Bug#926306: RFS: socklog/2.1.0-9

2019-04-17 Thread Mathieu Mirmont
On Wed, Apr 17, 2019 at 10:15:56AM +, Dmitry Bogatov wrote:
> 
> [2019-04-16 19:10] Mathieu Mirmont 
> > Both issues fixed, I re-uploaded the package to mentors (same links).
> 
> Uploaded. For next upload, please, take a look at following:
> 
>  * in `debian/rules' you skip dh_dwz step. Please write comment why.
>  * in socklog-init init script you mix tabs and spaces. It is of no
>importance to Policy, but looks wierd.
>  * file a bug aganist Lintian and either add comment into
>'d/socklog.lintian-override', or, hopefully, this file will be
>unneeded by time of next upload of `socklog'.

Sure, will do.

Thank for your review time and for uploading the package!

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-17 Thread Dmitry Bogatov


[2019-04-16 19:10] Mathieu Mirmont 
> Both issues fixed, I re-uploaded the package to mentors (same links).

Uploaded. For next upload, please, take a look at following:

 * in `debian/rules' you skip dh_dwz step. Please write comment why.
 * in socklog-init init script you mix tabs and spaces. It is of no
   importance to Policy, but looks wierd.
 * file a bug aganist Lintian and either add comment into
   'd/socklog.lintian-override', or, hopefully, this file will be
   unneeded by time of next upload of `socklog'.
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-16 Thread Dmitry Bogatov


[2019-04-13 11:11] Mathieu Mirmont 
> On Fri, Apr 12, 2019 at 05:22:35PM +, Dmitry Bogatov wrote:
> > [2019-04-10 23:48] Mathieu Mirmont 
> > >
> > > part 1 text/plain 434
> > > On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
> > > > You can repack it as new upstream version. New version would be
> > > > something like `2.1.0+repack-1'. Do not forget add clarification into
> > > > Debian.source.
> > >
> > > I've updated the package and uploaded to mentors:
> > >
> > > https://mentors.debian.net/package/socklog
> > > https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0+repack-1.dsc
> > 
> > Lintian (2.11.0) warnings:
> > 
> > W: socklog: missing-versioned-depends-on-init-system-helpers 
> > postinst:152 "update-rc.d defaults
> -disabled" needs i
> > nit-system-helpers >= 1.50
>
> Embarassingly I don't know how to get rid of this one. I've added a
> depencency on init-system-helpers (>= 1.50) naturally but the warning
> remains.
>
> Since ${misc:Pre-Depends} already includes init-system-helpers (>=
> 1.54~) I added lintian overrides. It's a bit dirty though, and the
> line number makes it a moving target. I'll fix the line numbers in the
> lintian overrides but if you know how to fix this in a better way I'm
> all ears.

I see. I suggest you to submit bug aganist Lintian, but in mean time I
am fine uploading with overrides. It may be useful to add reference to
that bug in override comment.

> > Oh, and extremely minor notice: in `debian/control' you align fields
> > with tabs. It does not look pretty, if tabstop is not 8. What about
> > expand(1)?
>
> Ah yeah sorry, I'll untabify this.

Thank you. By the way, you know about wrap-and-sort(1)?
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-16 Thread Mathieu Mirmont
On Fri, Apr 12, 2019 at 05:22:35PM +, Dmitry Bogatov wrote:
> [2019-04-10 23:48] Mathieu Mirmont 
> >
> > part 1 text/plain 434
> > On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
> > > You can repack it as new upstream version. New version would be
> > > something like `2.1.0+repack-1'. Do not forget add clarification into
> > > Debian.source.
> >
> > I've updated the package and uploaded to mentors:
> >
> > https://mentors.debian.net/package/socklog
> > https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0+repack-1.dsc
> 
> Lintian (2.11.0) warnings:
> 
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:152 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:165 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:178 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50
> 
> Oh, and extremely minor notice: in `debian/control' you align fields
> with tabs. It does not look pretty, if tabstop is not 8. What about
> expand(1)?

Both issues fixed, I re-uploaded the package to mentors (same links).

