Questions[lifecycle](server)

Is there a way to exclude certain tags/images from a cleanup policy?

E.g. say I have a repository with images and the related SHAs for the Pull Request which triggered their creation. When these are approved for deployment to an environment the image is tagged with that environment's name, resulting in something like this:

+---------------------------------------------------------------------------+
| Image Tags      | Image URI                                               |
+---------------------------------------------------------------------------+
| sha923456       | 000000.dkr.ecr.eu-west-1.amazonaws.com/myrepo:sha923456 |
+-----------------+---------------------------------------------------------+
| sha823456, test | 000000.dkr.ecr.eu-west-1.amazonaws.com/myrepo:test      |
+-----------------+---------------------------------------------------------+
| sha723456       | 000000.dkr.ecr.eu-west-1.amazonaws.com/myrepo:sha723456 |
+-----------------+---------------------------------------------------------+
| sha623456, prod | 000000.dkr.ecr.eu-west-1.amazonaws.com/myrepo:prod      |
+-----------------+---------------------------------------------------------+
| sha523456       | 000000.dkr.ecr.eu-west-1.amazonaws.com/myrepo:sha523456 |
+-----------------+---------------------------------------------------------+

I want to ensure that what's currently deployed to any environment doesn't get deleted; so those images tagged test or prod should be kept. In addition, the SHA tags against their images should be kept.

Beyond that, I want to keep all images created in the last 90 days with an SHA tag.

I'm happy for anything untagged to be removed.

Applying the below rules almost works; only it seems that because there's a match on the sha tag, the prod and test images would be expired.

{
  "rules": [
    {
      "rulePriority": 1,
      "description": "Remove untagged images",
      "selection": {
        "tagStatus": "untagged",
        "countType": "imageCountMoreThan",
        "countNumber": 1
      },
      "action": {
        "type": "expire"
      }
    },
    {
      "rulePriority": 100,
      "description": "Purge non-deployed images over 90 days old",
      "selection": {
        "countType": "sinceImagePushed",
        "countUnit": "days",
        "countNumber": 90,
        "tagStatus": "tagged",
        "tagPrefixList": [
          "sha"
        ]
      },
      "action": {
        "type": "expire"
      }
    }
  ]
}

I can't find any documentation on adding a "NOT" rule, and trying an exclamation doesn't work.

    {
      "rulePriority": 100,
      "description": "Purge non-deployed images over 90 days old",
      "selection": {
        "countType": "sinceImagePushed",
        "countUnit": "days",
        "countNumber": 90,
        "tagStatus": "tagged",
        "tagPrefixList": [
          "!test", "!prod"
        ]
      },
      "action": {
        "type": "expire"
      }
    }

I work in a small organization that uses a single file & print server with an Active Directory domain for user management.

The office admin who inducts new people is non-technical and generally creates the AD user account because there is no permanent IT support on-site and users need their accounts immediately. However, there is follow-up work that needs to be done after that (creating an extension on the PBX, setting up voicemail, form filling, etc, etc) which needs to be farmed out to various different people.

The manual process we have works for the most part, but has gaps and things sometimes get forgotten. I am not in any position to re-architect these business processes, such as they are, so for the purposes of this question, please can we take it as read that the business process is immutable. My solution needs to mesh with the existing process.

Essentially what I think I need is a way to generate email notifications upon manual creation of a new user in Active Directory - a sort of lightweight workflow engine that just emails people and/or my ticketing system with the follow-up tasks that need doing. Yes, I know it's not best practice, yes I know its not cool, but given my situation, I think that's the most pragmatic solution.

Is there a simple, lightweight way to do this? I'm thinking along the lines of a PowerShell script or something like that (there's no budget for a proper IDM system and, as I mentioned, the business process is immutable).

There is a server that is used from 4:30 am in the morning until ~ 22:00.

Should it be turned off? I think that it is a server and that it won't have a problem to stay on, but serious professors are telling me that it is dangerous and that HD can fail within 2 years. The server owner believes that his old server running from 1995 without backup and a single hard disk (if the hard disk fails he is screwed) had no problem because he used to turn it off at nights.

What do you believe for this?

Now it has a RAID 1 array, external hard disk backup, and serveral full hard disk backups on DVD and over the internet.

I am considering moving from a VPS to EC2. EC2 is elastic and so is its pricing; I could boot my instance on demand, then let it shut down if it isn't active by the end of some hourly period.

How much time does that process take? Assume a micro instance booting from EBS. Assume Linux (probably Ubuntu); the 10 minute times mentioned for Windows leave me unimpressed. The transition could be ec2-run-instance (with its pending state) or ec2-start-instance. Feel free to mention the start up times of other clouds if you know them.

LTO tapes, Maxell in this case, are often marketed as having 30 years or more shelf life when stored under "optimal conditions"

Is there a way to get a good estimation to the shelf life, given parameters such as relative humidity and temperature etc?

Obsolescence of the tapes aside, is there a way of determining the impact to shelf life of any deviance from the optimal. In other words how many years are lost when storing say 1 degree above the specified range?