Re: [DNG] Cockpit removal might make sense
On 5/26/21 5:23 PM, Mark Hindley wrote:
On Wed, May 26, 2021 at 04:23:56PM +0900, Simon Walter wrote:
On 5/26/21 12:37 AM, Mark Hindley wrote:
This is actually a Debian bug and should be fixed there.
If cockpit requires systemd, is should declare that dependency.
AFAICT, it requires a systemd socket. It doesn't require any systemd
packages (debs). Would that still be considered a dependency WRT to
packages?
I am no systemd expert ;) and have little first hand knowledge of it. But my
limited understanding is that systemd socket activation is the systemd
absorption/reimplementation of inetd(8) and requires systemd itself to be
running.
Cockpit uses a systemd socket get started on demand. The socket is not
needed for the cockpit-ws cockpit-bridge etc to run. However, I am told
that the version used in Beowulf is old and a newer version will fail to
work w/o systemd. So, I guess lets see how far it makes it on Devuan. It
may become totally useless, and then I'll file a bug upstream. You can
see my bug report here:
https://www.mail-archive.com/debian-bugs-dist@lists.debian.org/msg1806544.html
In an email not on bug tracker for some reason:
"More recent cockpit versions have a much more complex unit with
cockpit.{socket,service} controlling the cockpit-tls component, and that in
turn launches per-client-certificate cockpit-ws instances (as a separate
user)
through systemd socket activation. So this init script does not apply
at all
to current versions, and there is no reasonable way how to write one.
So in summary, there is no way of running cockpit in a non-systemd/Linux
environment that I'd be willing to support. For these I'd rather recommend
looking at webmin, ebox, or similar project."
Best regards,
Simon
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 5/26/21 12:53 AM, Rowland penny via Dng wrote: Why would you want to remove something that works ? It just needs an init script. I've been shown by Rowland that a lot of it does work without systemd and, yes, an init script is needed, which I've submitted upstream, thanks to Rowland. Thanks everyone for your feedback. Simon ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
Hi Mark (and everyone else on the list!), Mark Hindley writes: > On Tue, May 25, 2021 at 07:03:16PM +, g4sra via Dng wrote: >> Is this the rule for all packages that have dependency on SystemD for some >> functionality ? > > My personal opinion on this is yes. Add mine! > Non systemd init is still possible in Debian. So it appears some > current versions of cockpit will not work on a Debian system with > sysvinit, runit or openrc. To my mind, that makes it a Debian bug that > should be fixed there with an explicit dependency. As long as systemd is not `Essential: yes` (or `Priority: required`), anything that doesn't work without it needs to add a `Depends:`. But you already mentioned that below :-) FTR, for buster, bullseye and sid and as of writing, systemd is `Priority: important`, see https://sources.debian.org/src/systemd/247.3-5/debian/control/ (for bullseye and sid). >> I was under the impression that missing dependencies from Debian >> packages was to be expected, and that it was not considered >> 'important' by some of the Debian devs. That was a while ago though >> (Lenny ?)...maybe opinions have changed. > > Again, my take on this is that In Debian, systemd is the default, but it is > still not Essential in the Debian Policy technical sense.[1] Therefore > packages > (in this case cockpit) must explicitly depend on packages (in this case > systemd > or systemd-sysv) they require to function[2]. Debian Policy mandates this as a > 'must', so not complying should be an RC bug. Seeing that bullseye is in freeze, reporting this as an RC bug might actually get it fixed :-P ... On the other hand, you might just be perceived as a jerk/nitpick ... and get your bug downgraded because the maintainer wants to have it included in bullseye anyway. > However, as we have seen, some DDs are very quick to quote the Policy when it > suits them and will ignore it when it is inconvenient... Like I said :-) > [1] > https://www.debian.org/doc/debian-policy/ch-binary.html#essential-packages > > [2] https://www.debian.org/doc/debian-policy/ch-binary.html#dependencies Hope this helps, -- Olaf Meeuwissen, LPIC-2FSF Associate Member since 2004-01-27 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9 Support Free Softwarehttps://my.fsf.org/donate Join the Free Software Foundation https://my.fsf.org/join ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On Wed, May 26, 2021 at 04:23:56PM +0900, Simon Walter wrote: > On 5/26/21 12:37 AM, Mark Hindley wrote: > > This is actually a Debian bug and should be fixed there. > > > > If cockpit requires systemd, is should declare that dependency. > > AFAICT, it requires a systemd socket. It doesn't require any systemd > packages (debs). Would that still be considered a dependency WRT