[pve-devel] trousers error in latest test repo?

[Victor Hooi]

Hi,

Nope, I haven't edited /etc/init.d/trousers. (I did view it, because of this
post
<https://unix.stackexchange.com/questions/633563/can-not-start-trousers-service-giving-error-trousers-ioctl-25-inappropriat>
and
curiosity but didn't make any changes).

I just tested a reinstall, as suggested - I still get an error:

root at mgmt1:/etc/apt/sources.list.d# apt install --reinstall trousers
> Reading package lists... Done
> Building dependency tree... Done
> Reading state information... Done
> 0 upgraded, 0 newly installed, 1 reinstalled, 0 to remove and 0 not
> upgraded.
> Need to get 137 kB of archives.
> After this operation, 0 B of additional disk space will be used.
> Get:1 http://download.proxmox.com/debian/pve bullseye/pvetest amd64
> trousers amd64 0.3.14+fixed1-1.2+pve1 [137 kB]
> Fetched 137 kB in 1s (115 kB/s)
> (Reading database ... 49780 files and directories currently installed.)
> Preparing to unpack .../trousers_0.3.14+fixed1-1.2+pve1_amd64.deb ...
> Unpacking trousers (0.3.14+fixed1-1.2+pve1) over (0.3.14+fixed1-1.2+pve1)
> ...
> Setting up trousers (0.3.14+fixed1-1.2+pve1) ...
> Job for trousers.service failed because the control process exited with
> error code.
> See "systemctl status trousers.service" and "journalctl -xe" for details.
> invoke-rc.d: initscript trousers, action "restart" failed.
> ● trousers.service - LSB: starts tcsd
>      Loaded: loaded (/etc/init.d/trousers; generated)
>      Active: failed (Result: exit-code) since Wed 2021-10-06 01:37:06 PDT;
> 24ms ago
>        Docs: man:systemd-sysv-generator(8)
>     Process: 199367 ExecStart=/etc/init.d/trousers start (code=exited,
> status=30)
>         CPU: 9ms


Oct 06 01:37:06 mgmt1 systemd[1]: Starting LSB: starts tcsd...
Oct 06 01:37:06 mgmt1 tcsd[199372]: TCSD TDDL[199372]: TrouSerS ioctl: (25)
Inappropriate ioctl for device
Oct 06 01:37:06 mgmt1 tcsd[199372]: TCSD TDDL[199372]: TrouSerS Falling
back to Read/Write device support.
Oct 06 01:37:06 mgmt1 tcsd[199372]: TCSD TCS[199372]: TrouSerS ERROR: TCS
GetCapability failed with result = 0x1e
Oct 06 01:37:06 mgmt1 trousers[199367]: Starting Trusted Computing daemon:
tcsd
Oct 06 01:37:06 mgmt1 trousers[199373]:  failed!
Oct 06 01:37:06 mgmt1 systemd[1]: trousers.service: Control process exited,
code=exited, status=30/n/a
Oct 06 01:37:06 mgmt1 systemd[1]: trousers.service: Failed with result
'exit-code'.
Oct 06 01:37:06 mgmt1 systemd[1]: Failed to start LSB: starts tcsd.
Processing triggers for man-db (2.9.4-2) ...

What do you mean by whether I have a node?

root at mgmt1:/etc/apt/sources.list.d# ls -l /dev/tmp*
> ls: cannot access '/dev/tmp*': No such file or directory


Thanks,
Victor

On Wed, 6 Oct 2021 at 19:32, Thomas Lamprecht <t.lamprecht at proxmox.com>
wrote:

> Hi Victor,
>
> On 06.10.21 10:15, Victor Hooi wrote:
> > Ah - interesting - as a data-point, this was a new install of Proxmox
> 7.0-2
> > (from the ISO image), and then I ran an apt update, apt dist-upgrade
> > directly after first boot.
>
> very interesting indeed, I mainly tested on a cluster that was installed
> with PVE
> 6.2 and a development VM that was originally a Bullseye pre-release ISO,
> it worked
> out OK on both.
>
> >
> > I just did an apt update, apt dist-upgrade again - and there was one
> > package to update (trousers).
> >
> > I am still getting an error with this version:
>
> Hmm first I thought that could be the "stop-before-upgrade" of the old,
> broken
> version, but that makes not much sense as your error is on a "start".
> Did you edit `/etc/init.d/trousers` before the upgrade?
>
> And could you please still try `apt install --reinstall trousers` to
> verify if
> it was that.
>
> >
> > Unpacking trousers (0.3.14+fixed1-1.2+pve1) over (0.3.14+fixed1-1.2) ...
> >> Setting up trousers (0.3.14+fixed1-1.2+pve1) ...
> >> Installing new version of config file /etc/init.d/trousers ...
> >> Job for trousers.service failed because the control process exited with
> >> error code.
> >> See "systemctl status trousers.service" and "journalctl -xe" for
> details.
> >> invoke-rc.d: initscript trousers, action "start" failed.
> >> ● trousers.service - LSB: starts tcsd
> >>      Loaded: loaded (/etc/init.d/trousers; generated)
> >>      Active: failed (Result: exit-code) since Wed 2021-10-06 01:13:08
> PDT;
> >> 4ms ago
> >>        Docs: man:systemd-sysv-generator(8)
> >>     Process: 192905 ExecStart=/etc/init.d/trousers start (code=exited,
> >> status=30)
> >>         CPU: 9ms
> >
> > Oct 06 01:13:08 mgmt1 systemd[1]: Starting LSB: starts tcsd...
> >> Oct 06 01:13:08 mgmt1 tcsd[192910]: TCSD TDDL[192910]: TrouSerS ioctl:
> >> (25) Inappropriate ioctl for device
> >> Oct 06 01:13:08 mgmt1 tcsd[192910]: TCSD TDDL[192910]: TrouSerS Falling
> >> back to Read/Write device support.
> >> Oct 06 01:13:08 mgmt1 tcsd[192910]: TCSD TCS[192910]: TrouSerS ERROR:
> TCS
> >> GetCapability failed with result = 0x1e
> >> Oct 06 01:13:08 mgmt1 trousers[192905]: Starting Trusted Computing
> daemon:
> >> tcsd
> >> Oct 06 01:13:08 mgmt1 trousers[192911]:  failed!
> >> Oct 06 01:13:08 mgmt1 systemd[1]: trousers.service: Control process
> >> exited, code=exited, status=30/n/a
> >> Oct 06 01:13:08 mgmt1 systemd[1]: trousers.service: Failed with result
> >> 'exit-code'.
> >> Oct 06 01:13:08 mgmt1 systemd[1]: Failed to start LSB: starts tcsd.
> >
> > I should note that the spec sheet for the NUC7i5BNK says it does not
> have a
> > TPM - would that be related at all to these errors?
>
> Well yes, that was an issue with the old version, it checked that in a
> wrong way.
> But the newer, just uploaded package adapted that part, which was also
> tested by
> creating and dropping a fake /deb/tmp1 node, so it should work out.
>
> Or do you have a node?
>
> ls -l /dev/tmp*
>
> cheers,
> Thomas
>
>
>