@girish I'l disable auto-update. The version that works is Ghost 2.26.0 (Package v2.25.0). The version that wasn't working labelled only as Package v2.26.0 in the Backups dropdown. This looks weird since the working version is 2.26.0, but I guess there is some difference if the v is included. I recall it may have been Ghost 2.28.0, even though the latest version on https://github.com/TryGhost/Ghost/releases is 2.28.1.

@girish Ok, I've restored from a backup running 2.25.0 and things are running fine. I had already done that once, but Cloudron decided to auto-update it again and it was borked. I've re-restored and everything is fine. Thank goodness for thos backups!

PS: How do I earn 1 reputation? This would allow me to give responses on the Forum quicker.

@vjvanjungg Hi. No, I haven't run into this. There has been only one other instance of a problem, but I can't recall what it was, but I believe it was related to Ghost versions upgrading. FWIW, the new native Ghost install and update method was super easy and convenient when I was running it on a separate VPS, but Cloudron is just so amazing in so many ways I decided to export my Ghost blogs and import them to my Cloudron VPS and host them here.

So during an update of Ghost there seems to be a failure with the db. Here is what I think is the pertinent section from the logs:

Migrating database
INFO Finished database init!
INFO Creating database backup
INFO Database backup written to: /app/data/content/data/website.ghost.2019-08-18-17-46-12.json
ERROR Cannot read property 'schema' of undefined

Cannot read property 'schema' of undefined
"OuterError: insert into `integrations` (`created_at`, `created_by`, `description`, `icon_image`, `id`, `name`, `slug`, `type`, `updated_at`, `updated_by`) values ('2019-08-18 17:46:12', '1', 'Internal DB Backup integration', NULL, 'somekey', 'Ghost Backup', 'ghost-backup', 'internal', '2019-08-18 17:46:12', '1') - ER_BAD_FIELD_ERROR: Unknown column 'type' in 'field list'"

Or is the problem some media that's been uploaded??
Thank you for any help.

@girish For closure, this worked. Thank you!

So, I had to reinstall the OS on my VPS, and thought, 'no prob. I have my cloudron back upped to Exoscale.' The restore process goes well until "Looking to Restore". I enter all teh details and suddenly, this error" The bucket you are attempting to access must be addressed using the specified endpoint. Please send all future requests to this endpoint.
What does this mean? Have I lost my Cloudron??
Thank you.

@subven said in "socket hang up" when trying to display Memory Graph in Dashboard, no graphs displayed:

docker update --memory "256mb" --memory-swap "1g" graphite

There is no need to restart anything after that?
Thank you so much to the help!

@subven said in "socket hang up" when trying to display Memory Graph in Dashboard, no graphs displayed:

docker stats graphite

Thank you for sharing this info. I see that my MEM USAGE was 72.6 / 75 MB limit, so that would indicate why it crashes. How do I increase the limit then?

Why would this happen? I have a VPS with 16 GB of RAM!

Is it because ssdnodes uses the XFS file system, and not ext4? Remember, the only way I could get Cloudron to install was to comment out the check for ext4 in the install script.

This is just one of a few issues that really bug me, and I wish I knew if it was something to do with XFS or something entirely different. I mean, 16 GB.... why would there be any problems running cloudron????

Thank you for looking into this.


Much thanks to @girish for helping me figure out what was wrong. Sendy (sendy.co) is working well in the LAMP app. Yay! It was simply an issue of changing the
$dbHost = 'localhost';

to

$dbHost = 'mysql';