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-13 Thread Mathieu Mirmont
On Fri, Apr 12, 2019 at 05:22:35PM +, Dmitry Bogatov wrote:
> [2019-04-10 23:48] Mathieu Mirmont 
> >
> > part 1 text/plain 434
> > On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
> > > You can repack it as new upstream version. New version would be
> > > something like `2.1.0+repack-1'. Do not forget add clarification into
> > > Debian.source.
> >
> > I've updated the package and uploaded to mentors:
> >
> > https://mentors.debian.net/package/socklog
> > https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0+repack-1.dsc
> 
> Lintian (2.11.0) warnings:
> 
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:152 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:165 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50
>   W: socklog: missing-versioned-depends-on-init-system-helpers 
> postinst:178 "update-rc.d defaults-disabled" needs i
>   nit-system-helpers >= 1.50

Embarassingly I don't know how to get rid of this one. I've added a
depencency on init-system-helpers (>= 1.50) naturally but the warning
remains.

Since ${misc:Pre-Depends} already includes init-system-helpers (>=
1.54~) I added lintian overrides. It's a bit dirty though, and the
line number makes it a moving target. I'll fix the line numbers in the
lintian overrides but if you know how to fix this in a better way I'm
all ears.

> Oh, and extremely minor notice: in `debian/control' you align fields
> with tabs. It does not look pretty, if tabstop is not 8. What about
> expand(1)?

Ah yeah sorry, I'll untabify this.

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-12 Thread Dmitry Bogatov


[2019-04-10 12:46] Mathieu Mirmont 
> > Are you sure you need "Depends: runit"? Maybe it is just me, but I
> > thought that `socklog' is just an implementation of `syslog'. If
> > so, hard dependency on bin:runit is quite heavyweight.
>
> It heavily relies on runsv and svlogd, that's how it splits files by
> syslog service name. Essentially socklog is a data collector for svlogd.
>
> The alternative would be to re-implement runsv and svlogd by redirecting
> the output of socklog to a file and using logrotate, but clearly that
> would be a hack, it's not how it's meant to be run, and it would have
> the risk of losing messages when logrotate kicks in.
>
> The dependency on runit is less risky than that. Also the dependency is
> on runit alone, not on any of the runit-* packages that call runsvdir,
> so it's quite lite.

I see. Fine.
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-12 Thread Dmitry Bogatov
[2019-04-10 23:48] Mathieu Mirmont 
>
> part 1 text/plain 434
> On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
> > You can repack it as new upstream version. New version would be
> > something like `2.1.0+repack-1'. Do not forget add clarification into
> > Debian.source.
>
> I've updated the package and uploaded to mentors:
>
> https://mentors.debian.net/package/socklog
> https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0+repack-1.dsc

Lintian (2.11.0) warnings:

W: socklog: missing-versioned-depends-on-init-system-helpers 
postinst:152 "update-rc.d defaults-disabled" needs i
nit-system-helpers >= 1.50
W: socklog: missing-versioned-depends-on-init-system-helpers 
postinst:165 "update-rc.d defaults-disabled" needs i
nit-system-helpers >= 1.50
W: socklog: missing-versioned-depends-on-init-system-helpers 
postinst:178 "update-rc.d defaults-disabled" needs i
nit-system-helpers >= 1.50

Oh, and extremely minor notice: in `debian/control' you align fields
with tabs. It does not look pretty, if tabstop is not 8. What about
expand(1)?
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-10 Thread Mathieu Mirmont
On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
> You can repack it as new upstream version. New version would be
> something like `2.1.0+repack-1'. Do not forget add clarification into
> Debian.source.

I've updated the package and uploaded to mentors:

https://mentors.debian.net/package/socklog
https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0+repack-1.dsc

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-10 Thread Mathieu Mirmont
On Wed, Apr 10, 2019 at 08:20:30AM +, Dmitry Bogatov wrote:
>
> [2019-04-08 01:03] Mathieu Mirmont 
> > Done, the init scripts call daemon(1) and runsv(1) and they work
> > pretty nicely.
>
> Are you sure you need "Depends: runit"? Maybe it is just me, but I
> thought that `socklog' is just an implementation of `syslog'. If
> so, hard dependency on bin:runit is quite heavyweight.

It heavily relies on runsv and svlogd, that's how it splits files by
syslog service name. Essentially socklog is a data collector for svlogd.

The alternative would be to re-implement runsv and svlogd by redirecting
the output of socklog to a file and using logrotate, but clearly that
would be a hack, it's not how it's meant to be run, and it would have
the risk of losing messages when logrotate kicks in.

