Bug#863278: RFS: usbguard/0.7.0-1

2017-07-23 Thread gustavo panizzo

Hello

I did review the package, here are my comments

- i did compare usbguard-0.7.0 tag from upstream github with
 upstream/0.7.0+ds1 from your repo

- Why did you remove NOTES.md, doc/usbguard-vs-udev.md,  and ROADMAP.md?, they 
look useful.

- in your git repo there should be a tags with the debian releases, i only
 see upstream/XXX tags. please push, and sign, tags to your repo.

- why m4/* files haven't been updated since release 0.5, upstream no
 longer ship those files. Why keep shipping it? i can build
 successfully against stretch, with or without them.

- why src/GUI.qt/ is deleted?

- i'm glad you deleted the 3rd party code :)

- please import upstream tags on your git repo, that makes more easy for
 others to review (and hopefully send patches)

I was going to upload but the perl breakage make build the package
impossible giving us a few more days to finish above'

Best

On Sun, Jul 16, 2017 at 09:38:58PM +0200, Muri Nicanor wrote:

Hi,

On 07/16/2017 05:28 PM, gustavo panizzo wrote:

[...]
but i've removed those files from the archive for now, patched the
relevant testscripts and created a bug report to create the testdata
from source [0]

cool, excelent
please add a target gen-orig-xz (or gz, or bz2) to d/rules to generate
the source from upstream source.

done in git


d/control

- You have a d/gbp.conf but not Vcs-* headers on d/control
 would be cool if you publish your git packaging, and documented it on
 d/control

fixed


please push tags to you repo so i can compare between your tags and
upstream tags

done


the package fails to build with sbuild

it works here *if* i stop the usbguard-daemon on the system i'm building
the package. one of the tests tries to start a usbguard daemon which
fails if there is already a daemon running


please do the d/rules targe and/or the tag in the repo so i have a way
to validate the source you are providing before upload. i don't see
anything else
missing in the package :)


cheers,
muri






--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-07-23 Thread Muri Nicanor
Hi gustavo,

On 07/22/2017 08:43 AM, gustavo panizzo wrote:
> Hello
> 
> I did review the package, here are my comments
> 
> - i did compare usbguard-0.7.0 tag from upstream github with
>  upstream/0.7.0+ds1 from your repo
> 
> - Why did you remove NOTES.md, doc/usbguard-vs-udev.md,  and
> ROADMAP.md?, they look useful.
I did not- i'm not using the git repo from upstream as a source for
orig, but i'm using the tar.gz upstream provides. In this tar.gz (which
i imported using gbp import-orig --uscan), those files are not included

> - in your git repo there should be a tags with the debian releases, i only
>  see upstream/XXX tags. please push, and sign, tags to your repo.
Oke, i've created a signed tag for this release and pushed it

> - why m4/* files haven't been updated since release 0.5, upstream no
>  longer ship those files. Why keep shipping it? i can build
>  successfully against stretch, with or without them.
Hm, they are part of upstreams release, i'll ask if they can be removed
upstream

> - why src/GUI.qt/ is deleted?
Hm, what do you mean? i don't see that happening...

> - i'm glad you deleted the 3rd party code :)
> 
> - please import upstream tags on your git repo, that makes more easy for
>  others to review (and hopefully send patches)
I plan to switch to using upstreams git repo at some point, but at the
moment I feel more safe using the released .tar.gz files. Is it a big
advantage for reviewing the packaging stuff? In that case i'll try to
have a go at this before the next usbguard release...

cheers,
muri



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-07-16 Thread Muri Nicanor
Hi,

On 07/16/2017 05:28 PM, gustavo panizzo wrote:
>> [...]
>> but i've removed those files from the archive for now, patched the
>> relevant testscripts and created a bug report to create the testdata
>> from source [0]
> cool, excelent
> please add a target gen-orig-xz (or gz, or bz2) to d/rules to generate
> the source from upstream source.
done in git

>>> d/control
>>>
>>> - You have a d/gbp.conf but not Vcs-* headers on d/control
>>>  would be cool if you publish your git packaging, and documented it on
>>>  d/control
>> fixed
> 
> please push tags to you repo so i can compare between your tags and
> upstream tags
done

> the package fails to build with sbuild
it works here *if* i stop the usbguard-daemon on the system i'm building
the package. one of the tests tries to start a usbguard daemon which
fails if there is already a daemon running

> please do the d/rules targe and/or the tag in the repo so i have a way
> to validate the source you are providing before upload. i don't see
> anything else
> missing in the package :)

cheers,
muri



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-07-16 Thread gustavo panizzo

On Mon, Jul 10, 2017 at 03:52:12PM +0200, Muri Nicanor wrote:

On 06/29/2017 05:35 PM, gustavo panizzo wrote:

Hello, apologies for the late response


no worries, as you can see i'm not very fast myself

oke, so...

