Running on AWS?

AWS bill, outage or inherited account? We fix it.

Most AWS accounts that reach us were set up in a hurry by someone who has since moved on, and the bill has been climbing quietly ever since. We read the invoice line by line, work out what every running resource is actually for, and shut off the ones that are costing you money for nothing.

  • Since 2015Fixing technology
  • FreeEstimates
  • US teamReal engineers

aws-events.log

still failing

What it is

Amazon Web Services (AWS) at a glance

Amazon Web Services is Amazon's cloud platform, launched publicly in 2006 with S3 for storage and EC2 for virtual servers. It is the market share leader, with over 240 services covering compute, storage, databases, networking, analytics, machine learning and security.

Pricing is pay-as-you-go: per request, per gigabyte, per hour, per region, and often per data transfer leg. Organisations use it to run anything from a single virtual machine to multi-account Control Tower deployments spanning several continents, and the bill grows in step with that complexity.

Type
Public cloud platform
First released
2006
License
Proprietary, pay-as-you-go
Maintained by
Amazon Web Services, Inc.
Typically used for
Hosting web applications, data platforms, enterprise workloads, backups and machine learning

The honest version

Amazon Web Services (AWS) pros and cons

No technology is all upside. This is where Amazon Web Services (AWS) earns its place and where it tends to cause trouble.

Where Amazon Web Services (AWS) is strong

  • The deepest catalogueOver 240 services covering compute, storage, databases, queues, analytics and machine learning. Whatever a project needs, AWS has a service for it already.
  • Global regions and availability zonesDozens of regions across the world, each with multiple isolated zones. You can keep data in a specific country and still run workloads close to your users.
  • Strong security primitivesIAM, KMS and VPC give fine-grained control over who can do what and where traffic can flow. Encryption at rest is a checkbox on most services.
  • Real discounts for commitmentReserved Instances and Savings Plans knock a large chunk off steady-state compute costs for teams willing to commit to a one or three year term.
  • Mature compliance certificationsHIPAA, PCI DSS, SOC 2 and FedRAMP are available out of the box, which saves months of paperwork for regulated workloads.
  • Everything is API-drivenEvery service has an API and a CLI, so infrastructure can be described in code with Terraform or CloudFormation and rebuilt from scratch.

Where Amazon Web Services (AWS) bites

  • The bill is hard to readCharges land per request, per gigabyte, per hour and per availability zone data transfer. A single service can show up on a dozen line items, and surprise bills are common.
  • Resources outlive the people who created themTest clusters, snapshots and load balancers get spun up and then forgotten. Nothing stops the clock, and the account keeps billing for them month after month.
  • IAM is powerful and easy to over-grantWildcards in policies are quick to write and very hard to audit later. Most inherited accounts have at least one role that can do far more than its job needs.
  • Egress fees catch people outData going out of AWS, between regions or across availability zones is billed per gigabyte. A chatty workload in the wrong shape produces an invoice nobody expects.
  • Inherited accounts often have no ownerThe root email belongs to someone who left, the phone number for MFA is disconnected, and the credit card is in a personal name. Recovering control takes time.
  • us-east-1 outages cascadeSeveral global control planes still live in Northern Virginia. When that region has a bad day, services in other regions can wobble with it.

Sound familiar?

Common Amazon Web Services (AWS) problems we fix

The Amazon Web Services (AWS) problems that bring people to us, and what you have once each one is fixed.

bill: $8,412 (prev month $422)

The bill jumped this month

A new log group writing at debug level, a NAT gateway shuffling terabytes, or a developer who left a GPU instance running over a long weekend.

Line by line, cause found and stopped

root email: ex-employee@...

The AWS account was inherited and nobody owns the root

The developer who set it up is long gone. The root email is a mailbox no one monitors, MFA is tied to a lost phone, and the card on file is personal.

Root recovered, ownership in your name

s3:GetObject on *

Something sensitive might be public

A bucket set to public years ago for a one-off share, or a policy with a wildcard that grants more than anyone realised.