The dependency on runit is less risky than that. Also the dependency is
on runit alone, not on any of the runit-* packages that call runsvdir,
so it's quite lite.

> > There is one more issue that I noticed this weekend: the orig.tar.gz
> > file that is registered in debian archives is not the same as the
> > upstream tarball. It is in fact a tarball of the upstream tarball
> > (!). I don't know why it's done this way, and it pretty much breaks
> > breaks source format 3.0 (quilt) because I can't get dpkg-source to
> > unpack the tarball before applying the patches. Do you know how to
> > deal with that?
>
> You can repack it as new upstream version. New version would be
> something like `2.1.0+repack-1'. Do not forget add clarification into
> Debian.source.

Ah! You made my day, I didn't know this was possible!

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-10 Thread Dmitry Bogatov


[2019-04-08 01:03] Mathieu Mirmont 
> On Sat, Apr 06, 2019 at 07:13:56PM +, Dmitry Bogatov wrote:
> > 
> > [2019-04-04 13:30] Mathieu Mirmont 
> > > > * I know, it is pain, but there should be init.d script. You may want to
> > > >   take a look at bcron=0.11-8.
> > >
> > > Sure, no worries. How about systemd service files? It makes little sense
> > > to run socklog along with systemd I think, but for the principle it may
> > > be required to profile service files. What do you think?
> > 
> > Up to you. Presence of systemd unit files is not mandated by Policy,
> > unlike init.d scripts.
>
> Done, the init scripts call daemon(1) and runsv(1) and they work
> pretty nicely.

Are you sure you need "Depends: runit"? Maybe it is just me, but I
thought that `socklog' is just an implementation of `syslog'. If
so, hard dependency on bin:runit is quite heavyweight.

> > > >   I believe there should be separate sysuser for socklog-* services.
> > > >   Ideally, separate sysuser for /every/ from socklog-* service, but I do
> > > >   not know, whehter it is possible.
> > >
> > > Yeah good point. I tend to think that a single user for all socklog-*
> > > services would be enough, but if you prefer I can add one user per
> > > service.
> > 
> > Yes, I'd prefer as much separation, as possible.
>
> Done, one user per service.

I see. Thank you.

> There is one more issue that I noticed this weekend: the orig.tar.gz
> file that is registered in debian archives is not the same as the
> upstream tarball. It is in fact a tarball of the upstream tarball
> (!). I don't know why it's done this way, and it pretty much breaks
> breaks source format 3.0 (quilt) because I can't get dpkg-source to
> unpack the tarball before applying the patches. Do you know how to
> deal with that?

You can repack it as new upstream version. New version would be
something like `2.1.0+repack-1'. Do not forget add clarification into
Debian.source.
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-07 Thread Mathieu Mirmont
On Sat, Apr 06, 2019 at 07:13:56PM +, Dmitry Bogatov wrote:
> 
> [2019-04-04 13:30] Mathieu Mirmont 
> > > * I know, it is pain, but there should be init.d script. You may want to
> > >   take a look at bcron=0.11-8.
> >
> > Sure, no worries. How about systemd service files? It makes little sense
> > to run socklog along with systemd I think, but for the principle it may
> > be required to profile service files. What do you think?
> 
> Up to you. Presence of systemd unit files is not mandated by Policy,
> unlike init.d scripts.

Done, the init scripts call daemon(1) and runsv(1) and they work
pretty nicely.

> > >   I believe there should be separate sysuser for socklog-* services.
> > >   Ideally, separate sysuser for /every/ from socklog-* service, but I do
> > >   not know, whehter it is possible.
> >
> > Yeah good point. I tend to think that a single user for all socklog-*
> > services would be enough, but if you prefer I can add one user per
> > service.
> 
> Yes, I'd prefer as much separation, as possible.

Done, one user per service.

> > Thanks for the review!
> 
> My pleasure. By the way, you seems to forgot to add changelog entry
> about new maintainer. Something in lines:
> 
>   * Set myself as maintainer (Closes: #)

I had this line but somehow I messed up and accidentally squashed two
commits so the line disappeared (I use gbp dch to generate the
changelog). I've put it back.

There is one more issue that I noticed this weekend: the orig.tar.gz
file that is registered in debian archives is not the same as the
upstream tarball. It is in fact a tarball of the upstream tarball
(!). I don't know why it's done this way, and it pretty much breaks
breaks source format 3.0 (quilt) because I can't get dpkg-source to
unpack the tarball before applying the patches. Do you know how to
deal with that?

