Forum Replies Created

Viewing 4 replies - 1 through 4 (of 4 total)
  • Thread Starter oskapt

    (@oskapt)

    Looking at your code further, it seems that you’re simply trying to determine if a file is _already_ in the queue so that you don’t insert it twice. In that case, a better query would be:

    SELECT id from wp_w3tc_cdn_queue where remote_path=<path>;

    EDIT: My earlier comment about this being an indexed field is actually incorrect. You’re still searching on a field that is non-indexed, so it’ll still be slow. All this does is reduce the size of the response, which is insignificant.

    You need an index on remote_path.

    I don’t know what the minimum privileges are, but if you limit the plugin to a specific user and give it privileges on a specific bucket, you should be fine. I have the following for an IAM user whose sole purpose is to handle W3TC for our sites:

    {
        "Version": "2012-10-17",
        "Statement": [
            {
                "Effect": "Allow",
                "Action": "s3:*",
                "Resource": [
                    "arn:aws:s3:::my-prefix-w3tc-*",
                    "arn:aws:s3:::my-prefix-w3tc-*/*"
                ]
            },
            {
                "Effect": "Allow",
                "Action": [
                    "s3:ListAllMyBuckets"
                ],
                "Resource": "arn:aws:s3:::*"
            }
        ]
    }

    After spending all day working through the white screen of death when sending email via SES, I’ve rolled this back to 4.3.1. Thank <insert deity of choice here> for backups.

    Most of the complaints here are about the changes to forms and tags, but I’d like to raise a complaint that this completely broke a different plugin, one of 2 plugins that allow WP to send email through Amazon SES.

    I thought about posting this directly to the developer via their contact form, but they’re clear that “all support requests sent through this form will be ignored.” Does “your plugin broke another crucial plugin” count as a support request or a heads up?

    We’ll have to see if the developer rolls through here anytime soon and addresses any of these issues.

    This problem only appears in Chrome, and our solution to it was to use Safari. I’m curious why this has been relabeled as “resolved” without a comment as to what the fix was. If either of the posters has a resolution, please share it. If it was resolved by Automattic, please share the solution. If it was not resolved, please reset the post title.

Viewing 4 replies - 1 through 4 (of 4 total)