Store attachment differently and more effectively.
There is a cronjob that runs hourly. It checks for unused attachments and delete them.
For how long the attachment should be unused so the con to delete the attachment on s3?
If it is unused at the time the cron runs, it will be deleted right away. No wait time.
I've tried to go back and forth between the s3, and local...and it's not working out so well for me.
Media | Ohio Sportsman - Your Ohio Hunting and Fishing Resource
Check out all media from Ohio Sportsman - Your Ohio Hunting and Fishing Resource
I can see the files in the s3 bucket, but can't seem to get them to revert locally...or even back to the bucket url.
Can you describe the steps that you took? Between each switch, did you run the move tool?
Apparently your site blocks my IP
What country are you in and I’ll be sure to unblock it...we get a lot of spam
I updated to the latest add on. I turn off the ad on and run the move tool.
For turning it back on, I fill in all the details...and run the move tool. Takes about 8 hours when the tool is ran. When I move it back I keep a local copy. The urls of the images are almost correct it’s just the domain they pull from that is wrong
I'm in Vietnam.
You said you turned off the add-on and run the move tool? That doesn't sound right because if you turn it off, the move tool will disappear.
You also said after turning it back on, you fill in all the details... Did you put in the same details as before (before turn update/turning off)?
About "just the domain they pull from that is wrong". For S3, the URL domain came from the bucket unless you specifically told it to use another domain.
When I say turn it off, I do not mean disabling the add on but the option.
Okie, so let me describe the steps again:
1. You were running some older version of the addon with S3, without issue.
2. Update to latest version
3. Switch option from S3 to default
4. Make a backup and run move tool
5. Switch option from default to S3, fill in the details
6. Run move tool again, for 8 hours
7. Compare urls after step 6 with backup from step 4, they are almost correct except the domain
Am I correct here? Now, some questions:
For step 4, how long did the tool take? After step 4, how much space internal_data/attachments were taking?
The details in step 5 and the old configuration in step 1, are you sure that they are identical?
For step 7, can you share previous URL and current URL?
This add-on worked like a champ up to version 1.3.0, but the last couple of updates have messed up how attachments are handled using the S3 settings. There are a few people who have hinted or flat out said this, but to date it feels like the problem isn't being addressed. I just turned off S3 and set it back to default until this gets fixed in a new update. I have tried every piece of advice and procedure listed in your discussion threads and How To posts but nothing makes a bit of difference. Please take some time to fix what's wrong with how this add-on interfaces with S3 and why working configurations broke after upgrading to version 1.3.4 and beyond.
I'm actively working with admins to figure out what caused problem with their specific configuration. Maybe some combination of region is causing issue. We are running on S3 at xfrocks.com (Singapore region) and it's working without issue since the update...
Anyway, how did it stop working for your? No new upload or just flat out errors?
It was very simple as to how it stopped working for me. I upgraded the add-on like usual and didn’t change anything in the ACP settings. Before long one of our members who posts images frequently reported he was unable to successfully post his photos and that’s when it started. I’m probably 3 weeks into this and no closer to a solution.
I assume you have no server error log? Do you have delayed processing turned on, probably a no too?
I posted about this up thread and delayed processing has always been turned on. I have combed the server error logs in Cloudwatch and nothing stands out, only the S3-shippable-helper log was able to shed light on things - mainly the correct endpoint to use. The part I do not understand is it was showing the S3 region of east after the add-on upgrade to 1.3.4 when before it was set to west. That was not a setting I changed though, so I’m baffled as to how it automatically went to the east zone. Last night before I finally set it back to standard to get attachments working again I was getting a very similar error as was posted by here by Pete [bd] Attachment Store [Paid], but before that the same thing was occurring with us, the image would be there and then it would vanish after a few minutes, with no errors of any kind.
I can put the settings back to the way they were and post a screenshot of ACP, but not until this afternoon due to my schedule. If you want to have admin access to look around tell me what you need and I’ll work on it later today.
Edit: if you use Team Viewer we could work your access that way too, by you logging into my laptop where I have all the access you’ll ever need to check things out.
I will try TeamViewer then, please check your conversation. Thank you.
I have found the security issue, and removed it. Can you try the link again.
Separate names with a comma.