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-06 Thread Dmitry Bogatov


[2019-04-04 13:30] Mathieu Mirmont 
> > * I know, it is pain, but there should be init.d script. You may want to
> >   take a look at bcron=0.11-8.
>
> Sure, no worries. How about systemd service files? It makes little sense
> to run socklog along with systemd I think, but for the principle it may
> be required to profile service files. What do you think?

Up to you. Presence of systemd unit files is not mandated by Policy,
unlike init.d scripts.

> >   I believe there should be separate sysuser for socklog-* services.
> >   Ideally, separate sysuser for /every/ from socklog-* service, but I do
> >   not know, whehter it is possible.
>
> Yeah good point. I tend to think that a single user for all socklog-*
> services would be enough, but if you prefer I can add one user per
> service.

Yes, I'd prefer as much separation, as possible.

> Thanks for the review!

My pleasure. By the way, you seems to forgot to add changelog entry
about new maintainer. Something in lines:

  * Set myself as maintainer (Closes: #)
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Gerrit Pape
On Thu, Apr 04, 2019 at 01:40:16PM +0200, Mathieu Mirmont wrote:
> On Thu, Apr 04, 2019 at 12:48:43PM +0200, Adam Borowski wrote:
> > A nice new process is described here:
> >https://wiki.debian.org/PackageSalvaging
> > but it's probably an overkill for a clear case like here.  But, the old
> > maintainer being the upstream means you need to at least communicate with
> > him beforehand.
> 
> I did reach out to Gerrit Pape (previous maintainer & upstream) of
> course before doing anything. I offered to help and he was happy to hand
> over the package to me. I'll file an ITS if you guys think I should.
> 
> Adding Gerrit Pape to CC.

Hi all, this is okay with me!

I'm happy to hand over the package, sorry for not making this clear in
advance.

Best Regards, and have fun, Mathieu, with maintaining socklog.



Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Adam Borowski
On Thu, Apr 04, 2019 at 01:30:30PM +0200, Mathieu Mirmont wrote:
> On Thu, Apr 04, 2019 at 10:20:47AM +, Dmitry Bogatov wrote:
> > * I know, it is pain, but there should be init.d script. You may want to
> >   take a look at bcron=0.11-8.
> 
> Sure, no worries. How about systemd service files? It makes little sense
> to run socklog along with systemd I think, but for the principle it may
> be required to profile service files. What do you think?

No point in adding .service if you already have an init script; that'd be
useful only if you want to do something _different_ when running under
systemd, or when you hit a systemd bug that makes it unable to handle a
regular init script.


Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Did ya know that typing "test -j8" instead of "ctest -j8"
⢿⡄⠘⠷⠚⠋⠀ will make your testsuite pass much faster, and fix bugs?
⠈⠳⣄



Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Adam Borowski
On Thu, Apr 04, 2019 at 01:40:16PM +0200, Mathieu Mirmont wrote:
> I did reach out to Gerrit Pape (previous maintainer & upstream) of
> course before doing anything. I offered to help and he was happy to hand
> over the package to me.

Cool, that answers my question.  It'll be enough to write "New maintainer"
in the changelog then.

> I'll file an ITS if you guys think I should.

No point, you already have Gerrit's consent.


Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Did ya know that typing "test -j8" instead of "ctest -j8"
⢿⡄⠘⠷⠚⠋⠀ will make your testsuite pass much faster, and fix bugs?
⠈⠳⣄



Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Mathieu Mirmont
On Thu, Apr 04, 2019 at 12:48:43PM +0200, Adam Borowski wrote:
> On Wed, Apr 03, 2019 at 10:45:56AM +0200, Mathieu Mirmont wrote:
> > * Package name: socklog
> >   Version : 2.1.0-9
> >   Upstream Author : Gerrit Pape 
> 
> > Changes since the last upload:
> > 
> >   * Convert the package to debhelper (Closes: #857208)
> >   * patches: Import previous patches
> >   * patch: remove the chkshsgr test
> >   * watch: add the uscan watch file
> >   * socklog-run: migrate to dh-runit (Closes: #668718, #834089)
> >   * gitlab-ci.yml: add GitLab CI file
> >   * control: update the Vcs fields
> >   * doc-base: register the html documentation
> >   * lintian: add overrides
> 
> This is a package hijack.  In this case, the package is so neglected (no
> maintainer upload in 11 years, long-standing RC bugs) that insisting on the
> proper procedure would be a waste of time -- but it still should be done
> consciously.  No mention in the changelog suggests that it's not.
> 
> A nice new process is described here:
>https://wiki.debian.org/PackageSalvaging
> but it's probably an overkill for a clear case like here.  But, the old
> maintainer being the upstream means you need to at least communicate with
> him beforehand.

