Enhance MCM uptime.

[Blasthappy]

It is quite ridiculous lately how bad MCM is with uptime. It is no longer the once or twice a day conundrum, it is nearly all day lately. If Mick does not take action soon, this site will slowly die.

Current Situation:

This site is protected by the services of Cloudfront, a service from Amazon. As you can tell, this system is not working too well. I would like to debunk a common misconception, and that is that people seem to think that AWS (Amazon Web Services) are to blame for this. That is not the case as AWS is a hosting service, and from my current knowledge, Mick does not run off of AWS anyway.

Proposed solution:

Cloudflare. This service has served me well and even if it has its flaws, you can't really say that Cloudfront is better. In Mick's case, I'd suggest this plan:

Spoiler: Plan

This plan provides more than enough for this site's needs and it can quite easily keep this site up much longer than the current system can.

I hope this helps some people understand the situation a bit better and how there is a quite easily feasible solution to this, the only hurdle being Mick's stubbornness.

 

[Overlord]

Oh... you people are fucking amazing.

To establish some credentials here... I have more certifications, qualifications and experience in this area than most of you have in any area.

As BeBosny said, AWS is a world class cloud computing platform. I don't use that term lightly, since I think most providers are complete ass. AWS is used by governments and massive worldwide corporations. AWS is considered a competitor to the likes of Google. GOOGLE. Take that in for a second. Cloudflare does not have the resources to compete with AWS at all, not only do they do two completely different things but AWS serves a very large number of sites holding very private and mission critical data. I mean shit that should definitely never get out.

AWS is used by a lot of companies. You probably interact with sites using AWS on a daily basis. I know of banks on the AWS infrastructure, Stripe, Adobe, Airbnb, Atlassian, BMW, British Gas, TV channels like 4 and ITV, Hitachi, Sage, SoundCloud, Slack, J&J, Intuit, Netflix, Samsung, Siemens, Unilever...

I don't mean these sites use AWS for some little site for a little team in some random city in Iraq or something; I mean they use them for their core infrastructure, their direct websites. They rely on AWS infrastructure to deal with their systems. Additionally, I know for a fact banks and government agencies use AWS as well.

So before you want to trash on AWS services and say it's in-superior to Cloudflare: fortune 500 companies depend on AWS and AWS is the leader, by far, in the cloud computing market. By far. Google can't even get near AWS.

Based off that, do you really think you know more about the AWS network than the thousands of full-time engineers that are much more qualified than any of us? I mean, you've got to be kidding me, you barely know what Cloudflare is.

buildblox
I know, CloudFront is a separate service within AWS.

AWS is not a separate service within AWS, CloudFront IS part of the AWS cloud computing platform. If you didn't know, a cloud computing platform will contain many different services (e.g. compute and processing, content delivery, storage, networking).

--

AWS aside, now tell me what the problem with this site is, and what makes you think AWS services are the cause for it? You have no access to the back-end of this site and you're telling Mick your expert, informed opinion on the cause of downtime and why moving to Cloudflare will fix all the problems?

FYI, Cloudflare is a reverse proxy. If the back-end server is down, Cloudflare will not connect. Cloudflare edge locations probably have more issues than CloudFront, and Cloudflare *never* excels at any one area. Their DDoS protection is par, the WAF is par, the CDN isn't even a CDN... Cloudflare is a bunch of services for websites that can't be bothered to deal with individual services in each one of the web security and optimisation areas, and instead want one singular service to make its best attempt at it.

I mean...

I hope this helps some people understand the situation a bit better and how there is a quite easily feasible solution to this, the only hurdle being Mick's stubbornness.

Who the fuck do you think you are?

You're here with, probably, zero technical knowledge and no idea of the actual issues the site has had (if any) and you're telling them what you think happened and then decide to call him stubborn if he doesn't listen to your, clearly uninformed, opinion.

You're truly amazing. It's amazing.[DOUBLEPOST=1488838147][/DOUBLEPOST]

Either way, I agree with BeBosny, as stated in the post I made, I think the system admin team is quite terrible and un-qualified for a website of this capacity.

Tell me why you think that. Do you know what qualifications the team has? Do you know the work they do? Do you know anything about the back-end of this site at all?

 

[lAkjtzAZ0]

Tell me why you think that. Do you know what qualifications the team has? Do you know the work they do? Do you know anything about the back-end of this site at all?

Well, yes actually many people do know quite a bit about the "team" in place. The day Mick got the new system admin, there were major issues. Days after, more major issues. For months now there have been issues that never existed before Mick regained control with a new system admin. There were times when we were getting ZERO answers as to what was happening and a few "uh, we don't know"'s as well. There have been countless issues and I don't think a week has gone by where MCM hasn't had a few hours of downtime in a day, sometimes it happens for days on end. No one REALLY knows what's happening behind the scenes with MCM, however it's pretty clear that either there's a very, very serious issue that they don't know how to fix, or they're just not good at what they do and can't improve/fix simple issues (if it is a simple issue, again no one really knows). As for the AWS issue, you're right - it's not the problem. It MAY have been the cause for a short period of downtime whenever AWS had it's outage, however everything else is non-AWS related yet Mick/the sys admin is still trying to claim AWS is the issue in the recent 'Regarding recent downtime' thread. <- this is part of the reason people are calling him stubborn. Mick himself is putting blame on AWS hours after the outage was cleared up and fixed. Whatever the issue is, it's kind of sad and if it keeps going at this rate it's gonna hurt.

