@BrutalBirdie So I figured this out, and it's definitely NOT a Cloudron issue. Actually, it's a Thunderbird issue... so I'll probably go write a quick report on Bugzilla at some point.

It turns out that when you hook up two emails for Thunderbird that (1) go the same mail server and (2) share the same password, that causes Thunderbird to ignore the username.

What this means:

Let's say you set up two mailboxes in Cloudron, your.name@company.com and service@company.com. While your.name@company.com is accessible to you, service@company.com is instead accessible to a group you are a part of. You can use user account password to log into both.

Let's say you first add your.name@compay.com to Thunderbird. It will work fine. However when you add service@company.com, outgoing will be messed up (though incoming will work). Rather than outgoing being registered for service@company.com, it'll go out through your.name@company.com, which is not the correct behavior.

You can fix this by going into Outgoing Servers in Thunderbird, but it's a big pain it doesn't need to be, and I'm sure it's something that can be fixed on the Thunderbird end.

Anyways, that was a fun dive into some email issues... hopefully I can get this fixed and jump back into using groups, as that helps make security issues with emails a little easier to deal with (no having to reset everybody's shared password).

Hope you all have an awesome day!

@chris thanks for creating those!

@benborges Ah ok. Maybe you can wait for 7.3. Please update to that and try to update again and we can debug further. There are some cgroup v2 related fixes for Ubuntu 22 in 7.3 which affect docker container creation, it's possible you hit that case.

@girish said in Mail server - How to expose ports for internal network:

@ChicagoGregg said in Mail server - How to expose ports for internal network:

Is it possible to expose its ports (25,465,587) to the local network so that mail can be sent (relayed) through the mail server and let it handle it going outbound? Figuring that Cloudron has this capability, and I'd only have a few emails per day requiring this, it should be able to provide this.

If I understand correctly, you want to relay mail via another mail server. If so, just sent the relay using the External SMTP relay configuration.

I am not quite following the part about exposing port 25/476/587 though. These are already exposed on the server. You have to configure your router/firewall to port forward those ports to the Cloudron internal IP.

Sorry for the tardy reply here. Thanks for both of your replies @subven and @girish .

Originally, I was simply wanting to do something like

Home Network (App sends email) -> Cloudron (on home network) accepts mail for relay from local home network 10.x.x.x -> sends to server through Internet connection (to my already-configured mail relay server, External w/ authentication)

I was looking for an unauthenticated relay on the home network.

Well, I suppose that Cloudron can't provide an unauthenticated relay point. So, I suppose the workaround is simply to configure a user and relay using authentication using a domain that is configured.

I'm fine going this route... And I get why it's for that reason with security. 🙂

Thanks for the help!

-gregg

@Wishuwero do a systemctl restart box and refresh the browser. That should bring back the setup UI again.

@necrevistonnezr not sure about Roundcube as I moved over to SnappyMail, but I think it is similar.
Under Filters you can add a sieve script (maybe they call it 'custom')

And yes I think discard with message is equivalent. But not an expert.

Due to the excellent spamassassin rules posted here https://forum.cloudron.io/post/30512 by @d19dotca, I am not needing to use sieve in SnappyMail. But here is the script which I use in Protonmail. The from can be changed to to I guess.

NB : I put the reject part of the rule before the spam test. Arguably I should have a return after the reject - not sure why I did not do that.

require ["include", "environment", "variables", "relational", "comparator-i;ascii-numeric", "spamtest", "reject", "fileinto", "imap4flags"]; if address :contains "from" ["spammer@blah.co.uk", "spammer.com"] { reject "Spam"; } if spamtest :value "ge" :comparator "i;ascii-numeric" "5" { fileinto "spam"; } if exists "list-unsubscribe" { addflag "\\List"; } # Generated: Do not run this script on spam messages if allof (environment :matches "vnd.proton.spam-threshold" "*", spamtest :value "ge" :comparator "i;ascii-numeric" "${1}") { return; }

https://forum.cloudron.io/topic/7619/filesystem-mount-inside-container for others to follow along.

