@nebulon to be honest I'm not sure neither now. But for some reason, quite a few times I had to create specific A entry to have DNS resolving to finish. And that is with wildcard entry existing for a few months now.

Is there something I can grab and share with you next time I will encounter that?

@djxx ha ha, sorry I misunderstood 🙂

The bounce message is sent from Cloudron for undeliverable messages. You can send randomaddress@domain.com and you will get that bounce back message.

Going back to your original problem, if gmail is blocking you , the only way to unblock is to wait it out. From your original mail, you are getting blocked at IP level. This means that gmail most likely blocked your entire VPS provider range. You have to set up a relay to get out of this situation - https://docs.cloudron.io/email/#relay-outbound-mails . If your mail is only getting to spam (instead of IP blocked), then usually you have to just "wait" it out. From many years ago, we just sent email from cloudron.io to various gmail friends and asked the to mark as not-spam. After that, it started going through. But gmail is a black box, who knows what they do now.

@Klaus Which email provider are you testing with? We can try to implement it.

Worthwhile reads - ``Reply-To'' Munging Considered Harmful and Reply-to-List, or Reply-to-All?

@girish said in Elasticsearch:

@Nicolas Indeed, current plan is to implement Elastic search as an addon and not as an app.

Might be time to revisit that.

Elasticsearch is open source, again
https://www.elastic.co/blog/elasticsearch-is-open-source-again

@girish said in Backup time considers timezones incorrectly:

I think you are suggesting showing the Settings Timezone in the Backup Schedule UI, correct? If so, that's a good idea!

Yes, that was what I was suggesting

The timezone shown in the UI is based on browser timezone . If you had multiple admins, each in their timezones, then the UI shows the correct timestamps.

I understand the logic, however, have a look at the screenshot. screenshot-backup-time.png

Seemingly, the backup starts at 1am, but it finishes before 1am, which is confusing. In fact, it took 24 minutes in this case, but I am in a different time zone now, 1 hour back.

also, if any of you have tested this, let me know.
I wonna know how you tested and how it was.
i've also PM'd @3246 to see if he could test it while I'm still trying to find a server.
and I think I might have to open port636 but maybe not, because duo connects to your actual server, not tryed to authenticate there, it wants a connection to your actual real life server. i'm curious to see how this will actually work, or if cloudron has a docker container i'd put it in.

@nebulon the doc might need to be updated

systemctl stop|start docker docker.socket postgresql mysql nginx for the services.

rsync --archive /mnt/old_root/home/yellowtent/boxdata /mnt/data/cloudron-mail/ for copying the data (mv when fails produce a whole set of problems to be resolved and can't be resumed).

and the following commands to verify:

find /home/yellowtent/boxdata/ -type f -printf "." | wc -c find /mnt/data/cloudron-mail/boxdata/ -type f -printf "." | wc -c diff -r -q /home/yellowtent/boxdata /mnt/data/cloudron-mail/boxdata

For the most part, everyone else covered my use case on this.

Either the ability to download the file as a tar.gz, that isn't encrypted to store locally or external storage.

To take this functionality a bit further down the road:

Offload the app and environment settings to an external S3 storage, gui to view these on like the file explorer, view basic details (time offloaded, size, dates, view file contents). There have been times where I had a different version of an app, and I just needed to quickly grab the .env file. It would save a lot of time, to be able to view those files, without having to sift through backups and files.

@girish Yup, nothing to say that a Cloudron "App" has to be a Docker Container.

@systemaddict I would check the app's configuration. Some apps let you change the CORS. It's not correct for system adminsto change CORS settings, this is a security issue waiting to happen.

@nebulon a checklist of current apps will be ideal scenario

@fbartels , thank you! I need a DMZ-like proxy, that is the only self-hosted VM / server that is open to the internet, if that makes sense?

This is implemented in Cloudron 8

@ekevu123 the manifest error here refers to the docker image manifest. That part of the error message is from your upstream docker registry and not generated by cloudron.

I misread your original comment. I guess you are referring to the cloudron update command of the CLI tool. I thought this was about a cloudron (platform) update.

Currently, the focus of the CLI tool is app development and not for end users as such.... So, one can automate what you want already by writing a small script which calls cloudron restore if the update fails.

A temporary workaround: docker stop turn . Some platform updates might turn back the turn service on though (when updating the turn docker image).