail server.

But yes, this is stressing me out a lot!
Posted by tonkatoy, 12-27-2011, 01:45 PM
Im going to go with Google based on their spam control and other apps which are quite useful and definitely a new VPS provider. Ive had it with Daily. Just tried to SSH again, Network is unreachable - arrgghhhh!
Posted by MichaelGMorgan, 12-28-2011, 04:20 AM
Yay! My VPS is back online!
The total downtime for me was just over 5 days. Wow!
Posted by Juln, 12-28-2011, 05:04 AM
My VPS was also up yesterday afternoon, I am now in the process of setting up my new dedicated server with LCN, Daily gave me an extra month with the VPS in exchange for the loss of business. - What an amazing trade.

It was almost 5 days for me too; that's the worst downtime and service that I have ever experienced.
Posted by gordonrp, 12-28-2011, 05:28 AM
Where are your offsite backups?
Posted by hostbunny, 12-28-2011, 05:30 AM

Quote:

Originally Posted by Juln

My VPS was also up yesterday afternoon, I am now in the process of setting up my new dedicated server with LCN, Daily gave me an extra month with the VPS in exchange for the loss of business. - What an amazing trade.

Extra month my ass and such a load of ******** that you only get that extra month if you phone up and complain. Not a single email/message about the matter, as if it never happened for everyone else.

Thinking of moving my domain names ass well. Anyone know a good one? 123-reg were awful when I was with them. Thinking off Go Daddy as need to register/hold .nl (Netherlands) as well.
Posted by tonkatoy, 12-28-2011, 08:16 AM
Mine was the last (or one of) to come back online. No apology from Daily, no offer of compensation or discounts!

Just routing my emails through GoogleApps now as a precaution before finding a more suitable host.
Posted by TurboGuru, 12-28-2011, 09:43 AM
Like many of you, my VPS was down for 5 days.

It came back online late yesterday evening and that's when I...

[ Downloaded all my files ]

[ Erased all personal data ]

[ Filed a cancellation from the control panel ]

The cancellation procedure involves sending a CODE to the Daily helpdesk to confirm the cancellation and quote the code.

I sent them a ticket instead as there is no way I am wasting any more of my time waiting on the phone in a que. The cancellation appears to of been accepted.

It's been really difficult trying to find a decent VPS provider. Once I thought I had found one, I would then end up finding some recent bad reviews on WHT. In the end I discovered RackSRV .... VPS was setup within 5 minutes and everything is nice a fast.

Will see how it goes....
Posted by Juln, 12-28-2011, 12:29 PM
I don't know if any of you guys got an email from them, but here's what I got this afternoon:

Quote:

Hi,

This is an update that all the VPS platform was back online from the late evening on Tuesday 27th through to the early hours of Wednesday 28th for the last couple of hosts.

We are now starting the full investigation to find out what exactly caused the outage, a full check of the way it was handled, if there are any future preventative steps that can be put in place etc. and putting this into a post incident report. We are involving all parties involved for this, from our Support agents right up to the third party engineers to ensure no stone is left unturned.

We apologise sincerely for any problems an outage of this nature this will have caused you. Obviously we are an Internet based company ourselves, so are well aware of the affect any extended loss of service causes.

If you are still encountering any problems on your services, please let us now straight away via a Support Ticket from your MyDaily Control Panel (this flags as a VPS query so our team can identify it straight away and includes disclaimers allowing us to log in). We have put a post on dailystatus.co.uk explainign the specifics we required.

Suffice to say the odds of an event happening like this in the first place is generally unheard of, however plans are already afoot to reduce the chances of anything like this occurring again.

With regards to compensation for the outage: The duration was just over five days in total. As such, we would like to extend your VPS expiry period by a full week, thus ensuring you have not paid for the five days and also have some extra days without charge to ensure your VPS is working and can contact us for any issues you may have. This will be added to your VPS over the next week so you will see the new expiry date in your MyDaily Control Panel

If you have any questions, please let us know.

Regards,

Ray Supple

I particularly like:

Quote:

Obviously we are an Internet based company ourselves, so are well aware of the affect any extended loss of service causes.

Let me just make a quick point; assuming that one of my clients is a well established restaurant in central London, this restaurant serves about 100 guest per day, each guest spends at least £50 on a meal, that's £5,000 per day (minimum, I know they make a lot more but let's keep it simple). During the holiday season, particularly over Christmas and New Year, their guest numbers doubles if not triples, so now we have about £75,000 per day...