ZFS is not supported/untested. I would say please open a feature request in the feature request category if this is required and we have to analyze the benefits/use case.

Hi there, please send us a mail with your Cloudron ID and appstore account email to support@cloudron.io

@skeats this is fixed already but pending a Cloudron release. We got a bit behind schedule this time.

@jdaviescoates Thank you so much!

@nebulon it would be lovely to see those credentials, for example, on the system settings or service settings somewhere - could be hidden by default, unveiled only after the click on some button, like 'show me the password'

@nebulon perfect, thanks for confirming!

@humptydumpty said in No graphs on any of the apps:

On my 22.04 server, the graph in the app settings does not show

This is fixed in the next release. The upcoming release is a bit delayed but we are working on getting it out next week or so. (Sorry, some of the mail features are quite time consuming to implement).

@girish Thank you - now its working again.

Or if the app supports import/export of it's own, then install the latest version of the app and import the previous one into it.

Can you please also post the output of:

systemd-resolve --status | grep 'DNS Servers' -A2

and

systemctl list-units --state=failed

@nebulon thank you, hope to hear some wisdom from the community

@Ritesh I suspect the issue is the low side of your memory to begin with, but wanted to also add that if you add more memory and still experience issues where it seems the memory or CPU usage is constantly high, you can reach out to Vultr's support as the issue may be on the hypervisor where some "noisy neighbours" may be happening which is starving your VM of resources. I had a similar issue a few months ago and they addressed it on their end after a few days.

As the root cause of this was, that the backup was continuing even though the backup disk was not mounted, we were now able to find the bug which caused this and possibly other similar issues.

So the check itself for the mountpoint was correct, but this result was just ignored by the code. This oversight will be fixed for the next release and should avoid such cases for mounted backup volumes in the future.

@Stardenver Interested in this use case too but have yet to do any organized testing....

There are some main factors regarding how Cloudron runs things that I considered important

Containerized apps and core services (linux) Docker network (the network resides on one linux instance)

Given docker and the network are open in terms of running multi process/threads besides the CPU speed It mainly depends on how the app itself is built

A good rule of thumb would be to just keep a nice ratio between speed and amount of threads

If you can going with high speed and most amount of thread's is the most "optimal" in terms of specs, not so much price...

I know its an open ended answer but that's because there is no one "right" answer lol

+1

@BrutalBirdie said in Right setup for HDD/SSD mix:

@Stardenver said in Right setup for HDD/SSD mix:

Anybody here running Cloudron on a HDD only system?

Yep, one of my Backup Servers who is running +10x Minio apps for backup space with 10TB HDD space is running HDD only.
A restart can take some time, but with 10TB it is expected.

For such mixed systems I would always go for System SSD and Storage HDD.
But storage can also have part of the SSD.

In the end all is backuped somewhere in the clouds and if anything fails or needs to be changed 🤷 just restore.
(To be able to have this kind of mindset is one of the best features imo :D)

Thank you very much for taking the time to answer me. I already figured out that its pretty easy to add several kinds of external storage to Cloudron. To be honest.. I am thinking about getting a cheaper server now and add some cloud storage to it. I am using as storage solution from a german provider and it worked pretty good in the past. So I guess I will go with SSD only and add additional storage by mountig this storage box to my server.

Nothing that mentioned the _ (wildcard) cert at least, which is part of why I'm stumped. I don't know what is creating it, or where to look.

@subven Hey, thanks for the answer. I have the box connected via SSHFS, but I tried switching from rsync to Tarbal and I think that probably solved the problem. Thank you very much

@nebulon Sorry for the late reply, this topic has been resolved with your support. It was a misleading page. I will start a new thread on the backup issue. Please mark this as done or delete. All solved here. Thank you for your help.
https://forum.cloudron.io/topic/6895/the-site-is-deceptive/14?_=1661248303629

@girish bingo - that did it - next time I'll read the full error instead of freaking out - gotta love those "uncertified" apps! Thanks mate

@d19dotca I can't believe I missed that. Literally read right past it. Thank you!