I did reach out to Gerrit Pape (previous maintainer & upstream) of
course before doing anything. I offered to help and he was happy to hand
over the package to me. I'll file an ITS if you guys think I should.

Adding Gerrit Pape to CC.

Cheers,

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Mathieu Mirmont
On Thu, Apr 04, 2019 at 10:20:47AM +, Dmitry Bogatov wrote:
> * Please, do not use ${runit:Conflicts}. As suggested by Lintian, it is
>   very strong relations. Use ${runit:Breaks} only. The very
>   introduction of ${runit:Conflicts} was my mistake. It will make
>   Lintian override unneeded.

Will do.

> * Please, consider merging bin:socklog-run into bin:socklog. Option
>   `since' for dh_runit will be useful for it.

Ah that's what you meant, sorry I misunderstood. I will try to merge
them.

> * I know, it is pain, but there should be init.d script. You may want to
>   take a look at bcron=0.11-8.

Sure, no worries. How about systemd service files? It makes little sense
to run socklog along with systemd I think, but for the principle it may
be required to profile service files. What do you think?

> * Please, add description to 0002-import-patch-tryto. It is unclear,
>   what issue this patch resolves.

This one comes from the previous packaging scripts. I'll do some
digging.

> * In patch 0003-remove-chkshgrp you remove test, that fails on CI. Does
>   it also fails in sbuild? If not, probably it should only be disabled in
>   Gitlab CI?

It only fails in CI. I'll try to have it disabled in CI only, that would
be much better indeed.

> * It is matter of taste, but are you aware, that you can
> 
>   Build-Depends: debhelper-compat (= 12)
> 
>   and remove `debian/compat'?

I didn't know, thanks :)

> * Dep-5 would be nice.

Will do.

> * What is the purpose of 'log' user, you create with dh_sysuser? You
>   know, that bin:runit provides user `runit-log' since -20, don't you?

I overlooked that, thanks.

> * All services run as same user, 'nobody'. It is unfortunate, since nobody is
>   quite popular owner for NFS files.
> 
>   I believe there should be separate sysuser for socklog-* services.
>   Ideally, separate sysuser for /every/ from socklog-* service, but I do
>   not know, whehter it is possible.

Yeah good point. I tend to think that a single user for all socklog-*
services would be enough, but if you prefer I can add one user per
service.

> * I believe, README file is useless -- it contains copyright, authorship
>   and homepage information only, which is already present in Debian
>   package files.

Alright, I'll remove it.

Thanks for the review!

Cheers,

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature


Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Adam Borowski
On Wed, Apr 03, 2019 at 10:45:56AM +0200, Mathieu Mirmont wrote:
> * Package name: socklog
>   Version : 2.1.0-9
>   Upstream Author : Gerrit Pape 

> Changes since the last upload:
> 
>   * Convert the package to debhelper (Closes: #857208)
>   * patches: Import previous patches
>   * patch: remove the chkshsgr test
>   * watch: add the uscan watch file
>   * socklog-run: migrate to dh-runit (Closes: #668718, #834089)
>   * gitlab-ci.yml: add GitLab CI file
>   * control: update the Vcs fields
>   * doc-base: register the html documentation
>   * lintian: add overrides

This is a package hijack.  In this case, the package is so neglected (no
maintainer upload in 11 years, long-standing RC bugs) that insisting on the
proper procedure would be a waste of time -- but it still should be done
consciously.  No mention in the changelog suggests that it's not.

A nice new process is described here:
   https://wiki.debian.org/PackageSalvaging
but it's probably an overkill for a clear case like here.  But, the old
maintainer being the upstream means you need to at least communicate with
him beforehand.

Adding Mattia Rizzolo to CC -- he's not only handling MIA, but also happened
to NMU this package a couple of years ago.


