Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
On Sun, 2021-07-18 at 04:31 -0700, Craig Small wrote: > Which makes zero sense for procps which has no daemon. The restart action also makes zero sense for things that have no daemon. In the procps case, the Linux kernel is the software that is being restarted/reloaded, so reload actually makes *more* sense than restart, since restarting Linux means reboot or kexec but the procps init script doesn't do either of those things. > Why would you change from restart to reload in this case? I didn't change anything, procps changed to drop the reload action. -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 2021-07-16 at 02:49, p...@debian.org wrote: > Can you elaborate on what you mean by "Using reload is very wrong"? The scripts should be deliberate and specific. When you use reload you are saying "don't stop the daemon but do what you do for restart". So its changed from a mandatory has to be there restart to an optional reload. Which makes zero sense for procps which has no daemon. Why would you change from restart to reload in this case? - Craig -BEGIN PGP SIGNATURE- Version: FlowCrypt Email Encryption 8.1.0 Comment: Seamlessly send and receive encrypted email wsFzBAEBCgAGBQJg9BEiACEJEAIhZsD/PITjFiEEXT3w9TizJ8CqeneiAiFm wP88hOOqzQ//f9bBJkzedM0P9QV9ar7jJeRC+IhOgj6Z/iwDT5SDuqdHD6Ao c/jRIwxHz998m7/Z5Y9mk77Q90deVJYBtrANxGEuAcwqZAkpCbEVpTiR4y1H D2WC6U7zwsVxznVGn30ZaIusmpMoBaYEN2zNfgPOCuB4ZVgiDhpZxFkvvUbV e3nqxBPFSOC6J5sAzD7rGWbAqePXhys60M24whigpZmMby3Z0XmMERF1isFd vD/u+7AMItiXBtEG80aX23zdtNtNDzgkDB2K5aXJzthWMkQ7J6F1QCB3iDxx 7ayGjHAQue+ccE8zzoam32JISiHPgG0Pk2dC54RbOpygPejQgzzJKw8ew4k5 m9gJ6DG7f0WRzAO9GlV1XiDtcRH7d+XkfiA/7x7qRS5hNlJWOCJiO+guRtHq +qvXCCcmYK4c67V57V6W41q0WlrTqgzujz6pcCxKeU7OBaeka0omTEfqzLaH iGf/Ttn7xftkBWQWwlcJiylFakTTwDAUfqv5Zb7y2FvBhBBhI4FZMsSy7n7C Wlw9eNJ9FsrUc6Kk6Rh0c3hv40OAyn5gBQ/APMRhr8Hfxj6ILPg2gIatr6E8 VSMhyNdFAvCc2hq2tB//+qAnyw4XMfWVxU3VI6vm22zUognRDLrVEg9Wv2tw 3CXCS2PB2tI+7EiNdGLfQUqVrW2Yw2zLR7I= =uH4u -END PGP SIGNATURE- 0x3938F96BDF50FEA5.asc Description: application/pgp-keys
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
On Fri, 2021-07-16 at 12:41 +1000, Craig Small wrote: > I can add an alias easily enough. Using reload is very wrong so > corekeeper do the right thing but it's a one line change for procps. Can you elaborate on what you mean by "Using reload is very wrong"? -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
I can add an alias easily enough. Using reload is very wrong so corekeeper do the right thing but it's a one line change for procps. - Craig On Fri, 16 Jul 2021, 12:31 Paul Wise, wrote: > On Fri, 2021-07-16 at 02:25 +, Thorsten Glaser wrote: > > > … this isn’t right. This is an RC bug in corekeeper but nōn-RC > > in procps because of Policy §9.3.2: > > I still think it is RC as it is a feature regression breaking install > of reverse dependencies in supported configurations (sysvinit). > > > So I think it’d be better to clone the bugreport, asking procps nicely > > to implement “reload” while fixing corekeeper for bullseye first. > > If the procps maintainer doesn't plan to fix this in bullseye and > buster, then I guess I will have to workaround it in corekeeper. > > -- > bye, > pabs > > https://wiki.debian.org/PaulWise >
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
On Fri, 2021-07-16 at 02:25 +, Thorsten Glaser wrote: > … this isn’t right. This is an RC bug in corekeeper but nōn-RC > in procps because of Policy §9.3.2: I still think it is RC as it is a feature regression breaking install of reverse dependencies in supported configurations (sysvinit). > So I think it’d be better to clone the bugreport, asking procps nicely > to implement “reload” while fixing corekeeper for bullseye first. If the procps maintainer doesn't plan to fix this in bullseye and buster, then I guess I will have to workaround it in corekeeper. -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
Paul Wise dixit: >> Yes, the procps init script does not have the action reload. > >Looks like this is a regression in procps in buster and later. Hrm. OK, but… >I've bounced the thread to the procps maintainer and reassigned. … this isn’t right. This is an RC bug in corekeeper but nōn-RC in procps because of Policy §9.3.2: | The "start", "stop", "restart", and "force-reload" options should be | supported by all init scripts. Supporting "status" is encouraged. The | "reload" and "try-restart" options are optional. So I think it’d be better to clone the bugreport, asking procps nicely to implement “reload” while fixing corekeeper for bullseye first. bye, //mirabilos PS: I was unclear in the previous mail… I found the bug while crossgrading but was actually installing corekeeper on my systems recently; apt insisted on remove+install which is what triggered this. -- you introduced a merge commit│ % g rebase -i HEAD^^ sorry, no idea and rebasing just fscked │ Segmentation should have cloned into a clean repo │ fault (core dumped) if I rebase that now, it's really ugh │ wuahh
