Yes, we have that one. I’m putting final on that plus a couple more now.
This change was already staged but didn’t make the cutoff to be included.
I ask you to confirm if this will work for you,
# Independently, evaluate if in a LXC. This is used later in installation to avoid UDEV trigger bug
if [ "$(grep -i 'container=lxc' /proc/1/environ )" != "" ]; then
LinuxContainer=1
fi
That’s what’s already in the preinst file on my system. But /proc/1/environ contains container=systemd-nspawn, not lxc, so that doesn’t set LinuxContainer.
What does it set in /proc/1/cgroup ?
Is there something I can key on ?
11:blkio:/
10:devices:/
9:cpu,cpuacct:/
8:cpuset:/
7:pids:/
6:memory:/
5:rdma:/
4:freezer:/
3:net_cls,net_prio:/
2:perf_event:/
1:name=systemd:/init.scope
0::/init.scope
That sort of depends on what LinuxContainer is for. If it should match any container, I would presume that grep -zi "^container=" /proc/1/environ would be sufficient for any case. If instead this is a workaround for specific container types, then an else if [ "$(grep -i 'container=systemd-nspawn' /proc/1/environ )"... would solve the nspawn issue without impacting other runtimes.
While the grep might work for nspawn, it’s not a guarantee against false detection.
What I find more interesting is
1:name=systemd:/init.scope
Where one can get totally thrown off base is with Debian
I migrated to using /proc/1/cgroup so I don’t have to name-chase everyone’s docker executive.
It’s equally impossible to name-chase everyone’s LXC executive.
To your question about what LinuxContainer is for; The sole purpose of the variable is to detect and avoid getting failures from the udevadm bug. I was trying to be nice there. If grepping for container in environ is sufficiently guaranteed to identify a LXC, I’ll take it. If it’s not going to work as intended, I’ll suppress the error from trickling back up into the installation and causing issues.
PMS 1.18.6.2368-97add474d is available for download in the PlexPass downloads section.
It contains all the corrections and changes for issues presented here.
EDIT: Engineering rebuild: 1.18.6.2369
Can confirm this fixed the install error on docker. Thanks @ChuckPa! 
Get this error on a Ubuntu 18.04.4 LTS
any ideas?
sudo dpkg -i plexmediaserver_1.18.6.2368-97add474d_amd64.deb [sudo] password for plex: (Reading database ... 159543 files and directories currently installed.) Preparing to unpack plexmediaserver_1.18.6.2368-97add474d_amd64.deb ... Removed /etc/systemd/system/multi-user.target.wants/plexmediaserver.service. PlexMediaServer install: Pre-installation Validation. dpkg: error processing archive plexmediaserver_1.18.6.2368-97add474d_amd64.deb (--install): new plexmediaserver package pre-installation script subprocess returned error exit status 2 Created symlink /etc/systemd/system/multi-user.target.wants/plexmediaserver.service → /lib/systemd/system/plexmediaserver.service. Processing triggers for libc-bin (2.27-3ubuntu1) ... Errors were encountered while processing: plexmediaserver_1.18.6.2368-97add474d_amd64.deb
Can I see the contents of /tmp/plexinstaller.log ?
It will tell me what it was using as configuration parameters.
It’s the only information we have to go on since no errors were printed.
I’m trying to install 1.18.5.2309 and getting the following errors.
I have a systemd override conf file where I set a bunch of custom settings, and the preinst script seems to be getting confused by some of them. Here’s my override file in its entirety:
[Service]
ExecStartPre=
ExecStartPre=/bin/mkdir -p "${PLEX_MEDIA_SERVER_APPLICATION_SUPPORT_DIR}"
ExecStart=
Environment=LD_LIBRARY_PATH=/usr/lib/plexmediaserver/lib
Environment=PLEX_MEDIA_SERVER_INFO_VENDOR="Ubuntu"
Environment=PLEX_MEDIA_SERVER_INFO_DEVICE="PC"
Environment=PLEX_MEDIA_SERVER_INFO_MODEL="x86_64"
Environment=PLEX_MEDIA_SERVER_INFO_PLATFORM_VERSION="19.10 (Eoan Ermine)"
ExecStart="/usr/lib/plexmediaserver/Plex Media Server"
ProtectHome=true
PrivateTmp=true
ProtectControlGroups=true
ProtectKernelTunables=true
ProtectKernelModules=true
ProtectSystem=full
PrivateUsers=true
DeviceAllow=char-DVB rw
DeviceAllow=char-drm rw
DevicePolicy=closed
InaccessiblePaths=/mnt/games /mnt/backup /boot
SystemCallFilter=~@clock @cpu-emulation @debug @keyring @mount @obsolete @privileged @reboot @setuid @swap
SystemCallErrorNumber=EPERM
#SecureBits=noroot-locked
CapabilityBoundingSet=
User, it’s getting matched with PrivateUsers, gets the value of true, sees that there’s no such user, and complains.Group, it’s getting matched with PrivateControlGroups, gets the value of true, sees that there’s no such group, and complains.LD_LIBRARY_PATH setting there, but that’s because I like to keep the ExecStart field clean, and have environment variable settings go as separate entries in their own fields. I get the reasoning for the error, but any chance there can be an override setting to say “Don’t check my settings, I know what I’m doing”?Also, systemd shouldn’t have the “d” at the end capitalized. See this page.