to > packages? I am no systemd expert ;) and have little first hand knowledge of it. But my limited understanding is that systemd socket activation is the systemd absorption/reimplementation of inetd(8) and requires systemd itself to be running. Mark ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 5/26/21 12:37 AM, Mark Hindley wrote: This is actually a Debian bug and should be fixed there. If cockpit requires systemd, is should declare that dependency. AFAICT, it requires a systemd socket. It doesn't require any systemd packages (debs). Would that still be considered a dependency WRT to packages? If the dependency were present, amprolla would exclude cockpit from the Devuan archives. So I suggest you submit a bug to Debian's BTS asking for the explicit systemd dependency to be added. If it turns out to be the case, I will do so. Thanks, Simon ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 5/26/21 1:25 AM, Rowland penny via Dng wrote: On 25/05/2021 17:09, Tomasz Torcz wrote: ... Have you seen Cockpit working on Devuan system? Yes, I had it running on my old Samba AD DC's and I now have it running on a Devuan Unix domain member on my way to installing the Samba AD DC module. ... I am really interested in how you were able to do this. I've installed it on a fresh Beowulf installation and would appreciate some guidance/hints in setting it up without a systemd socket. Best regards, Simon ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On Tue, May 25, 2021 at 07:03:16PM +, g4sra via Dng wrote: > Is this the rule for all packages that have dependency on SystemD for some > functionality ? My personal opinion on this is yes. Non systemd init is still possible in Debian. So it appears some current versions of cockpit will not work on a Debian system with sysvinit, runit or openrc. To my mind, that makes it a Debian bug that should be fixed there with an explicit dependency. > I was under the impression that missing dependencies from Debian packages was > to be expected, and that it was not considered 'important' by some of the > Debian devs. > That was a while ago though (Lenny ?)...maybe opinions have changed. Again, my take on this is that In Debian, systemd is the default, but it is still not Essential in the Debian Policy technical sense.[1] Therefore packages (in this case cockpit) must explicitly depend on packages (in this case systemd or systemd-sysv) they require to function[2]. Debian Policy mandates this as a 'must', so not complying should be an RC bug. However, as we have seen, some DDs are very quick to quote the Policy when it suits them and will ignore it when it is inconvenient... Mark [1] https://www.debian.org/doc/debian-policy/ch-binary.html#essential-packages [2] https://www.debian.org/doc/debian-policy/ch-binary.html#dependencies ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
Thanks for chirping in Mark. ‐‐‐ Original Message ‐‐‐ On Tuesday, May 25, 2021 4:37 PM, Mark Hindley wrote: > This is actually a Debian bug and should be fixed there. > > If cockpit requires systemd, is should declare that dependency. > > If the dependency were present, amprolla would exclude cockpit from the Devuan > archives. > > So I suggest you submit a bug to Debian's BTS asking for the explicit systemd > dependency to be added. Is this the rule for all packages that have dependency on SystemD for some functionality ? I was under the impression that missing dependencies from Debian packages was to be expected, and that it was not considered 'important' by some of the Debian devs. That was a while ago though (Lenny ?)...maybe opinions have changed. > > If we want/need cockpit to work in Devuan that is a whole new problem ;) publickey - g4sra@protonmail.com - 0x42E94623.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 25/05/2021 17:09, Tomasz Torcz wrote: Dnia Tue, May 25, 2021 at 04:53:29PM +0100, Rowland penny via Dng napisał(a): After installing it on an fresh Beowulf installation, it does not run and to my knowledge will never be able to w/o systemd sockets. IIRC, the policy is not to remove anything related, but use stubs and let the user deal with half-broken software (ie. GNOME). Cockpit doesn't (currently) have dependencies on systemd and it's modules, but it requires a systemd socket to function. So, AFAICT, it is not even partially usable. Why would you want to remove something that works ? Have you seen Cockpit working on Devuan system? Yes, I had it running on my old Samba AD DC's and I now have it running on a Devuan Unix domain member on my way to installing the Samba AD DC module. There are minor problems, red-hat seems to think the only way to join to a domain is via realmd and sssd and you cannot use either with Samba >= 4.8.0 It just needs an init script. It's more than that. Cockpit uses systemd's API not only to listen on network socket, but also to manage services and other stuff. Just starting Cockpit without systemd requires helpers like https://github.com/cockpit-project/cockpit/commit/777c59095af6147af487bf6a5aa76b915b2463d6 It probably not worth (or even feasible) reimplemnting all those APIs for Cockpit. You might be correct, but it works for what I require 😁 Rowland ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