Public exposure closed and verified

status check failed 2/2

An EC2 instance is unreachable

Underlying host trouble, a kernel update that broke networking, or a disk that filled and took the operating system down with it.

Back online with a path to replace it

storage-full: database frozen

RDS has filled up and gone read-only

Autogrow disabled, binary logs piling up, or a table that has been growing for months without anyone watching.

Space reclaimed, growth alerted on

us-east-1 elevated error rates

A regional outage is taking us down

A workload pinned to one region with no fallback, or a dependency on a global service that routes through the affected one.

Resilience plan the business can afford

11 security findings

Trusted Advisor is full of red

Root account without MFA, security groups open to the world, access keys older than the laptop they were generated on.

Findings closed, drift watched for

no infrastructure as code

Nobody knows what is running

Resources created by hand in the console over years, with no record of what each one is for or who asked for it.

Account mapped, code describes it now

What we do

AWS help, from rescue to upkeep

Call us for one surprise bill, or hand us the whole account.

Cost rescue
We read the invoice, trace every big line item back to a resource, and switch off what nobody needs. Numbers before and after.
Account takeover
Recovering the root login from a previous developer, moving billing into your name, and cleaning up orphaned users and keys.
Security review
IAM, public exposure, encryption, logging and the Trusted Advisor list, turned into a short list of things actually worth doing.
Resilience
Working out which outages your business can survive and which it cannot, and spending money only on the ones that matter.
Migrations
Onto AWS from another provider, between AWS accounts, or off AWS when something else fits better.
Managed Tech
IAM and MFA hygiene, monthly bill reviews, patch windows on the stack we built with you, and an engineer who already knows the account when something pages at 2am.

How it works

From first message to fixed

  1. Look Free estimate

    Tell us what's going wrong. We look at it, then tell you plainly what's wrong and what it will cost.

  2. Fix

    We fix the cause, not the symptom. If something is down, getting it back up comes first.

  3. Harden

    Backups, monitoring, updates and access, so the same problem can't quietly come back.

  4. Hand back or stay on

    You get a plain-language write-up. Or we stay on for hosting, upkeep and new work.

Often found together

Running Amazon Web Services (AWS) alongside something else?

We work on the whole stack, so the problem can be wherever it actually is.

Questions

Amazon Web Services (AWS) questions

Something else? Call+1 737 727 7327

Why did my AWS bill jump this month?

The usual culprits are NAT gateway data transfer, CloudWatch logs written at debug level, oversized RDS or EC2 instances left running, snapshots that nobody prunes, and cross-region or cross-AZ traffic. We open the Cost Explorer breakdown with you and name the cause before we touch anything.

Can you take over an AWS account from a previous developer?

Yes. We work through Amazon's root account recovery process, change the billing contact and payment method into your name, rotate every access key and password, and audit who else still has a foothold. It usually takes a few days because AWS verifies identity carefully, and that is a good thing.

Is our S3 bucket publicly accessible?

We check every bucket against the block public access settings, the bucket policy and the ACLs, and we look at what IAM policies and presigned URLs are doing as well. If anything sensitive is exposed we lock it down and tell you what, if anything, was reachable.

How do we cut AWS costs without breaking production?

By finding resources nobody uses first, right-sizing the ones that are oversized second, and only then buying Savings Plans or Reserved Instances for the steady baseline. We make changes in that order so the risky step comes after the easy wins.

What do we do if root account access was lost?

Do not panic and do not delete anything. AWS has a documented recovery process that involves proving ownership of the account, and it works. We have walked several clients through it and can do the same for you.

How much does AWS help cost?

Estimates are free. Share your rough monthly bill and what is going wrong, and you get a price before any work begins.

Free estimate

AWS giving you trouble?

Tell us what you're running and what's going wrong. A person reads it and comes back with an estimate.

  • No charge to find out what it will take
  • If it's urgent, say so

Replies come from a person. By sending, you authorize us to email you back. No marketing.