On 06/18/2017 11:23 AM, gustavo panizzo wrote:

d/copyright

- please set the years to include 2016, 2017. there are some files owned
by RH which are copyright 2016 but in d/copyright it only says 2015

fixed


- src/Tests/Makefile.in, src/Tests/test-driver are copyright FSF

added in d/copyright


- would be cool to run files-without-copyright.sh

you mean in d/rules?





- src/Tests/USB/data/0001.bin src/Tests/UseCase/DummyDevices/root.tar.xz
 what's the source of that file? please regenerate them at build time

i'm not sure how to deal with this- the 0001.bin is a copy of a binary
USB Descriptor of one of upstreams devices.

which device? is it a physical device? is it a virtual device?


upstream says:

This is a copy of a binary USB descriptor of one of my devices. I

don't think I know anymore which one

but i've removed those files from the archive for now, patched the
relevant testscripts and created a bug report to create the testdata
from source [0]

cool, excelent
please add a target gen-orig-xz (or gz, or bz2) to d/rules to generate
the source from upstream source. 




- please clarify copyright for src/Tests/custom.supp

done so in [1] and upstream added a public domain header [2]. i've added
a paragraph about that file to d/copyright, although it does not yet
contain that public domain header in this release (the file containing
the public domain header is not part of a relaese)


d/control

- You have a d/gbp.conf but not Vcs-* headers on d/control
 would be cool if you publish your git packaging, and documented it on
 d/control

fixed


please push tags to you repo so i can compare between your tags and
upstream tags




 if you do that, please use pristine-tar to publish the tarballs

oke, i've imported the tarballs using pristine-tar commit
../usbguard_0.7.0+ds1.orig.tar.gz and used --git-pristine-tar with gbp


d/changelog

- please fix spelling in 'neede kernek'

fixed


* New upstream version 0.7.0
 This release contains a backwards incompatible
 change because it changes how the device hash is
 computed for Linux root hub devices

Please create a NEWS.Debian file, also please make a note to add a
notice in the release manual for buster (i know, looong time)
Users hate when things change without notice (i'm an usbguard user :)

i've added a d/NEWS file (dch --create --news) and added the information
there. did you mean i should make a note for myself or did you mean a
note in d/NEWS and if the latter, what is the syntax?

I meant the latter (so users can read it and fix their configs)

the package fails to build with sbuild


please do the d/rules targe and/or the tag in the repo so i have a way
to validate the source you are providing before upload. i don't see anything 
else
missing in the package :)



new version is on mentors ;)

cheers,
muri

[0] https://github.com/dkopecek/usbguard/issues/188
[1] https://github.com/dkopecek/usbguard/issues/175
[2]
https://github.com/dkopecek/usbguard/commit/faa84a648a575db7d8e382729ea7dcea079afe3a






--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-07-16 Thread Muri Nicanor
i've uploaded another version that removes an unused build-dependency



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-07-10 Thread Muri Nicanor
On 06/29/2017 05:35 PM, gustavo panizzo wrote:
> Hello, apologies for the late response

no worries, as you can see i'm not very fast myself

oke, so...

On 06/18/2017 11:23 AM, gustavo panizzo wrote:
> d/copyright
> 
> - please set the years to include 2016, 2017. there are some files owned
> by RH which are copyright 2016 but in d/copyright it only says 2015
fixed

> - src/Tests/Makefile.in, src/Tests/test-driver are copyright FSF
added in d/copyright

> - would be cool to run files-without-copyright.sh
you mean in d/rules?

>>> - src/Tests/USB/data/0001.bin src/Tests/UseCase/DummyDevices/root.tar.xz
>>>  what's the source of that file? please regenerate them at build time
>> i'm not sure how to deal with this- the 0001.bin is a copy of a binary
>> USB Descriptor of one of upstreams devices.
> which device? is it a physical device? is it a virtual device?

upstream says:
> This is a copy of a binary USB descriptor of one of my devices. I
don't think I know anymore which one

but i've removed those files from the archive for now, patched the
relevant testscripts and created a bug report to create the testdata
from source [0]

> - please clarify copyright for src/Tests/custom.supp
done so in [1] and upstream added a public domain header [2]. i've added
a paragraph about that file to d/copyright, although it does not yet
contain that public domain header in this release (the file containing
the public domain header is not part of a relaese)

> d/control
> 
> - You have a d/gbp.conf but not Vcs-* headers on d/control
>  would be cool if you publish your git packaging, and documented it on
>  d/control
fixed

>  if you do that, please use pristine-tar to publish the tarballs
oke, i've imported the tarballs using pristine-tar commit
../usbguard_0.7.0+ds1.orig.tar.gz and used --git-pristine-tar with gbp

> d/changelog
> 
> - please fix spelling in 'neede kernek'
fixed

> * New upstream version 0.7.0
>  This release contains a backwards incompatible
>  change because it changes how the device hash is
>  computed for Linux root hub devices
> 
> Please create a NEWS.Debian file, also please make a note to add a
> notice in the release manual for buster (i know, looong time)
> Users hate when things change without notice (i'm an usbguard user :)
i've added a d/NEWS file (dch --create --news) and added the information
there. did you mean i should make a note for myself or did you mean a
note in d/NEWS and if the latter, what is the syntax?

new version is on mentors ;)

cheers,
muri

[0] https://github.com/dkopecek/usbguard/issues/188
[1] https://github.com/dkopecek/usbguard/issues/175
[2]
https://github.com/dkopecek/usbguard/commit/faa84a648a575db7d8e382729ea7dcea079afe3a



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-29 Thread gustavo panizzo

Hello, apologies for the late response

On Sun, Jun 25, 2017 at 07:49:51PM +0200, Muri Nicanor wrote:

Hi,

On 06/18/2017 11:23 AM, gustavo panizzo wrote:

- src/Tests/USB/data/0001.bin src/Tests/UseCase/DummyDevices/root.tar.xz
 what's the source of that file? please regenerate them at build time


i'm not sure how to deal with this- the 0001.bin is a copy of a binary
USB Descriptor of one of upstreams devices. 

which device? is it a physical device? is it a virtual device?


The root.tar.xz is an
archive containing the files from
src/Tests/UseCase/DummyDevices/{bus,devices}, which itself is a copy of
upstreams /sys/bus/usb and /sys/devices, containing also some binary
files. And the files are only there for the tests.

looks to me like a bunch of mkdir, touch, etc can replace it easily



i would compare this with a JPG-file thats part of a graphics library
test-routine or something like that...

we try to generate those at build time

PS: i forgot to add to the last email we should target unstable instead
of experimental :)



cheers,
muri






--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-25 Thread Muri Nicanor
Hi,

On 06/18/2017 11:23 AM, gustavo panizzo wrote:
> - src/Tests/USB/data/0001.bin src/Tests/UseCase/DummyDevices/root.tar.xz
>  what's the source of that file? please regenerate them at build time

i'm not sure how to deal with this- the 0001.bin is a copy of a binary
USB Descriptor of one of upstreams devices. The root.tar.xz is an
archive containing the files from
src/Tests/UseCase/DummyDevices/{bus,devices}, which itself is a copy of
upstreams /sys/bus/usb and /sys/devices, containing also some binary
files. And the files are only there for the tests.

i would compare this with a JPG-file thats part of a graphics library
test-routine or something like that...

cheers,
muri



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-18 Thread gustavo panizzo


On Wed, May 24, 2017 at 07:43:35PM +0200, Muri Nicanor wrote:

Package: sponsorship-requests
Severity: normal

Dear mentors,

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



d/copyright

- please set the years to include 2016, 2017. there are some files owned
by RH which are copyright 2016 but in d/copyright it only says 2015

- src/Tests/Makefile.in, src/Tests/test-driver are copyright FSF
- would be cool to run files-without-copyright.sh
- src/Tests/USB/data/0001.bin src/Tests/UseCase/DummyDevices/root.tar.xz 
 what's the source of that file? please regenerate them at build time
 
- please clarify copyright for src/Tests/custom.supp


d/control

- You have a d/gbp.conf but not Vcs-* headers on d/control
 would be cool if you publish your git packaging, and documented it on
 d/control
 
 if you do that, please use pristine-tar to publish the tarballs


d/changelog

- please fix spelling in 'neede kernek'

* New upstream version 0.7.0
 This release contains a backwards incompatible
 change because it changes how the device hash is
 computed for Linux root hub devices

Please create a NEWS.Debian file, also please make a note to add a
notice in the release manual for buster (i know, looong time)
Users hate when things change without notice (i'm an usbguard user :) 




--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-14 Thread gustavo panizzo
Hi

I'll review the package this weekend

On June 14, 2017 6:51:06 PM GMT+08:00, Muri Nicanor  wrote:
>control: tags -1 - moreinfo
>
>hi,
>
>On 06/01/2017 12:07 PM, gustavo panizzo wrote:
>> the package builds fine with pbuilder but tests fails when i try to
>> build using sbuild.
>> 
>> FAIL UseCase/001_cli_policy.sh (exit status: 1)
>this fails if there is already a running usbguard instance- when i
>stopped mine the test ran fine
>
>
>> perhaps usbguard should build-depend on sudo?
>> 
>> UseCase/002_cli_devices.sh.log:  + sudo -n dd bs=4096 count=1
>> if=/dev/zero of=/tmp/usbguard_disk
>> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line
>81:
>> sudo: command not found
>> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe dummy_hcd
>> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line
>82:
>> sudo: command not found
>> UseCase/002_cli_devices.sh.log:  + sudo -n rmmod g_mass_storage
>> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line
>83:
>> sudo: command not found
>> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe g_mass_storage
>> file=/tmp/usbguard_disk iSerialNumber=555666111
>> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line
>84:
>> sudo: command not found
>> UseCase/002_cli_devices.sh.log:  + '[' 127 -ne 0 ']'
>> UseCase/002_cli_devices.sh.log:  + exit 77
>> UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh
>(exit
>> status: 77)
>
>this test is now disabled (skipped), as the neede kernel modules needed
>are not part of debian default kernel
>
>> this should be easy to fix
>> 
>> Packaging/spell-check.sh.log:  Skipping test because some dependecies
>> are missing
>> Packaging/spell-check.sh.log:  ASPELL: Packaging/spell-check.sh.log: 
>> PANDOC: /usr/bin/pandoc
>> Packaging/spell-check.sh.log:  SKIP Packaging/spell-check.sh (exit
>> status: 77)
>fixed.
>
>i've uploaded a new version to mentors,
>
>cheers,
>muri

-- 
IRC: gfa
GPG: 0x44BB1BA79F6C6333

Sent from my mobile. Please excuse my brevity and typos.



Bug#863278: RFS: usbguard/0.7.0-1

2017-06-14 Thread Muri Nicanor
control: tags -1 - moreinfo

hi,

On 06/01/2017 12:07 PM, gustavo panizzo wrote:
> the package builds fine with pbuilder but tests fails when i try to
> build using sbuild.
> 
> FAIL UseCase/001_cli_policy.sh (exit status: 1)
this fails if there is already a running usbguard instance- when i
stopped mine the test ran fine


> perhaps usbguard should build-depend on sudo?
> 
> UseCase/002_cli_devices.sh.log:  + sudo -n dd bs=4096 count=1
> if=/dev/zero of=/tmp/usbguard_disk
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 81:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe dummy_hcd
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 82:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n rmmod g_mass_storage
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 83:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe g_mass_storage
> file=/tmp/usbguard_disk iSerialNumber=555666111
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 84:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + '[' 127 -ne 0 ']'
> UseCase/002_cli_devices.sh.log:  + exit 77
> UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh (exit
> status: 77)

this test is now disabled (skipped), as the neede kernel modules needed
are not part of debian default kernel

> this should be easy to fix
> 
> Packaging/spell-check.sh.log:  Skipping test because some dependecies
> are missing
> Packaging/spell-check.sh.log:  ASPELL: Packaging/spell-check.sh.log: 
> PANDOC: /usr/bin/pandoc
> Packaging/spell-check.sh.log:  SKIP Packaging/spell-check.sh (exit
> status: 77)
fixed.

i've uploaded a new version to mentors,

cheers,
muri



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-04 Thread Muri Nicanor
Hi,

On 06/01/2017 11:54 AM, gustavo panizzo wrote:
> Hello
> 
> On Wed, May 24, 2017 at 07:43:35PM +0200, Muri Nicanor wrote:
> 
>>  * New upstream version 0.7.0
>>This release contains a backwards incompatible
>>change because it changes how the device hash is
>>computed for Linux root hub devices
> 
> How is it handled for users that already have a config file? do they
> need to create a new configuration file?
kind of. the hashes of the devices in the usbguard rules file will have
to be set to new values. in the old versions of usbguard, the Linux root
hub bcdDevice value was included in the hash value of the device- but
this value changes every new kernel release. so it is now excluded from
calculating the hash.

> Can the update to the config file by done automatically? (postinst)
i don't think so- postinst would have to calculate new hashes for every
device listed in the rules file but most devices won't be present at the
time of upgrading.

> does this version need a newer kernel than older versions?
no



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-04 Thread Muri Nicanor
hi,

On 06/01/2017 12:07 PM, gustavo panizzo wrote:
> control: owner -1 !
> control: tags -1 moreinfo
> 
> the package builds fine with pbuilder but tests fails when i try to
> build using sbuild.
thanks for the pointer, i totally missed that only using pbuilder. now i
also have a working sbuild schroot ;)

> FAIL UseCase/001_cli_policy.sh (exit status: 1)
hm, i'm not sure yet why this test is failing, looking into thath

> 
> some tests are skipped, i feel they should run
> 
> 
> perhaps usbguard should build-depend on sudo?
> 
> UseCase/002_cli_devices.sh.log:  + sudo -n dd bs=4096 count=1
> if=/dev/zero of=/tmp/usbguard_disk
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 81:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe dummy_hcd
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 82:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n rmmod g_mass_storage
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 83:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe g_mass_storage
> file=/tmp/usbguard_disk iSerialNumber=555666111
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 84:
> sudo: command not found
> UseCase/002_cli_devices.sh.log:  + '[' 127 -ne 0 ']'
> UseCase/002_cli_devices.sh.log:  + exit 77
> UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh (exit
> status: 77)

i tried with removing the sudo, but debian doesn't have dummy_hcd or
g_mass_storage included in the kernel, so these tests won't work at all :(

> 
> 
> this should be easy to fix
> 
> Packaging/spell-check.sh.log:  Skipping test because some dependecies
> are missing
> Packaging/spell-check.sh.log:  ASPELL: Packaging/spell-check.sh.log: 
> PANDOC: /usr/bin/pandoc
> Packaging/spell-check.sh.log:  SKIP Packaging/spell-check.sh (exit
> status: 77)

thanks! i've added aspell to the build-dependencies!

cheers,
muri



signature.asc
Description: OpenPGP digital signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-01 Thread gustavo panizzo


On June 1, 2017 6:13:06 PM GMT+08:00, Andrey Rahmatullin  
wrote:

>> UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh
>(exit status: 77)
>No, it's a bad idea to do things requiring root during tests,
>especially
>modprobe/rmmod

Maybe those tests can be run using autopkgtests

I'd prefer to run tests if they are available


>And please remember that just calling sudo is not enough to get root.

Of course, sudo's configuration was explicit in my mind, my bad I didn't 
mentioned it.

-- 
IRC: gfa
GPG: 0x44BB1BA79F6C6333

Sent from my mobile. Please excuse my brevity and typos.

signature.asc
Description: PGP signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-01 Thread Andrey Rahmatullin
On Thu, Jun 01, 2017 at 06:07:40PM +0800, gustavo panizzo wrote:
> some tests are skipped, i feel they should run
> 
> 
> perhaps usbguard should build-depend on sudo?
> 
> UseCase/002_cli_devices.sh.log:  + sudo -n dd bs=4096 count=1 if=/dev/zero 
> of=/tmp/usbguard_disk
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 81: sudo: 
> command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe dummy_hcd
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 82: sudo: 
> command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n rmmod g_mass_storage
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 83: sudo: 
> command not found
> UseCase/002_cli_devices.sh.log:  + sudo -n modprobe g_mass_storage 
> file=/tmp/usbguard_disk iSerialNumber=555666111
> UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 84: sudo: 
> command not found
> UseCase/002_cli_devices.sh.log:  + '[' 127 -ne 0 ']'
> UseCase/002_cli_devices.sh.log:  + exit 77
> UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh (exit 
> status: 77)
No, it's a bad idea to do things requiring root during tests, especially
modprobe/rmmod.
And please remember that just calling sudo is not enough to get root.


-- 
WBR, wRAR


signature.asc
Description: PGP signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-01 Thread gustavo panizzo

control: owner -1 !
control: tags -1 moreinfo

the package builds fine with pbuilder but tests fails when i try to
build using sbuild.

FAIL UseCase/001_cli_policy.sh (exit status: 1)

some tests are skipped, i feel they should run


perhaps usbguard should build-depend on sudo?

UseCase/002_cli_devices.sh.log:  + sudo -n dd bs=4096 count=1 if=/dev/zero 
of=/tmp/usbguard_disk
UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 81: sudo: 
command not found
UseCase/002_cli_devices.sh.log:  + sudo -n modprobe dummy_hcd
UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 82: sudo: 
command not found
UseCase/002_cli_devices.sh.log:  + sudo -n rmmod g_mass_storage
UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 83: sudo: 
command not found
UseCase/002_cli_devices.sh.log:  + sudo -n modprobe g_mass_storage 
file=/tmp/usbguard_disk iSerialNumber=555666111
UseCase/002_cli_devices.sh.log:  ./UseCase/002_cli_devices.sh: line 84: sudo: 
command not found
UseCase/002_cli_devices.sh.log:  + '[' 127 -ne 0 ']'
UseCase/002_cli_devices.sh.log:  + exit 77
UseCase/002_cli_devices.sh.log:  SKIP UseCase/002_cli_devices.sh (exit status: 
77)


this should be easy to fix

Packaging/spell-check.sh.log:  Skipping test because some dependecies are 
missing
Packaging/spell-check.sh.log:  ASPELL: 
Packaging/spell-check.sh.log:  PANDOC: /usr/bin/pandoc

Packaging/spell-check.sh.log:  SKIP Packaging/spell-check.sh (exit status: 77)

thanks!

--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-06-01 Thread gustavo panizzo

Hello

On Wed, May 24, 2017 at 07:43:35PM +0200, Muri Nicanor wrote:


 * New upstream version 0.7.0
   This release contains a backwards incompatible
   change because it changes how the device hash is
   computed for Linux root hub devices


How is it handled for users that already have a config file? do they
need to create a new configuration file?
Can the update to the config file by done automatically? (postinst)

does this version need a newer kernel than older versions?

--
IRC: gfa
GPG: 0X44BB1BA79F6C6333



signature.asc
Description: Digital Signature


Bug#863278: RFS: usbguard/0.7.0-1

2017-05-24 Thread Muri Nicanor
Package: sponsorship-requests
Severity: normal

Dear mentors,

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

* Package name: usbguard
  Version : 0.7.0-1
  Upstream Author : Daniel Kopeček
* Url : https://dkopecek.github.io/usbguard/
* Licenses: CC-BY-SA-3.0,FSFULLR,GPL-3+,GPL-2+
  Section : utils

It builds those binary packages:

  * libusbguard0
  * usbguard
  * usbguard-applet-qt

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

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

Alternatively, one can download the package with dget using this command:
dget -x
https://mentors.debian.net/debian/pool/main/u/usbguard/usbguard_0.7.0-1.dsc

Or clone the package from https://0xacab.org/muri/debian-usbguard

More information about usbguard can be obtained from
https://dkopecek.github.io/usbguard/

Changes since last upload:

  * New upstream version 0.7.0
This release contains a backwards incompatible
change because it changes how the device hash is
computed for Linux root hub devices
  * Add support for bash-completion
  * Delete add-unistd.patch, its not needed anymore
  * Compile the library static, to make `make check` work
  * Bump compat version to 10
  * Add d/usbguard.dirs

Regards,
  Muri Nicanor




signature.asc
Description: OpenPGP digital signature