Backblaze "Snapshot" folder: multiple file copies retained

[humpty]

I had my BB bill grow every month until I decided to wipe all backups and triggered a full backup from the Cloudron side. It seemed to have fixed the issue as my bill has been pretty consistent afterwards. I'm not sure if it's a bug or due to the server migration. In any case, manually purging things once did the trick.

You mentioned multiple copies are kept. How many backups do you have scheduled per day and what's the use-case to have more than one?

[p44]

@humptydumpty thanks for your feedback.

It depends on istance, I've at least two daily backup, with a week retention policy. Total backup size (for all Cloudron istances) is around 400GB daily, around 2.8TB data per week.

I moved from Sshfs to Backblaze. Now I moved to iDrive e2 and all works perfectly.

I moved away from Hetzner Sshfs due to issues about backup integrity.

[humpty]

@p44 I've been procrastinating about moving away from BB. I get recurring failed backup notifications across my Cloudrons. I'll have to look into iDrive e2. Thanks for sharing.

[p44]

@humptydumpty I strongly suggest you to try iDrive e2, it seems to have better performances than BB... I also recently had backups fail with BB but with Idrive e2 everything is going well

[robi]

Another option would be to use a storage node, and either run another Cloudron with Minio, or just Minio as the generic S3 backup.

I run the former. It's also flawless.

[p44]

@robi Interesting, but for a backup I think is better to rely to outside service provider than a self hosted instance on Minio.

[humpty]

@p44 @robi that's not a bad idea, actually. I always disliked the fact that my backups are with a 3rd party, but thought it's "OK" since it's pass-protected. I think I'm going to go with a combo of the two, iDrive e2 for regular apps and Minio for the sensitive ones like Vaultwarden, Nextcloud, etc. I can't wait for the multi-backup destination update.

BTW, any idea if it's possible to "easily" run two Cloudrons on the same network? By easily, I mean not going through a complicated set up that's over my technical skills, or introduce another weak/fail point. I'd like to have a dedicated server in-house for Minio only instead of having it reside on one of my existing homeservers (have one at home and at work).

And I just remembered why I've been procrastinating, I've been WAITING for multi-backup destination

[p44]

@humptydumpty we all are waiting for multi-backup destination and, eventually, for Sshfs integrity check.

iDrive e2 is going very good, since 48hours installing that service.

A question for you: do you encrypt backups? If yes, Vaultwarden should be "safe" also on a 3rd party service...

[humpty]

@p44 Yes, definitely. I'm backing up all my apps to BB all pass-protected including VW. But, why take the extra risk when I can have it on-premise?

BTW, 48 hours is too short of a time to tell. Please let me know if you end up having a "failed backup" notification in coming weeks.

[p44]

@humptydumpty Yes, 48 hours is too short, let's see what will happens ... of course I'll inform in case of "failed backup".

[humpty]

I just noticed BB's pricing has changed to have a bare minimum $6/TB/mo charge. I wonder if they're deliberately causing the failed backups to kick me off the grandfathered plan

[p44]

@humptydumpty Or, more likely, they've limited resources for those with legacy plans... Who knows...

Anyway, by doing this they lose customers...

It's no longer the cost of $5 or $6 per TB, but the hidden cost of time to have to examine all the failed uploads that time out (and lately it was happening almost every day).

Failed backups, plus multiple file copies retained, they led me to change...

[p44]

@humptydumpty Check this post about Minio: https://forum.cloudron.io/topic/13844/minio-removing-the-interface-for-community-edition

[humpty]

@p44 well, that s*cks.

[robi]

@humptydumpty @p44 it won't make a difference for Cloudron backups.

[p44]

@humptydumpty After a week, more or less, I can confirm: with iDrive e2 no any failed backup...

[humpty]

@p44 Perfect Thanks for the follow-up.

[p44]

@humptydumpty hello, after about 20 days from last post, and about one month about use, I've a negative feedback to share about iDrive e2.

I experienced an unplanned outage on EU (Frankfurt) data center.

Service was down, yet their status page showed "All Operational."

No prior notice or updates during the outage. I had to contact support to even confirm the issue.

Support acknowledged the problem but provided no clear timeline or explanation.

As a new customer, this kind of lack of transparency and communication makes me question their reliability.

Of course, I'm looking for alternatives...

[humpty]

@p44 It seems to be a trend in the hosting world. I ran into the same issue with Contabo, but unlike iDrive, they kept denying the outage Thanks for sharing!

[p44]

@humptydumpty Thanks for your answer.

Wow... also with Contabo?