The question often comes up as to why RAID 5 is so dramatically warned against and considered deprecated for use with traditional Winchester hard drives (aka spinning rust) and yet is often recommended for use with more modern SSDs (solid state drives.) Like with anything, it is actually a combination of many factors that come together to make one use case so bad and the other so generally good. Here is the run down:

• UREs are the primary risk factor for traditional hard drives in RAID 5 arrays, but UREs are not a risk (so far) on SSDs. So this one fact alone completely changes the "risk game" between Winchester drives and SSDs
• Time to Resilver is hugely reduced with traditional arrays often taking days or even weeks to resilver. The move to SSDs often cuts that to a small fraction of the original time. That resilver time is not just a performance impact to the environment, in some cases actually making the array useless until it has completely, but also means that the array is completely at risk of secondary drive failure during that window. Reducing that window greatly reduces that risk.
• Resilver Impact is much reduced as SSDs handle non-sequential data access so well meaning that even during a typical resilver an all SSD RAID 5 array may continue to function extremely well while still performing a high speed resilver operation meaning that the risk of performance impact to the environment is much smaller.
• Parity Resilver secondary drive failure risk does not exist. This is a rather sizeable risk to Winchester drives. The parity resilver operation often induces other drives to fail during the resilver operation due to the large strain placed on them for the operation. This does not impact SSDs making the resilver operation far safer.
• Performance is very different between Winchester drives and SSDs. The move from Winchester drives to SSD is a many orders of magnitude jump in performance. The write performance difference between RAID 5 and its key competitor, RAID 10, is small by comparison. So while the latency impact of a parity calculation is great in relation to the IOPS of the SSDs, the overall speed increase is generally so immense that the performance loss to the parity system is often of no consequence except in the most demanding environments and in those environments it is common to move to different models of data protection other than traditionally managed RAID arrays.
• Cost is very different with SSDs having a high capacity cost but a low performance cost, the opposite of Winchester drives. This means that using RAID 5 often results in a large cost savings while maintaining high performance and high reliability that does not exist with Winchester drives.

So this question came up (albeit a bit poorly) in another thread as to who is Hostadillo. Hostadillo has been mentioned once or twice on the community as a passing mention about website hosting. Hostadillo is a LAMP application host, focused primarily on WordPress. The site has not been linked or gone public yet, because it was not ready yet. But there was enough interest that some people had some concerns about who was involved with Hostadillo.

I had thought that it was obvious that I was involved in Hostadillo as I built the hosting publicly here on ML in a LAMP engineering thread that even had Hostadillo's name in things like certificate files. There was never an attempt to disassociate me from Hostadillo. Hostadillo is my own project done via the DCH group in Texas which it is very public that I am involved with (we even talked about this last night at MangoCon and I'm on the speaking list for tomorrow as a DCH rep - don't think that I hid my involvement there.) There has been some thought that @dominica (that everyone knows is my wife) is also involved with Hostadillo but she is not any further than being Hostadillo's web designer as that is what she does, designs websites.

Hostadillo came about when, as NTG's web app maintainer, I did a search for a web hosting product that really met NTG's needs (which are extensive as we do hosting work for clients, as well as internal work.) We have an existing web host that we still use primarily while we test Hostadillo for NTG use. It made sense for me to go down this path with my own non-NTG project because Dominica does web design and regularly needs access to a quality web host and none that I tested were found to be as good of a fit.

So I hope that that clarifies that I run Hostadillo as my own project in conjunction with DCH and that short of major independent announcement, has been relatively public since the project has not yet done a public announcement yet. But having it brought up in another thread today suddenly brought it to many peoples' attentions before it was really ready to announce so now it needs to be pointed out publicly that I'm involved there so that nothing is being hidden.

That being said... Hostadillo is the awesome choice for dedicated WordPress hosting. Hosting starts at just $10/mo and comes with full management. It's a good time to go public and let people know that we are ready to be hosting your websites! Check us out.

So long, 3DTV - we won't miss you
http://www.bbc.co.uk/news/technology-38778244

Good article on what we already knew, the whole thing was way too much of a gimmick.

