Hello,
For quite some time, I was stuck in ncp 1.20.0, and got some notifications saying that the system was updated to 1.20.0 (again). It felt weird to me after the 100th occurence of it. I logged into my Rpi, which tells me that 1.20.0 is outdated and that I should update to 1.20.7.
Well, it did not work. I got a similar message as here : Can't update from ncp 1.20.0 -> 1.20.1. So I followed @OliverV’s instruction (in that thread) to update to Buster.
Then, it started to get complicated for me.
I identified 2 issues in the process:
1/ php7.3 issue :
The following packages have unmet dependencies:
php7.3-bz2 : Depends: php7.3-common (= 7.3.14-1~deb10u1) but 7.3.14-5+0~20200202.52+debian9~1.gbpa71879 is to be installed
php7.3-curl : Depends: php7.3-common (= 7.3.14-1~deb10u1) but 7.3.14-5+0~20200202.52+debian9~1.gbpa71879 is to be installed
...
E: Unable to correct problems, you have held broken packages.
Something went wrong. Fix it and try again
<End of the ncp-dist-upgrade command>
2/ Apache server not starting
Apache log in ncp-dist-upgrade
Setting up apache2 (2.4.38-3+deb10u3) ...
Installing new version of config file /etc/apache2/mods-available/deflate.conf ...
Installing new version of config file /etc/apache2/mods-available/mime.conf ...
Installing new version of config file /etc/logrotate.d/apache2 ...
Job for apache2.service failed because the control process exited with error code.
See "systemctl status apache2.service" and "journalctl -xe" for details.
invoke-rc.d: initscript apache2, action "restart" failed.
â—Ź apache2.service - The Apache HTTP Server
Loaded: loaded (/lib/systemd/system/apache2.service; enabled; vendor preset: enabled)
Active: failed (Result: exit-code) since Thu 2020-02-20 20:28:12 UTC; 77ms ago
Docs: https://httpd.apache.org/docs/2.4/
Process: 27053 ExecStart=/usr/sbin/apachectl start (code=exited, status=1/FAILURE)
Feb 20 20:28:12 nextcloudpi apachectl[27053]: (98)Address already in use: AH00072: make_sock: could not bind to address [::]:80
Feb 20 20:28:12 nextcloudpi apachectl[27053]: (98)Address already in use: AH00072: make_sock: could not bind to address [::]:443
Feb 20 20:28:12 nextcloudpi apachectl[27053]: (98)Address already in use: AH00072: make_sock: could not bind to address 0.0.0.0:443
Feb 20 20:28:12 nextcloudpi apachectl[27053]: no listening sockets available, shutting down
Feb 20 20:28:12 nextcloudpi apachectl[27053]: AH00015: Unable to open logs
Feb 20 20:28:12 nextcloudpi apachectl[27053]: Action 'start' failed.
Feb 20 20:28:12 nextcloudpi apachectl[27053]: The Apache error log may have more information.
Feb 20 20:28:12 nextcloudpi systemd[1]: apache2.service: Control process exited, code=exited, status=1/FAILURE
Feb 20 20:28:12 nextcloudpi systemd[1]: apache2.service: Failed with result 'exit-code'.
Feb 20 20:28:12 nextcloudpi systemd[1]: Failed to start The Apache HTTP Server.
So (I think) I fixed the first problem by doing:
sudo apt-get -f remove php7.3-common
and then re-running sudo ncp-dist-upgrade
which reinstall php7.3 correctly (for deb10)
But the issue remains for Apache, with some message giving me hope like
NOTICE: Not enabling PHP 7.3 FPM by default.
NOTICE: To enable PHP 7.3 FPM in Apache2 do:
NOTICE: a2enmod proxy_fcgi setenvif
NOTICE: a2enconf php7.3-fpm
NOTICE: You are seeing this message because you have apache2 package installed.
which I did, but did not work.
I also try systemctl reload apache2
but did not work because apache was not active.
I dig up a bit on internet for why apache could not start with a could not bind to address
error, trying to see whether something else was using these http(s) ports. (e.g. https://stackoverflow.com/questions/9143964/make-sock-could-not-bind-to-address-443-when-restarting-apache-installi)
but I couldn’t find a duplicate Listen 80/443
in my conf file:
pi@nextcloudpi:~ $ grep -r Listen /etc/apache2/
/etc/apache2/ports.conf:Listen 80
/etc/apache2/ports.conf:Listen [::]:80
/etc/apache2/ports.conf: Listen 443
/etc/apache2/ports.conf: Listen [::]:443
/etc/apache2/ports.conf: Listen 443
/etc/apache2/ports.conf: Listen [::]:443
/etc/apache2/sites-available/ncp.conf:Listen 4443
and the 2 Listen 443
are because of <IfModule ssl_module>
or <IfModule mod_gnutls.c>
. I thought it was not a pb, but maybe it is ?
I was stuck there, not knowing what could I do, so since I saw in the output of the ncp-dist-upgrade a line with :
Setting up dbus (1.10.28-0+deb9u1) ...
A reboot is required to replace the running dbus-daemon.
Please reboot the system when convenient.
I thougt that it might me convenient by now, and in bonus, it might fix the apache issue.
so: sudo reboot
but then I couldn’t connect through ssh to the Rpi (connection timed out), and the Rpi was not detectable by the internet box. I thought maybe this is because it takes some time to replace the dbus stuff or whatever, so I let it like this overnight.
This morning, again, I could not ssh in. So I just unplug it and back in, and now it worked, I could ssh in directly.
But surprise, the HDD was not mounted anymore and not even detected. I unplug the usb cable and put it back but it did not change.
I check the syslog to see what was going on overnight, but 99% of it was:
Feb 21 07:16:41 nextcloudpi bash[1088]: Device "" does not exist.
Feb 21 07:16:44 nextcloudpi bash[1088]: Device "" does not exist.
Feb 21 07:16:48 nextcloudpi bash[1088]: Device "" does not exist.
And I also notice this line (which occured already before the reboot):
Feb 20 19:15:45 nextcloudpi kernel: [ 3.615346] usb-storage 1-1.5:1.0: USB Mass Storage device detected
Feb 20 19:15:45 nextcloudpi systemd[1]: Started Monitor /media for mountpoints and create USBdrive* symlinks.
Feb 20 19:15:45 nextcloudpi systemd[1]: Starting Automount USB drives...
Feb 20 19:15:45 nextcloudpi bash[338]: rmdir: failed to remove '/media/USBdrive': Not a directory
Feb 20 19:15:45 nextcloudpi systemd[1]: Started Automount USB drives.
This rmdir
kind of scared me.
I tried a sudo ncp-dist-upgrade
just in case it would magically fix everything. But it didn’t.
So I thought why not reboot ?
So I did, and now I’m back to having only the apache issue, the HDD is mounted again.
(see next post for output of ncp-report
)