Obviously not every guest will have been referred by the website and some guests would still have tried to call up despite the website being down, but we can safely assume that over 5 of the most important days in the year, at least 100 people would have seen that the website was down and therefore would not have tried to make a booking (either because they thought it was shut down or could not confirm that it was closed over the holidays).

So that client lost at least £10,000 in business; bearing in mind the post from before with Daily's last year's revenue, I'd be surprised if Daily took this much in over 5 months for VPS, let alone 5 days. They really have NO IDEA of the EFFECT that their downtime has caused, that's for damn sure.

I also like the other quote, like someone was going to stay with them after all that downtime:

Quote:

With regards to compensation for the outage: The duration was just over five days in total. As such, we would like to extend your VPS expiry period by a full week, thus ensuring you have not paid for the five days and also have some extra days without charge to ensure your VPS is working and can contact us for any issues you may have. This will be added to your VPS over the next week so you will see the new expiry date in your MyDaily Control Panel

Posted by TurboGuru, 12-28-2011, 01:09 PM
Wow - I would of thought that you would be given at least a month free...

Well I didn't get that email, and that's probably a good thing as it means my cancellation has gone through.
Posted by dazmanultra, 12-28-2011, 01:25 PM

Quote:

Let me just make a quick point; assuming that one of my clients is a well established restaurant in central London, this restaurant serves about 100 guest per day, each guest spends at least £50 on a meal, that's £5,000 per day (minimum, I know they make a lot more but let's keep it simple). During the holiday season, particularly over Christmas and New Year, their guest numbers doubles if not triples, so now we have about £75,000 per day...

I'm not absolving Daily of any responsibility in this instance, but for a business during an important period, it is important to consider contingency plans and scenarios such that if outages of this type do occur they are mitigated and you have backup plans to bring up alternative hosting solutiosn. That could be as simple as hosting a simple "contact us page" on a small hosting account elsewhere (keeping your TTLs low so as to enable you to re-direct the site).

So how much were you paying for your hosting for a site that deals with or is in some way partly responsible for £75k a day in revenue?

This type of outage is possible with almost any provider - see what happened when The Planet was down for a week when a truck drove in to the side of the datacentre.

I do think in this instance, Daily's offer of compensation/SLA is pitiful and they should be working harder to re-earn everyone's trust.
Posted by Juln, 12-28-2011, 03:09 PM

Quote:

Originally Posted by dazmanultra

I'm not absolving Daily of any responsibility in this instance, but for a business during an important period, it is important to consider contingency plans and scenarios such that if outages of this type do occur they are mitigated and you have backup plans to bring up alternative hosting solutiosn. That could be as simple as hosting a simple "contact us page" on a small hosting account elsewhere (keeping your TTLs low so as to enable you to re-direct the site).

So how much were you paying for your hosting for a site that deals with or is in some way partly responsible for £75k a day in revenue?

This type of outage is possible with almost any provider - see what happened when The Planet was down for a week when a truck drove in to the side of the datacentre.

I do think in this instance, Daily's offer of compensation/SLA is pitiful and they should be working harder to re-earn everyone's trust.

Thanks for the advice dazmanultra, however the website I am referring to was already only a holding page for the restaurant (I'm still developing it), with no more than basic contact information on a fancy background and festive season opening times.

On the other hand, how could I possibly have known or even anticipated that it would go offline for 5 days?

Quoting their website:

Quote:

Serious hosting requires fanatical attention to detail from Daily. We monitor our network and systems 24 hours a day 7 days a week to ensure your systems are running at peak performance.

Hosting should be a hell of a lot more reliable than this if they're going to use selling points like that, and slogans like "great hosting" and so-on.

I think my future will point to Cloud solutions rather than multiple dedicated servers now, I just can't afford it.

To answer your question (ignoring the obvious parody), I was on the VPS Pro package which is about £27/month (+cPanel); quite an attractive package if you ask me, no bandwidth restrictions, 4gigs of RAM... can't really complain at all, I use it for project development and my personal websites/email. Never again.
Posted by MichaelGMorgan, 12-29-2011, 03:46 AM
This morning it seems my VPS is down again. I cannot SSH, pings are failing, all hosted sites and email is down. Anyone else having trouble with their Daily VPS still?