 

[Zyr]

CloudFlare free plan > CloudFront

Just kidding, I don't know anything about this stuff but I definitely don't support pineapples on pizza.

 

[Yeti]

Yeti
Alright, shit posting is just plain dumb on this thread, keep this professional.

I'd appreciate it if you refrain from harassment in the future as it may lead to dire consequences.

 

[Overlord]

Well, yes actually many people do know quite a bit about the "team" in place. The day Mick got the new system admin, there were major issues. Days after, more major issues. For months now there have been issues that never existed before Mick regained control with a new system admin. There were times when we were getting ZERO answers as to what was happening and a few "uh, we don't know"'s as well. There have been countless issues and I don't think a week has gone by where MCM hasn't had a few hours of downtime in a day, sometimes it happens for days on end. No one REALLY knows what's happening behind the scenes with MCM, however it's pretty clear that either there's a very, very serious issue that they don't know how to fix, or they're just not good at what they do and can't improve/fix simple issues (if it is a simple issue, again no one really knows). As for the AWS issue, you're right - it's not the problem. It MAY have been the cause for a short period of downtime whenever AWS had it's outage, however everything else is non-AWS related yet Mick/the sys admin is still trying to claim AWS is the issue in the recent 'Regarding recent downtime' thread. <- this is part of the reason people are calling him stubborn. Mick himself is putting blame on AWS hours after the outage was cleared up and fixed. Whatever the issue is, it's kind of sad and if it keeps going at this rate it's gonna hurt.

hm, maybe you should have him create a https://www.statuspage.io/ instance to track uptime of the site and its services

i don't frequent this site, but yeah, sounds like an issue then. you definitely cannot blame it on aws, though.

 

[buildblox]

Oh... you people are fucking amazing.

To establish some credentials here... I have more certifications, qualifications and experience in this area than most of you have in any area.

As BeBosny said, AWS is a world class cloud computing platform. I don't use that term lightly, since I think most providers are complete ass. AWS is used by governments and massive worldwide corporations. AWS is considered a competitor to the likes of Google. GOOGLE. Take that in for a second. Cloudflare does not have the resources to compete with AWS at all, not only do they do two completely different things but AWS serves a very large number of sites holding very private and mission critical data. I mean shit that should definitely never get out.

AWS is used by a lot of companies. You probably interact with sites using AWS on a daily basis. I know of banks on the AWS infrastructure, Stripe, Adobe, Airbnb, Atlassian, BMW, British Gas, TV channels like 4 and ITV, Hitachi, Sage, SoundCloud, Slack, J&J, Intuit, Netflix, Samsung, Siemens, Unilever...

I don't mean these sites use AWS for some little site for a little team in some random city in Iraq or something; I mean they use them for their core infrastructure, their direct websites. They rely on AWS infrastructure to deal with their systems. Additionally, I know for a fact banks and government agencies use AWS as well.

So before you want to trash on AWS services and say it's in-superior to Cloudflare: fortune 500 companies depend on AWS and AWS is the leader, by far, in the cloud computing market. By far. Google can't even get near AWS.

Based off that, do you really think you know more about the AWS network than the thousands of full-time engineers that are much more qualified than any of us? I mean, you've got to be kidding me, you barely know what Cloudflare is.


AWS is not a separate service within AWS, CloudFront IS part of the AWS cloud computing platform. If you didn't know, a cloud computing platform will contain many different services (e.g. compute and processing, content delivery, storage, networking).

--

AWS aside, now tell me what the problem with this site is, and what makes you think AWS services are the cause for it? You have no access to the back-end of this site and you're telling Mick your expert, informed opinion on the cause of downtime and why moving to Cloudflare will fix all the problems?

FYI, Cloudflare is a reverse proxy. If the back-end server is down, Cloudflare will not connect. Cloudflare edge locations probably have more issues than CloudFront, and Cloudflare *never* excels at any one area. Their DDoS protection is par, the WAF is par, the CDN isn't even a CDN... Cloudflare is a bunch of services for websites that can't be bothered to deal with individual services in each one of the web security and optimisation areas, and instead want one singular service to make its best attempt at it.

I mean...

Who the fuck do you think you are?

You're here with, probably, zero technical knowledge and no idea of the actual issues the site has had (if any) and you're telling them what you think happened and then decide to call him stubborn if he doesn't listen to your, clearly uninformed, opinion.

You're truly amazing. It's amazing.[DOUBLEPOST=1488838147][/DOUBLEPOST]
Tell me why you think that. Do you know what qualifications the team has? Do you know the work they do? Do you know anything about the back-end of this site at all?

Thank you for reiterating! Hopefully people will listen, since it appears my post was clearly not enough.

 

[AntiMatter]

lol, it's funny to see all the people hating on AWS/CloudFront and wanting MCM to switch to Cloudflare because they think Amazon is the issue :rofl:

I even heard this site got booted off successfully when it had Cloudflare before.

 

[Wvisoecj]

Yet another suggestion by someone who doesn't understand what AWS and CloudFront nor how HTTP requests work.

Your stated "fix" won't do anything.

 

[Mick]

Thanks for the help Overlord.

Denied, thanks for the suggestion.