Meow!
-- 
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Did ya know that typing "test -j8" instead of "ctest -j8"
⢿⡄⠘⠷⠚⠋⠀ will make your testsuite pass much faster, and fix bugs?
⠈⠳⣄



Bug#926306: RFS: socklog/2.1.0-9

2019-04-04 Thread Dmitry Bogatov


[2019-04-03 10:45] Mathieu Mirmont 
> Package: sponsorship-requests
> Severity: normal [important for RC bugs, wishlist for new packages]
>
> Dear mentors,
>
> I am looking for a sponsor for my package "socklog"
>
> * Package name: socklog
>   Version : 2.1.0-9
>   Upstream Author : Gerrit Pape 
> * URL : http://smarden.org/socklog
> * License : BSD
>   Section : admin
>
> It builds those binary packages:
>
> socklog - system and kernel logging services (programs)
> socklog-run - system and kernel logging services
>
> To access further information about this package, please visit the
> following URL:
>
>   https://mentors.debian.net/package/socklog
>
>
> Alternatively, one can download the package with dget using this
> command:
>
>   dget -x 
> https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0-9.dsc
>
> Changes since the last upload:
>
>   * Convert the package to debhelper (Closes: #857208)
>   * patches: Import previous patches
>   * patch: remove the chkshsgr test
>   * watch: add the uscan watch file
>   * socklog-run: migrate to dh-runit (Closes: #668718, #834089)
>   * gitlab-ci.yml: add GitLab CI file
>   * control: update the Vcs fields
>   * doc-base: register the html documentation
>   * lintian: add overrides


* Please, do not use ${runit:Conflicts}. As suggested by Lintian, it is
  very strong relations. Use ${runit:Breaks} only. The very
  introduction of ${runit:Conflicts} was my mistake. It will make
  Lintian override unneeded.

* Please, consider merging bin:socklog-run into bin:socklog. Option
  `since' for dh_runit will be useful for it.

* I know, it is pain, but there should be init.d script. You may want to
  take a look at bcron=0.11-8.

* Please, add description to 0002-import-patch-tryto. It is unclear,
  what issue this patch resolves.

* In patch 0003-remove-chkshgrp you remove test, that fails on CI. Does
  it also fails in sbuild? If not, probably it should only be disabled in
  Gitlab CI?

* It is matter of taste, but are you aware, that you can

Build-Depends: debhelper-compat (= 12)

  and remove `debian/compat'?

* Dep-5 would be nice.

* What is the purpose of 'log' user, you create with dh_sysuser? You
  know, that bin:runit provides user `runit-log' since -20, don't you?

* All services run as same user, 'nobody'. It is unfortunate, since nobody is
  quite popular owner for NFS files. 

  I believe there should be separate sysuser for socklog-* services.
  Ideally, separate sysuser for /every/ from socklog-* service, but I do
  not know, whehter it is possible.

* I believe, README file is useless -- it contains copyright, authorship
  and homepage information only, which is already present in Debian
  package files.
-- 
Note, that I send and fetch email in batch, once every 24 hours.
 If matter is urgent, try https://t.me/kaction
 --



Bug#926306: RFS: socklog/2.1.0-9

2019-04-03 Thread Mathieu Mirmont
Package: sponsorship-requests
Severity: normal [important for RC bugs, wishlist for new packages]

Dear mentors,

I am looking for a sponsor for my package "socklog"

* Package name: socklog
  Version : 2.1.0-9
  Upstream Author : Gerrit Pape 
* URL : http://smarden.org/socklog
* License : BSD
  Section : admin

It builds those binary packages:

socklog - system and kernel logging services (programs)
socklog-run - system and kernel logging services

To access further information about this package, please visit the
following URL:

  https://mentors.debian.net/package/socklog


Alternatively, one can download the package with dget using this
command:

  dget -x 
https://mentors.debian.net/debian/pool/main/s/socklog/socklog_2.1.0-9.dsc

Changes since the last upload:

  * Convert the package to debhelper (Closes: #857208)
  * patches: Import previous patches
  * patch: remove the chkshsgr test
  * watch: add the uscan watch file
  * socklog-run: migrate to dh-runit (Closes: #668718, #834089)
  * gitlab-ci.yml: add GitLab CI file
  * control: update the Vcs fields
  * doc-base: register the html documentation
  * lintian: add overrides

Cheers,

-- 
Mathieu Mirmont 


signature.asc
Description: PGP signature