@Drew said:

Great Post.

I've often seen people over at spiceworks blindly put down anyone suggesting RAID 5 as an option when brought up in SSD discussions. They've learned RAID 5 = bad period not RAID 5 = bad because of the above reasons.

Which is how we got into the problematic position in the first place because everyone read the white papers from 1998 that said "RAID 5 is generally the best choice because of cost savings" and interpreted that as RAID 5 = good and failed to learn the reasons that RAID 5 was chosen on small Winchester disks in the late 1990s.

It's amazing how quickly the pattern of attempting to learn IT as a simple set of rules rather than "good starting points combined with an understanding of the technology so that it is understood when to deviate from the norm."

The worst part is probably that so many people think that IT is so simplistic that we can have a simple set of rules that anyone without training could follow blindly rather than needing to actually understand the technology and apply that knowledge as needed. This is what leads, I think, to say many posts asking "What's the best X" as if there is always one single answer and that the secret to IT is getting this closely guarded knowledge rather than needing to understand products, techniques, risk, cost, etc.

IM ran a piece that I wrote last week on worker safety in disaster planning.

Youtube Video

At three this morning, Microsoft arbitrarily decided to turn off all of our Azure resources. No billing issues, no problems, no communications.... just turned them off. Everywhere, all regions. Everything failed. Multi-region protection failed, everything. Backups gone, everything. Every feature that Microsoft touts as protecting you on Azure - failed.

Seven hours later, we are still struggling just to get someone who knows what Azure is (no kidding!!) on the phone. We've had a Microsoft concierge on with us for a while and even he cannot get anything approximating support.

This is an epic outage. This isn't a datacenter blip that MS is working on. This is a massive, total outage for us as a customer that no one is working to fix. We can't sit back and wait for it to get resolved, we can't even find a way to get someone to understand that we are down.

This is the biggest lack of competence and of support that I have ever seen from a vendor like this. This is absolutely ridiculous. We have no idea how to get someone to provide support or get this working. Apparently Azure is too fragile and MS has lost the ability to provide support to use. Even with Rackspace's support issues, things like this never happen.

We are getting to wits' end here. Has Microsoft abandoned Azure? Why is there no support channel for it? Why is anyone allowed to just shut off systems?

Also posted to SW

@guyinpv said in Anybody in online retail and warehousing I need ERP advice:

They handle accounting a little differently. They use QuickBooks but do NOT import every sale and order and invoice etc. The only thing they record is the daily batches, that is, deposits.

Replace QB You don't normally use an ERP with QB, you replace QB with your ERP.

Share what you are reading currently.

The top feature that everyone has been clamoring for here in MangoLassi, the ability to get emails sent as soon as you get a new post to a thread that you are following or when you get an IM, has been added! For many of our users who have lots of experience with SW, this is expected behaviour that we have not been able to offer. Now the system can optionally act in a manner to which many of you are accustomed which should make it far easier for everyone to see activity and respond in a more prompt manner.

This is perfect timing giving all of the activity that we have going on this week.

@IRJ said in Cylance Unbelievable Tour Lives Up to Name, Can Cylance Be Trusted?:

@scottalanmiller For those of us that werent at SW '15 can you fill in the blanks?

Sure... I'll try to keep it as accurate as possible as some of the pieces are claims after the fact by the vendors.

• Cylance advertises a very technical security session on white hat hacking. It's the only extremely technical session advertised at SW '15.
• The session draws huge interest both before the event and during the event. Everyone is looking forward to finding out who Cylance is and getting one technical session in a sea of high level fluff. It's the only session that I attended other than Scale's storage replication talk.
• Later, Cylance states that during the conference at this time they found that the audience was not technical like they had been led to believe, so they decided to chance the talk topic from a highly technical one about hacking where they promised to show live hacking examples to a non-technical talk aimed at grandparents and consumers talking about how to spot malware in your phone's online store.
• No one changes any official details for the session. No announcement is made, the signage is not changed. The people standing at the doors signing people in don't mention anything. Every person who attends the session is there because they were told it was a technical hacking session.
• The session starts, no mention of them changing anything. Suddenly the talk is non-technical, embarrassing and insulting to the audience. Cylance looked like utter morons as if they don't even know exactly what malware is. Clearly they were not marketing to IT pros, but consumers.
• People storm out of the biggest session at SW en masse, furious that they were lied to and that the vendor was totally condescending.
• Later, Cylance back peddles and throws SW under the bus claiming that they misled them as to the audience. Which we can't verify, but doesn't matter because that doesn't explain the idiocy of the bait and switch. No matter what the audience was perceived to be, they were promised a show of Cylance's technical prowess and got a middle school presentation on the dangers of downloading random apps from the app store on your phone.

