Install Cloudron, it failed due Collectd not in repo....
-
I just found this forum post after happily following the documentation: https://docs.cloudron.io/guides/upgrade-ubuntu-22/ and ran into the problems afterwards.
There is no hint in the documentation about the
collectd
-missing problem or any suggestion to wait with upgrades to 22.04.For the moment being I installed collectd and collectd-core manually as was suggested in the referenced launchpad forum. Fingers crossed that this is a stable solution, because
a) after rebooting the machine collectd initially failedcollectd.service: Failed with result 'start-limit-hit'.
, manually restarting it with systemctl worked
b) installing the collectd packages with apt saidN: Download is performed unsandboxed as root as file '/root/collectd-core_5.12.0-9_amd64.deb' couldn't be accessed by user '_apt'. - pkgAcquire::Run (13: Permission denied)
.We'll see.
-
@girish I followed the notes exactly (on this day I updated from Ubuntu 18 to 20 to 22) and I'm impressed how well and smooth everything worked out in general (thanks for the documentation work).
It's just the
collectd
thing that should be updated at the moment up until there is some feedback from the maintainers. -
@nebulon Oh ok because that's what I'm attempting to do (intranet) but I'm stuck on this very collectd error
running ./cloudron-setup
echo "==> Install collectd" ==> Install collectd # without this, libnotify4 will install gnome-shell apt-get install -y libnotify4 libcurl3-gnutls --no-install-recommends Reading package lists... Building dependency tree... Reading state information... libcurl3-gnutls is already the newest version (7.81.0-1ubuntu1.16). libnotify4 is already the newest version (0.7.9-3ubuntu5.22.04.1). 0 upgraded, 0 newly installed, 0 to remove and 2 not upgraded. # https://bugs.launchpad.net/ubuntu/+source/collectd/+bug/1872281 if [[ "${ubuntu_version}" == "22.04" ]]; then readonly launchpad="https://launchpad.net/ubuntu/+source/collectd/5.12.0-9/+build/23189375/+files" cd /tmp && wget -q "${launchpad}/collectd_5.12.0-9_amd64.deb" "${launchpad}/collectd-utils_5.12.0-9_amd64.deb" "${launchpad}/collectd-core_5.12.0-9_amd64.deb" "${launchpad}/libcollectdclient1_5.12.0-9_amd64.deb" cd /tmp && apt install -y --no-install-recommends ./libcollectdclient1_5.12.0-9_amd64.deb ./collectd-core_5.12.0-9_amd64.deb ./collectd_5.12.0-9_amd64.deb ./collectd-utils_5.12.0-9_amd64.deb && rm -f /tmp/collectd_*.deb echo -e "\nLD_PRELOAD=/usr/lib/python3.10/config-3.10-x86_64-linux-gnu/libpython3.10.so" >> /etc/default/collectd else if ! apt-get install -y --no-install-recommends collectd collectd-utils; then # FQDNLookup is true in default debian config. The box code has a custom collectd.conf that fixes this echo "Failed to install collectd, continuing anyway. Presumably because of http://mailman.verplant.org/pipermail/collectd/2015-March/006491.html" fi if [[ "${ubuntu_version}" == "20.04" ]]; then echo -e "\nLD_PRELOAD=/usr/lib/python3.8/config-3.8-x86_64-linux-gnu/libpython3.8.so" >> /etc/default/collectd fi fi
I get stuck at this point when I launch the install
-
Damn, stupid me, in a corporate network setting the proxy for wget or curl was configured for the user but not for the root user, which the setup needs to perform the first install.
so after copying .wgetrc and .curlrc to /root/ the collectd step went just fine !
-
@nebulon I hit many road blocks indeed, but managed to move quite well forward
now i'm at the part of the npm rebuild and I can't do the db-migrate because I don't see any npm install being done that provides db-migrate within nodes_modules/.bin folderI have documented the changes I had to do in the different install files to get it working, i can share this to the support email if you're interested