Also, rather than manually parsing these files, I think systemctl can be used to get the values (presumably, if .service files are being installed, systemctl will exist). Something like systemctl show -p User plexmediaserver should work (hopefully in containers as well).
Which environment is this being used on ? (I see the ubuntu)
As for the SystemD , thanks for catching those typos. (there were 2)
I’m doing a lot of bullet-proofing now.
I think I see where the grep can be tightened as part of the regex to improve.
As FYI, hard setting the variable PMS picks up dynamically is against their purpose.
You also have LD_LIBRARY_PATH – It’s now illegal given RPATH is defined.
Confusion between PrivateUsers and Users, or any other wording combination has been corrected.
DorkUser is no longer considered.
I have Ubuntu 16.04 with init and I’m getting the following error:
dpkg -i plexmediaserver_1.18.6.2368-97add474d_amd64.deb
Selecting previously unselected package plexmediaserver.
(Reading database … 227526 files and directories currently installed.)
Preparing to unpack plexmediaserver_1.18.6.2368-97add474d_amd64.deb …
PlexMediaServer install: Pre-installation Validation.
PlexMediaServer install: Unsupported init/systemd hybrid control configuration.
dpkg: error processing archive plexmediaserver_1.18.6.2368-97add474d_amd64.deb (–install):
subprocess new pre-installation script returned error exit status 1
Processing triggers for libc-bin (2.23-0ubuntu11) …
Errors were encountered while processing:
plexmediaserver_1.18.6.2368-97add474d_amd64.deb
Is it init or is it upstart or was the system upgraded from Ubuntu 15 ?
Please go look at where pid 1’s executable is pointing. and what it’s value is.
I don’t support upstart anymore because Ubuntu dropped it in 2016.
it was a “one release wonder”
Here is the actual code to the test I perform:
# If this looks like init, be careful of redirectioon
elif [ "$(cat /proc/1/comm)" = "init" ]; then
# Verify not a redirected init -> systemd
if [ "$(readlink /sbin/init)" = "" ]; then
Init=1
elif [ $(readlink /sbin/init | grep systemd | wc -l) -gt 0 ]; then
Systemd=1
else
Output both "Unsupported init/systemd hybrid control configuration."
exit 1
fi
Bare-metal Ubuntu 19.10, with systemd.
Interesting; the .service file at /usr/lib/plexmediaserver/lib/plexmediaserver.service (I’m guessing this is the one that would get installed in a normal systemd installation case?) still has LD_LIBRARY_PATH being set. I did verify, though, that the binary itself has RPATH set, and so can start without that.
I am aware of the status of the service file, thank you.
There is a migration taking place here.
I am quite capable of handling the platform.
Hi @ChuckPa
Upgraded to latest PMS version and not able to remotely access the server.
Server is hosted in a VPS.
Running Native package
Upgrade installation worked but cannot find my server on Plex web.
Attempted to stop/start service & rebooted server several times but still nothing.
As with any remotely-hosted server.
SSH-tunnel to it.
ssh -L 8888:127.0.0.1:32400 ip.addr.of.remote
Sign in the ssh tunnel and let it sit idle.
Now open http://127.0.0.1:8888/web in your local browser
You might have gotten caught in the plex.tv outage of yesterday and need reclaim it.
Howdy
I actually have a standalone Microserver set up with just OMV and the OS coders of the software have a Plex Media server built into the software. You just have to enable it. but before you do that i suggest that you first install all off the needed file system plug-ins as well as set up your server certificates for the server. OMV is a very powerful OS and it is actually my first introduction into using and learning Linux. I have my server set up as a network file server for my entire home network and I was using another Media server software that I ended up hating and when i found OMV and saw that Plex was built into it I tried it and now 3 years later that server is still going strong and works well with the Plex Pass as well as Tidal on all of my streaming devices in the home as well as outside of my home network I am still able to access my media files for viewing. and way the OMV forum suggests that you don’t try to install updates that have not been released by the OMV team as the Plex software coding even though OMV is based on Debian the coding is different and you run the chance of crashing your system. OMV so far have been pretty good about keeping Plex current and up-to-date. Hope this info helped a little.
Rob
I get the following on Ubuntu 18.04:
sudo dpkg -i plex_.deb
(Reading database ... 437561 files and directories currently installed.)
Preparing to unpack plex_.deb ...
Removed /etc/systemd/system/multi-user.target.wants/plexmediaserver.service.
PlexMediaServer install: Pre-installation Validation.
dpkg: error processing archive plex_.deb (--install):
new plexmediaserver:i386 package pre-installation script subprocess returned error exit status 2
Created symlink /etc/systemd/system/multi-user.target.wants/plexmediaserver.service → /lib/systemd/system/plexmediaserver.service.
Errors were encountered while processing:
plex_.deb
No /tmp/plexmediaserver.log is created, only an empty folder named pms-0c1093b3-d36a-473a-b667-7190a202c704, with the UID on the end there changing on each installation.
Any ideas on other things to check?