Now what really happened? We don't know. There is a bit of excuses and he said, she said stuff. What we know is that Cylance came across as pathetic and incapable; while also being insulting and offensive to the audience. SW came out looking terrible with their promised most technical session turning into their least technical one. It significantly skewed the value of the conference as it had been the keystone to justifying the technical education level of the event.

End of day, Cylance came out looking very, very bad.

Year later, they did far better at the event. But having a good booth and good interactions is only a tiny beginning step towards correcting a massive blunder.

Has anyone noticed the dates? November 1 - 3, 2016. Going to be so much better for weather. I'm very excited at the date change.

@Kimberley-Paessler congrats on the new job.

If you like, you could get the account renamed and stick around. No need for your account to be tied to the company.

Some folks asked for a follow up on the impact of MangoLassi Day so here I am, following up. Overall the day was a great success. As a technical community, numbers are never as big as you would think. It actually only takes a small number of active people to produce a lot of activity. That's why every new member matters a lot, it is all about the network effect.

The real goal of the day was exposure. We had a lot of people visiting from Microsoft's MSP community and even though few people joined during the day, a lot of people were watching us and that first exposure gets the community into the consciousness of the potential community.

There are definitely new users. A few from the days ahead of the event that we knew were signing up to participate. This is pretty much the new user list from MLD1:

Not a bad list at all. New vendors, new reps, new community members.... a good mix. And a lot of those people were posting during MLD1. A very healthy bit of user growth. Not thousands of new users or anything like that, but those are not realistic numbers for an IT community of this nature. We are very happy with the new users who joined during our push.

We are asking a direct question, just need a direct answer.

Did Ars Technica report something false? Or did Cylance actually do what they said?

We do self signed because.... users need to shut up.

So the question is then.... how many users does ML actually have?

And the answer is... 260. Two hundred and sixty real, live humans using the site. That might seem like a low number to some, but it is actually a very high, very healthy number. For a site with no advertising and no product to have 260 real people using it, and that number is always increasing (a little less than one a day, of course, but still, increasing) that is a really great number. We are very happy with the community growth and userbase.

@DustinB3403 said in FCC Net Neutrality Insanity Continues:

@coliver said in FCC Net Neutrality Insanity Continues:

@DustinB3403 said in FCC Net Neutrality Insanity Continues:

@coliver Fortunately none of the comments made this week will be taken into account as a part of some odd rule.

It looks like he is doing classic deflection, embrace some derogatory or inflammatory arguments to invalidate the entire base of arguments while ignoring the overwhelming majority of good arguments, or as it is colloquially known "American Politics".

The fact that thousands of companies have come out on the pro-net neutrality side, and he is saying the "19 ISPs" are against it just speaks volumes...

Ajit Pai I assume you're speaking about. Yes he needs to be removed from office. He used to work Verizon where he handled matters of competition, regulation and business counseling.

If that isn't a conflict of interest I don't know what is. His past clearly makes him pro business, anti basic regulations.

His past makes it highly possible. But very few of us are tied to former jobs in that way. I used to work as a journalist (two different careers within that two different businesses) but am aware of no media bias, even though they were good jobs. I don't go around telling people to listen to the media.

Likely he's corrupt and his former job was the basis for his connections to kickbacks. Or he just hates America. Or both. But just because he worked at Verizon means little other than he knew where to look for kickbacks.