Dnia Tue, May 25, 2021 at 04:53:29PM +0100, Rowland penny via Dng napisał(a): > > After installing it on an fresh Beowulf installation, it does not run > > and to my knowledge will never be able to w/o systemd sockets. > > > > IIRC, the policy is not to remove anything related, but use stubs and > > let the user deal with half-broken software (ie. GNOME). Cockpit doesn't > > (currently) have dependencies on systemd and it's modules, but it > > requires a systemd socket to function. So, AFAICT, it is not even > > partially usable. > > > > Why would you want to remove something that works ? Have you seen Cockpit working on Devuan system? > It just needs an init script. It's more than that. Cockpit uses systemd's API not only to listen on network socket, but also to manage services and other stuff. Just starting Cockpit without systemd requires helpers like https://github.com/cockpit-project/cockpit/commit/777c59095af6147af487bf6a5aa76b915b2463d6 It probably not worth (or even feasible) reimplemnting all those APIs for Cockpit. -- Tomasz TorczOnly gods can safely risk perfection, to...@pipebreaker.pl it's a dangerous thing for a man. — Alia ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 25/05/2021 05:42, Simon Walter wrote: Hi all, I wanted to see if Cockpit would by some unknown magic run on Devuan. The reason I wanted to do this is because the packages are available in the default repos. After installing it on an fresh Beowulf installation, it does not run and to my knowledge will never be able to w/o systemd sockets. So, may I suggest that it is removed from the default repos, as it is misleading for beginners and may lead to unnecessary bug reports. IIRC, the policy is not to remove anything related, but use stubs and let the user deal with half-broken software (ie. GNOME). Cockpit doesn't (currently) have dependencies on systemd and it's modules, but it requires a systemd socket to function. So, AFAICT, it is not even partially usable. Best regards, Simon ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng Why would you want to remove something that works ? It just needs an init script. Rowland ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
This is actually a Debian bug and should be fixed there. If cockpit requires systemd, is should declare that dependency. If the dependency were present, amprolla would exclude cockpit from the Devuan archives. So I suggest you submit a bug to Debian's BTS asking for the explicit systemd dependency to be added. If we want/need cockpit to work in Devuan that is a whole new problem ;) Thanks Mark ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On 25-05-2021 17:22, g4sra via Dng wrote: > <--snip--> > >>> I sympathise, but am not sure this would (currently) be the best use of >>> resources. >>> There is an ever growing list of packaged applications that do not function >>> without SystemD. >>> If every one were identified, then removal of all from the repos may be >>> justified. >>> Who is going to volunteer to be the Devuan developer that continuously >>> performs this duty ? >>> I see this as being the work for a downstream distribution, to take the >>> freedom that Devuan provides and tailor it (reducing choice) for a specific >>> target use. >> There should be soe way of warning potential users once it has been >> discovered that a package is useless. We're wasting their time >> otherwise. > Are you volunteering to collate and maintain a list of SystemD dependant > application packages ? > If so, I am sure some Users would be grateful and Devuan could add it > alongside the release notes. > > Devuan is not a distribution for 'newbie's, there is a benefit from not > stripping them out. Anyone who is really passionate about a particular broken > package will either attempt a fix and submit a patch upstream or pay someone > else to do so. > Maintaining such list is not feasible for one person. Personally I was thinking of a wiki page with packages broken by systemd with version number and date. Grtz. Nick signature.asc Description: OpenPGP digital signature ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
