Auto Upgrade NextCloud Pi to Nextcloud 21.0.4

ncp_upgrade_nc_21.0.4_failure_analysis.md

Auto Upgrade NextCloud Pi to Nextcloud 21.0.4

Hi,

the auto-upgrade to nc 21.0.4 failed and left me with stall nextcloud.

I'm running on Raspberry Pi 4 (4GB) with docker. Current Nextcloud version 20.0.12.1 Available Nextcloud version 21.0.4

Here my analysis commenting log (/var/log/ncp.log):

...
notify_push 0.2.4 installed  **--> line 200 of ncp-update-nc**
notify_push enabled
notify_push already enabled
/usr/local/etc/ncp-templates/nextcloud.conf.sh: line 21: /usr/local/bin/ncp/SYSTEM/metrics.sh: No such file or directory
/usr/local/etc/ncp-templates/nextcloud.conf.sh: line 22: tmpl_metrics_enabled: command not found

--> line 202: for those 2 errors above I have no idea lines do not match with error and it should not have been run as I'm on docker.

--> as we can see below there is an issue with the generated nextcloud.conf file it took the wrong domain.

--> reson for it, I use self generated letsencrypt certificate (DNS method), but it considered active the letsencrypt configuration, but it is not:

--> letsencrypt.cfg has: "name": "Active", value": "no"

Apache self check:
AH00526: Syntax error on line 11 of /etc/apache2/sites-enabled/nextcloud.conf:
SSLCertificateFile: file '/etc/letsencrypt/live/mycloud.ownyourbits.com/fullchain.pem' does not exist or is empty
Action '-t' failed.
The Apache error log may have more information.

--> Apache commands failed as it re-gernerate nextcloud.conf with the wrong domain name in the path.

Abort --> rollback is called line 158.
Clean up...
Rolling back to backup /var/www/nextcloud-bkp_20210917_1631854237-20.0.12.1.tar.gz...
{"reqId":"D8J5HwD3WWHHlUizVMyS","level":2,"time":"2021-09-17T07:44:49+02:00","remoteAddr":"","user":"--","app":"no app in context","method":"","url":"--","messa
ge":"Temporary directory /var/www/nextcloud/data/tmp is not present or writable","userAgent":"--","version":"21.0.4.1"}
Can only restore from ext/btrfs filesystems

--> line 167: ncp-restore fails on check for "ext/btrfs filesystems" as on docker it is:

--> command "stat -fc%T ncp-restore.loGwXx" returns: overlayfs

Rollback failed! Data left at /var/www/ncp-data.Ia9BQq

--> game over !

Additional findings:

stat -fc%T /data
ext2/ext3

root@ad7dda4836ab:/var/www# ls -l /data
total 136940
drwxr-xr-x  3 root     root          4096 Dec 11  2019 backups
drwxr-xr-x  3 root     root          4096 Apr  6  2020 bin
drwxr-xr-x  5 mysql    mysql         4096 Aug 11 18:10 database
drwxr-xr-x  9 root     root          4096 Dec 10  2019 etc
drwxr-xr-x  3 root     www-data      4096 Sep 18 14:18 ncp
drwxr-x--- 14 www-data www-data      4096 Sep 17 07:44 nextcloud
-rw-r--r--  1 root     root     140200496 Aug  3 17:46 nextcloud.tar.bz2

--> maybe the upgrade did not run in the right location ? (above we see an old nextcould software archive)