Processed: Re: Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
Processing control commands: > reassign -1 procps 3.3.15-2 Bug #991151 [corekeeper] corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed. Bug reassigned from package 'corekeeper' to 'procps'. No longer marked as found in versions corekeeper/1.7. Ignoring request to alter fixed versions of bug #991151 to the same values previously set Bug #991151 [procps] corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed. There is no source info for the package 'procps' at version '3.3.15-2' with architecture '' Unable to make a source version for version '3.3.15-2' Marked as found in versions 3.3.15-2. > retitle -1 procps: dropped the reload option from the init script, breaking > corekeeper Bug #991151 [procps] corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed. Changed Bug title to 'procps: dropped the reload option from the init script, breaking corekeeper' from 'corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.'. > affects -1 corekeeper Bug #991151 [procps] procps: dropped the reload option from the init script, breaking corekeeper Added indication that 991151 affects corekeeper -- 991151: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=991151 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
Control: reassign -1 procps 3.3.15-2 Control: retitle -1 procps: dropped the reload option from the init script, breaking corekeeper Control: affects -1 corekeeper On Fri, 2021-07-16 at 01:15 +, Thorsten Glaser wrote: > Yes, the procps init script does not have the action reload. Looks like this is a regression in procps in buster and later. I've bounced the thread to the procps maintainer and reassigned. -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
Paul Wise dixit:
>On Thu, 2021-07-15 at 21:34 +0200, Thorsten Glaser wrote:
>
>> invoke-rc.d: initscript procps, action "reload" failed.
>
>I don't have this problem on amd64 with systemd,
>can you reproduce it on amd64 with sysvinit?
Yes, the procps init script does not have the action reload.
| Usage: /etc/init.d/procps {start|stop|status|restart|try-restart|force-reload}
>I'm thinking of switching to systemd-coredump,
>are you interested in adopting corekeeper?
I’m trying to not increase the amount of time I sink into
Debian at this moment, sorry; corekeeper i̲s̲ useful though;
it just works as-is.
bye,
//mirabilos
--
Gestern Nacht ist mein IRC-Netzwerk explodiert. Ich hatte nicht damit
gerechnet, darum bin ich blutverschmiert… wer konnte ahnen, daß SIE so
reagier’n… gestern Nacht ist mein IRC-Netzwerk explodiert~~~
(as of 2021-06-15 The MirOS Project temporarily reconvenes on OFTC)
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
On Thu, 2021-07-15 at 21:34 +0200, Thorsten Glaser wrote: > invoke-rc.d: initscript procps, action "reload" failed. I don't have this problem on amd64 with systemd, can you reproduce it on amd64 with sysvinit? I'm thinking of switching to systemd-coredump, are you interested in adopting corekeeper? -- bye, pabs https://wiki.debian.org/PaulWise signature.asc Description: This is a digitally signed message part
Bug#991151: corekeeper: postrm: invoke-rc.d: initscript procps, action "reload" failed.
Package: corekeeper
Version: 1.7
Severity: serious
Justification: does not uninstall
X-Debbugs-Cc: t...@mirbsd.de
Removing corekeeper:x32 (1.7) ...
Usage: /etc/init.d/procps {start|stop|status|restart|try-restart|force-reload}
invoke-rc.d: initscript procps, action "reload" failed.
dpkg: error processing package corekeeper:x32 (--remove):
installed corekeeper:x32 package post-removal script subprocess returned error
exit status 3
dpkg: too many errors, stopping
Errors were encountered while processing:
corekeeper:x32
-- System Information:
Debian Release: 11.0
APT prefers unreleased
APT policy: (500, 'unreleased'), (500, 'buildd-unstable'), (500, 'unstable'),
(100, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: x32, i386
Kernel: Linux 5.10.0-6-amd64 (SMP w/4 CPU threads)
Kernel taint flags: TAINT_FIRMWARE_WORKAROUND
Locale: LANG=C, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)
Versions of packages corekeeper depends on:
ii procps 2:3.3.17-5
corekeeper recommends no packages.
corekeeper suggests no packages.
-- no debconf information
