What's coming in Cloudron 6.3
-
@d19dotca There is not option to ignore spam rules per forwarding list. A work around is to simple create a mailbox and then create a forward (this forwards everything). Let me look into adding an option per mailing list for 6.3.4. This will take me a day or two, so hopefully the workaround of creating a mailbox and a sieve forwarding works in the intermiitent time.
-
@girish Yes please. Spam filtering should never be forced without any option to bypass it, so it should definitely have always been an option to toggle on or off for mailboxes and mailing lists (likely defaulting to on I suspect). Thanks Girish if you can add it in the next release.
-
-
@girish I'll have to see, in case that other issue is applicable still when I convert it back to a mailbox again where spam isn't really handled in any sieve filters. I'll try to get this sorted out in the coming hours and confirm. I'd suggest this may apply to others than just me though so the sooner the better to fix that please.
-
@d19dotca spam still cannot be handled by user's sieve filters because the spam filtering is done before user sieve filters are run. so, for this specific address alone, you can just convert it into a mailbox and setup forwarding. the rest can remain as lists i guess.
(As for the fix, I am not sure about adding a "do not forward spam" option. I think it's best not to ever forward spam like it is now but make spam detection itself more configurable which I agree is quite rigid now. So, this is why it might take a bit more time).
-
FYI it looks like, at least in Firefox, the update button might be inside the <a> tag for the app itself? It always opens the app in a new tab in addition to taking me to the update page.
-
@girish Weird... I'll wait for a new app update an check again with cache cleared. It seemed to still be an issue with 6.3.3 but again, could be cache
-
@girish said in What's coming in Cloudron 6.3:
spam still cannot be handled by user's sieve filters because the spam filtering is done before user sieve filters are run
Just for clarity... are you saying Spam will never forward from a mailbox either then if marked as Spam by SpamAssassin since it "cannot be handled by users sieve filters"? Or are you saying if the mailbox is set to just redirect all messages that it'll still include spam message forwarding then too?
For what it's worth, until all the spam filtering improvements appear in a future release, it may be better to focus on making that optional for mailing lists until such spam improvements are made. Spam filtering should never be forced on anyone (I've made calls for that too when discussing Spamhaus for example), there should always be a way to bypass spam filtering when needed whether mailbox or mailing list, and that option should exist for any admins.
To be fair, I just re-read my original feature request and don't think I made it obvious at all it should have been an option, so that's on me as I may have unintentionally mislead you to thinking it should be a forced setting for all. It was definitely my intention for it to be optional. Similarly, I had also noted in that very same feature request "the ability to set a threshold number to avoid false-positives" specifically because if it's marked as spam to the mailing list it's then completely lost and can't be seen later unlike a mailbox that still keeps it in the spam folder for viewing, meaning it may be necessary to set a less-rigid spam threshold (from 5 default to say 8 for example) when it comes to rejecting mail to mailing lists.
Just some food for thought anyways.
In the meantime, I've created a mailbox again (moved away from mailing list) for this particular user and hoping that all messages (including spam) are redirected to them.
Sorry again Girish for possibly misleading the feature request and having an issue right away, haha, I admit it's a bit of an edge case because of the sensitivity of this particular recipient's job (a doctor) which means that all spam filtering should be on their side, not mine which is why I needed that option. -
@atridad I fixed a different glitch there. However first I thought I got what you meant with the update indicator, but now I am a bit lost. Also neither in chrome nor firefox I can reproduce the link opening pattern you are describing.
-
@d19dotca I think a mailing list not forward spam is correct (same way there is no option to not filter spam for mailboxes). I think what is missing is letting admins define what spam is and have more spam control (through this they can also define nothing is spam too). That said, I will see how much work all this is and decide on a short term checkbox or long term spam control.
-
@girish I agree that the spam filtering needs to be improved and in most cases it'd be fine to block spam to mailing lists (I basically wrote those feature requests after all haha), but that's part of the issue here too... in order for that new feature of not forwarding spam on to mailing lists to be effective and non-problematic (keyword there), we need those extra controls around identifying spam and setting all the different SpamAssassin options and such.
Until those extra controls arrive (it's understandable that'll take a while), the new feature can cause some unintended consequences for some since it didn't come with the ability to at least disallow spam filtering on a mailing list. There will certainly be cases where the spam filtering should be done on the recipients end, not Cloudron's end because at least if their spam filter catches it they can still see it in their junk box, but if the mailing list on Cloudron catches it then it's gone forever, so there's a bigger risk of Cloudron doing it for some types of accounts.
I would strongly encourage that ability to disallow spam filtering on mailing lists for those who need it at least until the better spam filtering is added to Cloudron along with all the other important mail features we've been clamouring for.
Right now, that feature - while useful in probably 90-95% of cases - is also now a detriment to the remaining percentage without extra controls in place.To be fair, maybe I'm misusing mailing lists though. I used them because it was much easier to manage for accounts which need to simply forward on to a different mailing address. Maybe I always should have been using mailboxes for that use-case? So I'll admit maybe part of the issue here is more the way I'm using it, but I still stand by the fact that the feature is a bit "half baked" without the ability to disallow spam filtering on them for sensitive accounts, and I hope that will be fixed until the time that the spam features are added down the road.
-
-
@girish Okay, thanks Girish for understanding.
In the meantime I'm using the mailbox route again for the few addresses where I don't want to be doing spam filtering / preventing the recipient from receiving spam messages.Relevant thread spin-off: https://forum.cloudron.io/topic/5277/spam-messages-forwarding-correctly-via-forwarding-sieve-rule/1
-
@atridad said in What's coming in Cloudron 6.3:
FYI it looks like, at least in Firefox, the update button might be inside the <a> tag for the app itself? It always opens the app in a new tab in addition to taking me to the update page.
Actually I am seeing this issue now as well, not sure why it didn't happen when I tested this yesterday. Anyways I've pushed a fix for that now.
-
Something for the stats: 10 minutes for a Cloudron instance with 6 cores, 32 GB RAM, Ubuntu 18.04.2 LTS. 42 apps from Cloudron 6.2.8 to Cloudron 6.3.4
-
Problem for me after updating to v6.3.4 running on Ubuntu 18.04.4 LTS.
Matrix (Synapse server) seems to be running, but sending messages on Android and Windows not working anymore...edit:
i moved this topic to https://forum.cloudron.io/topic/5291/problem-with-sending-and-receiving-message-s-android-windows -
@dantheman this maybe be better discussed in the matrix/synapse forum section as most likely this has nothing to do with the Cloudron update. If you have any further logs or error messages, please create a topic at https://forum.cloudron.io/category/50/matrix-synapse-element