<--snip--> > > I sympathise, but am not sure this would (currently) be the best use of > > resources. > > There is an ever growing list of packaged applications that do not function > > without SystemD. > > If every one were identified, then removal of all from the repos may be > > justified. > > Who is going to volunteer to be the Devuan developer that continuously > > performs this duty ? > > I see this as being the work for a downstream distribution, to take the > > freedom that Devuan provides and tailor it (reducing choice) for a specific > > target use. > > There should be soe way of warning potential users once it has been > discovered that a package is useless. We're wasting their time > otherwise. Are you volunteering to collate and maintain a list of SystemD dependant application packages ? If so, I am sure some Users would be grateful and Devuan could add it alongside the release notes. Devuan is not a distribution for 'newbie's, there is a benefit from not stripping them out. Anyone who is really passionate about a particular broken package will either attempt a fix and submit a patch upstream or pay someone else to do so. publickey - g4sra@protonmail.com - 0x42E94623.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
On Tue, May 25, 2021 at 11:45:49AM +, g4sra via Dng wrote: > ‐‐‐ Original Message ‐‐‐ > On Tuesday, May 25, 2021 5:42 AM, Simon Walter wrote: > > > Hi all, > > > > > I wanted to see if Cockpit would by some unknown magic run on Devuan. > > The reason I wanted to do this is because the packages are available in > > the default repos. > > > > > After installing it on an fresh Beowulf installation, it does not run > > and to my knowledge will never be able to w/o systemd sockets. > > > > > So, may I suggest that it is removed from the default repos, as it is > > misleading for beginners and may lead to unnecessary bug reports. > > > > > IIRC, the policy is not to remove anything related, but use stubs and > > let the user deal with half-broken software (ie. GNOME). Cockpit doesn't > > (currently) have dependencies on systemd and it's modules, but it > > requires a systemd socket to function. So, AFAICT, it is not even > > partially usable. > > > > > Best regards, > > > > > Simon > > > > > I sympathise, but am not sure this would (currently) be the best use of > resources. > There is an ever growing list of packaged applications that do not function > without SystemD. > If every one were identified, then removal of all from the repos may be > justified. > Who is going to volunteer to be the Devuan developer that continuously > performs this duty ? > > I see this as being the work for a downstream distribution, to take the > freedom that Devuan provides and tailor it (reducing choice) for a specific > target use. There should be soe way of warning potential users once it has been discovered that a package is useless. We're wasting their time otherwise. -- hendrik ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
Re: [DNG] Cockpit removal might make sense
‐‐‐ Original Message ‐‐‐ On Tuesday, May 25, 2021 5:42 AM, Simon Walter wrote: > Hi all, > > I wanted to see if Cockpit would by some unknown magic run on Devuan. > The reason I wanted to do this is because the packages are available in > the default repos. > > After installing it on an fresh Beowulf installation, it does not run > and to my knowledge will never be able to w/o systemd sockets. > > So, may I suggest that it is removed from the default repos, as it is > misleading for beginners and may lead to unnecessary bug reports. > > IIRC, the policy is not to remove anything related, but use stubs and > let the user deal with half-broken software (ie. GNOME). Cockpit doesn't > (currently) have dependencies on systemd and it's modules, but it > requires a systemd socket to function. So, AFAICT, it is not even > partially usable. > > Best regards, > > Simon > I sympathise, but am not sure this would (currently) be the best use of resources. There is an ever growing list of packaged applications that do not function without SystemD. If every one were identified, then removal of all from the repos may be justified. Who is going to volunteer to be the Devuan developer that continuously performs this duty ? I see this as being the work for a downstream distribution, to take the freedom that Devuan provides and tailor it (reducing choice) for a specific target use. publickey - g4sra@protonmail.com - 0x42E94623.asc Description: application/pgp-keys signature.asc Description: OpenPGP digital signature ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
[DNG] Cockpit removal might make sense
Hi all, I wanted to see if Cockpit would by some unknown magic run on Devuan. The reason I wanted to do this is because the packages are available in the default repos. After installing it on an fresh Beowulf installation, it does not run and to my knowledge will never be able to w/o systemd sockets. So, may I suggest that it is removed from the default repos, as it is misleading for beginners and may lead to unnecessary bug reports. IIRC, the policy is not to remove anything related, but use stubs and let the user deal with half-broken software (ie. GNOME). Cockpit doesn't (currently) have dependencies on systemd and it's modules, but it requires a systemd socket to function. So, AFAICT, it is not even partially usable. Best regards, Simon ___ Dng mailing list Dng@lists.dyne.org https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng
