Re: Ubuntu 22.04 LTS support

[Christian Ehrhardt]

On Mon, Feb 6, 2023 at 12:00 AM alfredo.nodo
 wrote:
>
> Hi,
> I see here that you are the maintainers of the package jami 
> https://packages.ubuntu.com/focal/jami
> Are you planning to port it to Ubuntu 22.04 LTS?

Hi Alfredo,
this binary package belongs to source ring which has many many bugs
and crashes [1].
Hence it was removed from Ubuntu [2] and Debian [3] over and over again.
There now is a new version [4] but it seems to break just as much.

Due to that it is unlikely to make it into the next release 23.04 and
even more unlikely to then be backported into 22.04.

[1]: https://bugs.launchpad.net/ubuntu/+source/ring
[2]: https://bugs.launchpad.net/ubuntu/+source/ring/+bug/1885646
[3]: https://tracker.debian.org/news/1270862/ring-removed-from-testing/
[4]: https://launchpad.net/ubuntu/+source/ring/20210112.2.b757bac~ds1-2

I hope that clarifies why you miss it post focal.


> Thank you
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: libmodbus-dev

[Christian Ehrhardt]

On Sat, Jan 14, 2023 at 10:35 PM walter graham
 wrote:
>
>
>
> Hi there!
>
> I installed libmodbus-dev from https://libmodbus.org/getting_started/
>
> However, I am new to this and have no idea how to "run" libmodbus or even 
> what the command is.
> Can you point me in the right direction?  The website seems like it's only 
> for installing.

Hi Walter,
libmoddbus is "only" a library, it won't do much for you on its own.
Packages with the -dev suffix [1] in particular are usually containing
headers, pkgconfig, and man pages of that library, so that you can
start to write a program that uses it.
The actual library [2] with code is currently in the `libmodbus5`
package, but again - a library on its own won't do much.

Since you seem to look for a tool, more than a library we can have a
look which packages in the Ubuntu Archive use that library via

$ reverse-depends --release=lunar libmodbus5
Reverse-Recommends
* collectd
Reverse-Depends
* libmodbus-dev
* linuxcnc-uspace
* mbpoll
* nut-modbus
* zabbix-agent

Of these, description-wise, mbpoll might be closer to what you look for:

$ apt-cache show mbpoll
...
Description-en: command line utility to communicate with ModBus slave
(RTU or TCP)
 mbpoll uses libmodbus (http://libmodbus.org/).
 Although the syntax of these options is very close modpoll proconX program,
 it is a completely independent project.
 .
 mbpoll can:
 .
  - read discrete inputs
  - read and write binary outputs (coil)
  - read input registers
  - read and write output registers (holding register)
 .
 The reading and writing registers may be in decimal, hexadecimal or
 floating single precision.

Not sure if that was the tool you were looking for, but understanding
the difference between library and programs will be an important step
#1 in any case.

[1]: 
https://www.debian.org/doc/debian-policy/ch-sharedlibs.html#development-files
[2]: 
https://www.debian.org/doc/debian-policy/ch-sharedlibs.html#shared-libraries


> My only experience is setting up master and slave back in 1993 on msdos.  I 
> am not quite sure where to start with Linux.
>
> I could not find anything in my desktop start menu so tried Terminal command 
> line using libmodbus and libmodbus-dev and those commands are not found.  I 
> re-installed libmodbus-dev using Synaptic Package manager and no change.
>
> Thank you,
> Walt
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> "They ceased to wear their prosperity with moderation and the universe struck 
> them down" Plato
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



--
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Xen support in the upcoming 22.04 LTS version

[Christian Ehrhardt]

On Fri, Sep 9, 2022 at 8:26 PM Éliás Tamás  wrote:
>
> Dear Christian.
>
> As per my current understanding you should be able to commit some work
> on the xen package in ubuntu.
>
> https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2022-January/019169.html
>
>
> I'm just wondering, that I have seen for kinect there is an updated and
> well patched xen version. Should it be possible to do the same for 22.04
> LTS? If somewhere, in an LTS this could be more than important.
>
> I'm suffering with a missing pygrub ZSTD compression from pygrub and a
> couple of security fixes.
>
> pls let me know if there are any future plans on the update. Thx.

Hiho,
thanks for reaching out.

Indeed we (Luis and others helped along the way) got it updated in
Kinetic to 4.16 which was some work.
Backporting that (4.16) fully to Jammy is against the SRU rules [0] as
it would introduce as much regressions as it would help.
But the bug you are waiting for is [1] itself which is about
backporting just this missing feature to fix boot and not all of 4.16.

As far as I'm concerned I considered Mauricio assigned to [1], the bug
state also still shows that.
Not sure what blocks this, maybe it was getting 4.16 into Kinetic
which is done now, maybe something else.
Adding Mauricio to this thread to answer.

[0]: https://wiki.ubuntu.com/StableReleaseUpdates
[1]: https://bugs.launchpad.net/ubuntu/+source/xen/+bug/1956166

> --
>
> Éliás Tamás
> Thomas Elias
>
> ETIT[nwpro] KFT, Ügyvezető-Hálózatbiztonsági specialista
> ETIT[nwpro] Ltd, General Manager-Network security specialist
>
> Tel. HU: +36/30-497-1626
> OpenPGP pubkey: http://etit.hu/doc/et-pub.asc
>
> Okleveles mérnök-informatikus (MSC)
> Master of Science in Information Technology (MSC)
> Licenced Penetration Tester (TM15-047)
>
> Kapcsolat: http://etit.hu/index.php/hu/kapcsolat
> Jogi nyilatkozat: http://etit.hu/disclaimer-email-hu.txt
> Contact: http://etit.hu/index.php/en/contact
> Disclaimer: http://etit.hu/disclaimer-email-en.txt
>
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Package TOMCAT 9.0.54 for Ubuntu 20.04

[Christian Ehrhardt]

On Wed, Jul 6, 2022 at 11:36 AM Rhea Moubarak  wrote:
>
> Hello,
>
>
>
> Can you please integrate TOMCAT 9.0.54 in the official repositories of Ubuntu 
> 20.04 ?
>
>
>
> As listed on https://packages.ubuntu.com/focal/tomcat9, the current version 
> in the official repo is 9.0.31
>

Hi Rhea,
but gladly this isn't plain 9.0.31.
There was a similar bug request [1] which got resolved a while ago in
[2] and I think has solved all those security issues in the 9.0.31
version that is in Focal.

On the other side there is the SRU policy [3] which prevents too big
version jumps unless there is extra focus on stability and testing
which needs further effort and dedication which for tomcat9 being only
in universe isn't provided by anyone at the moment.

[1]: https://bugs.launchpad.net/ubuntu/+source/tomcat9/+bug/1915911
[2]: https://launchpad.net/ubuntu/+source/tomcat9/9.0.31-1ubuntu0.2
[3]: https://wiki.ubuntu.com/StableReleaseUpdates

>
> It would be helpful to add the version of TOMCAT 9.0.54 as it fixes major 
> security bugs on TOMCAT installations.
>
>
>
> Thank you in advance.
>
>
>
> Regards,
>
> Rhea
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Please update liburing package.

[Christian Ehrhardt]

On Sun, Jul 3, 2022 at 7:17 PM 김태인  wrote:
>
> Hello!
> I'm just programmer in Korea. I'm using liburing-dev & liburing2 package in 
> ubuntu 22.04. There were important updates on liburing 2.2 version. I need it.

Hi,
packages are in general updated in the new releases of Ubuntu.
The upcoming 22.10 already has 2.2-1 about to be added.

 liburing | 2.1-2build1  | kinetic  | source
 liburing | 2.2-1| kinetic-proposed | source

For reliability and stability changes to existing releases have to
follow the SRU process [1].
So if you have an individual bug you might file a request so it can be
checked if it can be fixed.
But if you just need the full new version please consider upgrading to
the newest release once available.
Or already use the version that is in development right now, there are
always daily images, system containers and so much more available, you
do not have to wait for the release date (it is just not yet as stable
or complete as the final release).

[1]: https://wiki.ubuntu.com/StableReleaseUpdates

> Please update this feature. Thank you!
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: How to Remove Ubuntu Advantage from my computer?

[Christian Ehrhardt]

On Sun, Jul 3, 2022 at 7:17 PM J .W. Jansen
 wrote:
>
> Hi,
>
> I discover the Ubuntu Advantage Tools package o my (private,
> non-business) computer (Currently at latest update level for XUbuntu
> 20.04). I certainly never asked for this software!
>
> When I try to remove this package I get a list of other packages which
> would be removed along with it, consisting of software which seems
> necessary to me for the overall functioning of my software. The list
> goes as follows:
>
>  ttf-mscorefonts-installer
>  ubuntu-minimal
>  ubuntu-release-upgrader-gtk
>  update-manager
>  update-manager-core
>  update-notifier
>  update-notifier-common
>
> How can I get rid of this unwanted (i.e. BLOATWARE...) stuff?

Hi,
this was already discussed at [1] and [2] and a few other places
linked from there.
If you have new technical arguments I'd ask you to consider replying
to that mail thread so that they can be considered in the same context
where the discussion took place so far.

[1]: https://lists.ubuntu.com/archives/ubuntu-devel/2022-June/042115.html
[2]: 
https://bugs.launchpad.net/ubuntu/+source/ubuntu-advantage-tools/+bug/1950692

> Several other packages react in the same way when I try to remove them,
> so I would be grateful for suggestions to achieve this!
>
> Thanks,
>
> Hans Jansen.
>
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Senior Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Apache2 mpm_event bug still present in Ubuntu repositories

[Christian Ehrhardt]

On Fri, Jun 24, 2022 at 6:53 PM Mike Luttikhuis
 wrote:
>
> Hi,
>
> A bug in the apache2 package where mpm_event does not always gracefully kill 
> child processes after a graceful reload of the service is not patched in the 
> Ubuntu repositories. This is a nasty bug because if for example a SSL 
> certificate is replaced inside a vhost configuration, then older child 
> processes will still serve the older certificate, resulting in certificate 
> errors. Please check out PR 63169. Apache fixed this bug in 2.4.49, but the 
> bugfix is still not included in the Ubuntu repository. Since this bug is 
> pretty frustrating, I was wondering if this particular bugfix can be patched 
> in Ubuntu's version of the apache2 package.

Hi Mike,
thanks for your report, at a first glimpse the case looks valid and
doable, therefore I have converted this into a proper bug [1] to be
tracked.
You could help the team to have a look by providing steps to recreate
the situation via an update on that bug.

[1]: https://bugs.launchpad.net/ubuntu/+source/apache2/+bug/1979942

> Kind regards,
>
> Mike
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Reporting Security Vulnerabilities - libcurl4,curl

[Christian Ehrhardt]

On Wed, Jun 15, 2022 at 11:47 PM Ruelo, Christine M. L. <
christine.m.l.ru...@accenture.com> wrote:

> Hello libcurl4,curl Maintainers,
>
>
>
> Good day, We have used the libcurl4,curl package and perform a security
> scan using Palo Alto Network – Prisma Cloud and these vulnerabilities below
> are reported.
>

Hi,
not sure which Ubuntu releases you scanned but all these were handled and
fixed quite a while ago.
To help you find that information yourself for this and any other cases let
me point you to USN [1].

There you can enter your CVE numbers and will find which release was
affected and in which package versions it got fixed.
If you want to do such checks automatically there is also oval data [2].

If your scanners still report the issue you'll need to check that in
detail, but form my personal experience in 9 out of 10 cases the problem is
that they only perform "if version > X" which doesn't always work well for
fixes backported to versions that are in Distributions (for example
libcurl3-gnutls - 7.74.0-1ubuntu2 becomes libcurl3-gnutls -
7.74.0-1ubuntu2.1 due to the fix - but the scanner might just check > 7.75).

[1]: https://ubuntu.com/security/notices
[2]: https://ubuntu.com/security/oval


> We would like to report it and let us know once the fix is available so we
> can update accordingly.
>
>
>
> CVE-2021-22898
>
> CVE-2021-22947
>
> CVE-2021-22946
>
> CVE-2021-22945
>
> CVE-2021-22924
>
>
>
> Thank you
>
>
>
> Regards,
>
> *I* CHRISTINE MAE RUELO
>
> *I* ATCP | Data + AI
>
> *I* Global One Eastwood
>
> *I* E: christine.m.l.ru...@accenture.com
>
> *I* M: +63 927 088 6796
>
> Accenture Confidential
>
> *PTO: *
>
> *Holiday: *
>
> *Training: *
>
>
>
> --
>
> This message is for the designated recipient only and may contain
> privileged, proprietary, or otherwise confidential information. If you have
> received it in error, please notify the sender immediately and delete the
> original. Any other use of the e-mail by you is prohibited. Where allowed
> by local law, electronic communications with Accenture and its affiliates,
> including e-mail and instant messaging (including content), may be scanned
> by our systems for the purposes of information security and assessment of
> internal compliance with Accenture policy. Your privacy is important to us.
> Accenture uses your personal data only in compliance with data protection
> laws. For further information on how Accenture processes your personal
> data, please see our privacy statement at
> https://www.accenture.com/us-en/privacy-policy.
>
> __
>
> www.accenture.com
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>


-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Request to update 'ethtool' package

[Christian Ehrhardt]

On Sun, Jun 5, 2022 at 11:16 PM Bryant Eadon  wrote:
>
> It appears the latest ethtool package is v5.4 for Ubuntu 20.04 
> (https://packages.ubuntu.com/focal/ethtool ) .  The latest from kernel.org is 
> 5.17.  Is this normally something that is updated for 20.04 LTS ?
>
> Can the ethtool package be updated by a maintainer for these Ubuntu versions?

Hi Bryant,
the ethtool package is maintained by the Kernel team and usually
aligned with the kernel versions.
Since we have -hwe [1] kernels it might be reasonable to think about
also backporting a newer ethtools.
In the same way one could think about other packages aligned to the
available kernel version like iproute2.

But while this would be great for -hwe users it will OTOH be a rather
large regression risk for those
not using it. That will certainly need some discussions between the
Kernel and SRU team.

Maybe those discussions already happened and ended with "let us not do
it", I do not know.
Setting Stefan "to" as I know he sometimes looks after these packages
and might know
more (FYI - He's out for a few more days, do not expect the fastest reply).

[1]: https://wiki.ubuntu.com/Kernel/LTSEnablementStack

> Thank you,
> Bryant
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Call for votes: Developer Membership Board restaffing

[Christian Ehrhardt]

On Wed, Mar 23, 2022 at 12:06 PM Robie Basak  wrote:
>
> The Developer Membership Board (DMB) has started a restaffing vote for
> three vacant seats. The DMB is responsible for reviewing and approving
> new Ubuntu developers. It evaluates prospective Ubuntu developers and
> decides when to entrust them with developer privileges. There are five
> candidates:
>
> William 'jawn-smith' Wilson (https://wiki.ubuntu.com/jawn-smith/RunningForDMB)
> Brian Murray
> Sebastien Bacher
> Lucas Kanashiro
> Utkarsh Gupta
>
> The vote closes on 2022-03-30 at approximately 18:00 UTC.
>
> As usual, the election is being run using the Condorcet Internet Voting
> Service. All members of the ~ubuntu-dev team in Launchpad are eligible
> to vote. However, recent changes to CIVS require you to *opt-in* to
> using your email address to vote. You cannot receive a ballot until you
> have opted in. Please opt in here:
>
> https://civs1.civs.us/cgi-bin/opt_in.pl
>
> You must use the same email address as the one I have collected for you
> for your ballot. This is generally @ubuntu.com if that is an
> alias that you have published in Launchpad publicly. More complicated
> matching rules apply otherwise. If you are a member of ~ubuntu-dev and
> are struggling to acquire your ballot, please contact me privately. I
> may need to identify the email alias I am using for you directly.

Thank you so much Robie for outlining all this in detail.
Since I just activated mine, but needed to try a few addresses I
wanted to describe how it looks.
I hope it will help that everybody knows what to expect.

If you happen to activate the "wrong" email you'll just see:

```
  Email address successfully activated.
```

But if you activate the right one (just as Robie said, usually the
@ubuntu.com one) you'd in the Web UI of CIVS right after clicking
"Complete activation" see this:

```
Email address successfully activated.
  Pending poll invitations:
Ubuntu Developer Membership Board restaffing
```

The latter line is a link leading you to your vote.


> This announcement is being sent to a moderated announcement mailing
> list. For discussion, Ubuntu developers should use
> ubuntu-de...@lists.ubuntu.com.
>
> On behalf of the DMB,
>
> Robie
> --
> ubuntu-devel-announce mailing list
> ubuntu-devel-annou...@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-announce



--
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Xen support in the upcoming 22.04 LTS version

[Christian Ehrhardt]

On Wed, Jan 26, 2022 at 3:47 PM Oliver Linden  wrote:
>
> Hi folks,
>
> checking the current packages for the 22.04 LTS version I recognized
> that the Xen packages are still on version 4.11 as maintained in 20.04 LTS.
>
> Since Debian bullseye is already with Xen version 4.14 I would suggest a
> sync with the Debian repository before freeze to get a more recent,
> performant and secure Xen shipped with the new LTS release.

As part of [1] which I also referred to you on the other mail thread
that you started, I'm working on xen 4.16 for Jammy.
There are even PPAs on the bug if you want to sniff-test things.

To be clear, I'm unable to commit to making xen in jammy super-shiny,
but better than the totally outdated 4.11 it should be.

[1]: https://bugs.launchpad.net/ubuntu/+source/xen/+bug/1956166

> Best, Oliver
>
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: current ubuntu kernel compression method is breaking VMs using pygrub (Xen)

[Christian Ehrhardt]

On Wed, Jan 26, 2022 at 3:47 PM Oliver Linden  wrote:
>
> With at least the two most recent kernel versions
>
> linux-image-5.13.0-25-generic
> linux-image-5.13.0-27-generic
>
> ubuntu switched to ZSTD as default kernel compression method breaking all Xen 
> installations using "pygrub", regardless whether it's a Dom0 or a DomU.
>
> Please switch back to a "pygrub" supported kernel compression method.

Hi Oliver,
I'm not going into the pygrub side of things here, but I wanted to let
you know that there is [1] which I think reported this issue for xen.
There is work ongoing that tries to get a newer xen into jammy that
can work with that.

There might be SRUs later that fix this in older releases, but given
that xen is "only" in universe I do not want to commit/promise that.

[1]: https://bugs.launchpad.net/ubuntu/+source/xen/+bug/1956166

> Best, Oliver
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [integer-Ticket #81335] Log4J Sicherheitslücke

[Christian Ehrhardt]

On Tue, Dec 14, 2021 at 10:17 PM integer GmbH  wrote:

> Hello Ubuntu-Team,
> can you please tell me if the follwoing software is affected by the Log4J
> exploit?
>

*disclaimer: I'm not from the security team and this is not a definitive or
formal answer*

Hi,
In general for CVEs you'd want to check the https://ubuntu.com/security
entry for it.
It will mention its status, affected packages and link to further
ressources one should know about.
In this case the links to USN and the wiki page are very helpful as well.

In this case that is at: https://ubuntu.com/security/CVE-2021-44228

The TL;DR could be, you do not list the affected package "apache-log4j1.2",
so not affected.
But TBH your customers website clearly runs apache2 + php which isn't
listed here.
Might there also be some java or any other solution (like an appliance
which sometimes
aren't transparent what they use internally) that uses log4j2, no one here
would know.

python3.8
> python3.8-minimal
> python3-appdirs
> python3-apt
> python3-certifi
> python3-chardet
> python3-crypto
> python3-dbus
> python3-distlib
> python3-distro
> python3-distro-info
> python3-distupgrade
> python3-distutils
> python3-dnspython
> python3-filelock
> python3-gi
> python3-gpg
> python3-idna
> python3-importlib-metadata
> python3-ldb
> python3-lib2to3
> python3-markdown
> python3-minimal
> python3-more-itertools
> python3-netifaces
> python3-packaging
> python3-pkg-resources
> python3-pygments
> python3-pyparsing
> python3-requests
> python3-samba
> python3-six
> python3-talloc
> python3-tdb
> python3-update-manager
> python3-urllib3
> python3-virtualenv
> python3-yaml
> python3-zipp
> python3.6-minimal
> readline-common
> rename
> resolvconf
> rsync
> rsyslog
> samba
> samba-common
> samba-common-bin
> samba-dsdb-modules
> samba-libs
> samba-vfs-modules
> sed
> sensible-utils
> shared-mime-info
> socat
> squid
> squid-common
> squid-langpack
> ssl-cert
> sudo
> systemd
> systemd-sysv
> systemd-timesyncd
> sysvinit-utils
> tar
> tcpd
> tdb-tools
> thermald
> tzdata
> ubuntu-advantage-tools
> ubuntu-minimal
> ubuntu-release-upgrader-core
> ucf
> udev
> update-inetd
> update-manager-core
> usb.ids
> usbutils
> util-linux
> vim-common
> vim-tiny
> virtualenv
> wget
> whiptail
> winbind
> xauth
> xdg-user-dirs
> xkb-data
> xxd
> xz-utils
> zerofree
> zlib1g
> tasksel
> tasksel-data
>
> Our client Hopfenveredlung St. Johann is using this software and we want
> to make sure they are not affected by the Log4J exploit.
>
> Best Regards
> Jonas Böck
>
>
> _ _ _
> integer GmbH Support
> Telefon 08252 - 96031 - 10
> |
> E-Mail: supp...@integer-it.de
> <https://integer.de/> <http://www.integer-it.de/>
> Hans-Sachs-Weg 25
> |
> 86529
> Schrobenhausen
> Registergericht: Amtsgericht Ingolstadt
> Registernummer: HRB 7821
> Geschäftsführer: Luise Krammer
> Allgemeine Datenschutzhinweise:
> *https://integer-it.de/ds.html* <http://www.integer-it.de/ds.html>
> Folgen Sie uns auf: [image: Facebook] <https://www.facebook.com/integerit>
>  [image: Instagram] <https://www.instagram.com/integergmbh/>F
>
>
> <https://heyalter.com/schrobenhausen/>
> <https://heyalter.com/schrobenhausen/>
> <https://heyalter.com/schrobenhausen/>
> _ _ _
>
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>


-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Remove CloudInit from Ubuntu Server

[Christian Ehrhardt]

On Mon, Nov 29, 2021 at 10:24 PM Jeffrey Walton  wrote:
>
> Hi Everyone,
>
> I'm testing Ubuntu Server in a VM. I noticed the server edition
> includes CloudInit by default. I don't need a VM inside a VM so I
> would like to remove it.

There might be a misunderstanding. The CloudInit installed in your
current VM (level 1) isn't responsible for a potential VM (level 2)
further inside it.

It is responsible to allow customization of your current VM (level 1).
And depending on where/how you run/host your current VM (level 1) it
might already be used to do customization when you have spawned it.

> The CloudInit docs don't say if it is Ok to remove
> (https://help.ubuntu.com/community/CloudInit).
>
> Is it Ok to remove CloudInit?

We generally consider the ability to customize a system a core feature
of a server image, therefore ubuntu-server-minimal depends on
cloud-init and removing it will thereby wreak havoc as it would
(auto)remove plenty of other things then.
Even when not using CloudInit, it isn't much of a burden - it is small
and designed to quickly check if there is no datasource and then does
nothing.
If even that is too much you can completely disable its start, for
that check out [1]

[1]: https://cloudinit.readthedocs.io/en/latest/topics/boot.html#generator

> Thanks in advance.
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: updated package ausweisapp2 needs to be imported to ubuntu

[Christian Ehrhardt]

On Sat, Nov 13, 2021 at 10:43 PM Gunnar Hjalmarsson  wrote:
>
> On 2021-11-13 08:02, ciphert...@gmx.de wrote:
> > Hi there,
> >
> > in https://bugs.launchpad.net/ubuntu/+source/ausweisapp2/+bug/1942649
> > the maintainer of the upstream debian pacakge to ausweisapp2 recommends
> > "to ping the Ubuntu maintainers to import the latest version of the
> > package from Debian", because a major functionality of the package is
> > broken since about april this year but obviously fixed in the newer
> > version of the upstream package.
> >
> > Is this the right address and way to do so?
>
> Yes, it's a proper way to reach out to Ubuntu developers.
>
> However, the latest ausweisapp2 version in Debian has already been
> sync'ed into Ubuntu, and is available in the Ubuntu 21.10 archive as
> well as the archive for the coming 22.04.
>
> It's not in 20.04, though. To get it into 20.04, someone needs to
> propose a stable release update in accordance with the related guidelines:
>
> https://wiki.ubuntu.com/StableReleaseUpdates
>
> Normally a new upstream release is not a proper candidate for a stable
> release update. OTOH, if the current version in focal is non-working,
> there may still be a chance...

Ack to all that Gunnar has said, but I just wanted to let you know that there
also is a snap of ausweisapp2 [1] which helps to have the new version
available on 20.04 and many other places.

But if - due to external changes - the version in 20.04 really became
dysfunctional then the SRU rules would allow it to push a new major version.
The complex tradeoff begins when it has many functions and only one is
inhibited,
but the update might break valid users of other functions.

[1]: https://snapcraft.io/ausweisapp2-ce

> HTH
>
> --
> Gunnar Hjalmarsson
> https://launchpad.net/~gunnarhj
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Release vmlinuz and initrd alongside iso

[Christian Ehrhardt]

On Wed, Nov 10, 2021 at 2:51 AM Michael Hudson-Doyle
 wrote:
>
> Hi,
>
> On Tue, 9 Nov 2021 at 02:19, MonkZ  wrote:
>>
>> Hi,
>>
>> currently enabling booting via ipxe (https://ipxe.org/) over http needs
>> a dedicated mirror that has vmlinuz and initrd extracted from the iso.
>>
>> Would it be possible to release those files - already extracted from the
>> iso - alongside those very isos?
>
>
> This is something we should do, yes. I've created an internal ticket to look 
> into it (after failing to find the one I was *sure* already existed) -- I 
> personally am not very familiar with the relevant bits and pieces.

Wasn't that in the past already available and discoverable through
simplestreams for MAAS doing PXE boots?
Here an old example what I remember
https://lists.ubuntu.com/archives/ubuntu-server-bugs/2015-October/137272.html

IIRC this isn't used exactly the same way nowadays, but the following
still discovers me a set of kernels and can be further restricted as
needed:

sstream-query 
http://images.maas.io/ephemeral-v3/stable/streams/v1/com.ubuntu.maas:stable:v3:download.json
kflavor=generic ftype=boot-kernel release=focal arch=amd64
sstream-query 
http://images.maas.io/ephemeral-v3/stable/streams/v1/com.ubuntu.maas:stable:v3:download.json
kflavor=generic ftype=boot-initrd release=focal arch=amd64

The Maas team might have better info how to (or why you should not)
use those nowadays.

> Cheers,
> mwh
>
>>
>> This would enable ubuntu to create one iso, that can boot every version
>> available directly from one usb stick or pxe server.
>>
>> #!ipxe
>> dhcp
>> set http-server https://releases.ubuntu.com
>> kernel
>> http://${http-server}/ubuntu/21.10/ubuntu-21.10-desktop-amd64.iso.vmlinuz
>> initrd
>> http://${http-server}/ubuntu/21.10/ubuntu-21.10-desktop-amd64.iso.initrd
>> imgargs ubuntu-21.10-desktop-amd64.iso.vmlinuz root=/dev/ram0
>> ramdisk_size=300 boot=casper ip=dhcp netboot=url
>> url=http://${http-server}/os/ubuntu/21.10/ubuntu-21.10.x-desktop-amd64.iso
>> boot || shell
>>
>> Something similar is done by Arch Linux. But they included the Let's
>> Encrypt Root Certificate too, to enable boot via https, and they
>> chainloaded a menu from their mirror to just have one iso or usbstick
>> image to work indefinitely, as the critical information, what to boot
>> would be loaded after the boot of ipxe.
>>
>>
>> Regards
>>
>> MonkZ
>>
>>
>> --
>> Ubuntu-devel-discuss mailing list
>> Ubuntu-devel-discuss@lists.ubuntu.com
>> Modify settings or unsubscribe at: 
>> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Build limits for specific package

[Christian Ehrhardt]

On Fri, Sep 24, 2021 at 7:53 AM Christian Ehrhardt
 wrote:
>
> On Fri, Sep 24, 2021 at 7:50 AM Christian Ehrhardt
>  wrote:
> >
> > On Thu, Sep 23, 2021 at 8:37 PM Nicholas Guriev  wrote:
> > >
> > > Hi folks!
> > >
> > > What is actual RAM limit on Launchpad for building packages for Ubuntu?
> >
> > Hi Nicholas,
> > Default, 1.5GB 1 vCPU
> > Can be bumped to 8GB 4vCPU
> >
>
> Arr, I misread as you talk about build and I answered for testing :-/
> There is something similar for that though, let me look for the links 

Unfortunately I found nothing useful and eventually got the
confirmation (thanks Wgrant) that all builders of a given architecture
share the same specs and no method to vary it.
I'll track down what I thought to remember because I'm curious now,
but that might not help you and will take a while...

> > > I noticed some builds of the telegram-desktop package failed with an
> > > error "virtual memory exhausted: Cannot allocate memory". For example:
> > >
> > > https://launchpad.net/ubuntu/+source/telegram-desktop/2.9.2+ds-1ubuntu2/+build/22012030
> > >
> > > At last, I was forced to lower debug info by the -g1 compiler flag in
> > > attempting to reduce main memory consumption. But sadly this makes
> > > generated stacktraces hardly readable.
> > >
> > > Is there a chance to raise the limit for the package?
> >
> > IMHO [1] will be a good start paragraph to read and provide you all
> > the links you will need to get to the details and how to change it per
> > package.
> >
> > [1]: 
> > https://github.com/canonical/ubuntu-maintainers-handbook/blob/main/ProposedMigration.md#test-framework-timeouts-and-out-of-memory
> >
> > > --
> > > Ubuntu-devel-discuss mailing list
> > > Ubuntu-devel-discuss@lists.ubuntu.com
> > > Modify settings or unsubscribe at: 
> > > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
> >
> >
> >
> > --
> > Christian Ehrhardt
> > Staff Engineer, Ubuntu Server
> > Canonical Ltd
>
>
>
> --
> Christian Ehrhardt
> Staff Engineer, Ubuntu Server
> Canonical Ltd



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Build limits for specific package

[Christian Ehrhardt]

On Fri, Sep 24, 2021 at 7:50 AM Christian Ehrhardt
 wrote:
>
> On Thu, Sep 23, 2021 at 8:37 PM Nicholas Guriev  wrote:
> >
> > Hi folks!
> >
> > What is actual RAM limit on Launchpad for building packages for Ubuntu?
>
> Hi Nicholas,
> Default, 1.5GB 1 vCPU
> Can be bumped to 8GB 4vCPU
>

Arr, I misread as you talk about build and I answered for testing :-/
There is something similar for that though, let me look for the links 


> > I noticed some builds of the telegram-desktop package failed with an
> > error "virtual memory exhausted: Cannot allocate memory". For example:
> >
> > https://launchpad.net/ubuntu/+source/telegram-desktop/2.9.2+ds-1ubuntu2/+build/22012030
> >
> > At last, I was forced to lower debug info by the -g1 compiler flag in
> > attempting to reduce main memory consumption. But sadly this makes
> > generated stacktraces hardly readable.
> >
> > Is there a chance to raise the limit for the package?
>
> IMHO [1] will be a good start paragraph to read and provide you all
> the links you will need to get to the details and how to change it per
> package.
>
> [1]: 
> https://github.com/canonical/ubuntu-maintainers-handbook/blob/main/ProposedMigration.md#test-framework-timeouts-and-out-of-memory
>
> > --
> > Ubuntu-devel-discuss mailing list
> > Ubuntu-devel-discuss@lists.ubuntu.com
> > Modify settings or unsubscribe at: 
> > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>
>
>
> --
> Christian Ehrhardt
> Staff Engineer, Ubuntu Server
> Canonical Ltd



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Build limits for specific package

[Christian Ehrhardt]

On Thu, Sep 23, 2021 at 8:37 PM Nicholas Guriev  wrote:
>
> Hi folks!
>
> What is actual RAM limit on Launchpad for building packages for Ubuntu?

Hi Nicholas,
Default, 1.5GB 1 vCPU
Can be bumped to 8GB 4vCPU

> I noticed some builds of the telegram-desktop package failed with an
> error "virtual memory exhausted: Cannot allocate memory". For example:
>
> https://launchpad.net/ubuntu/+source/telegram-desktop/2.9.2+ds-1ubuntu2/+build/22012030
>
> At last, I was forced to lower debug info by the -g1 compiler flag in
> attempting to reduce main memory consumption. But sadly this makes
> generated stacktraces hardly readable.
>
> Is there a chance to raise the limit for the package?

IMHO [1] will be a good start paragraph to read and provide you all
the links you will need to get to the details and how to change it per
package.

[1]: 
https://github.com/canonical/ubuntu-maintainers-handbook/blob/main/ProposedMigration.md#test-framework-timeouts-and-out-of-memory

> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Misplaced packages?

[Christian Ehrhardt]

On Tue, Aug 31, 2021 at 11:49 PM Gunnar Hjalmarsson  wrote:
>
> On 2021-08-27 16:19, Robert Spanjaard wrote:
> > Greetings Ubuntu Developers!
> >
> > I was just browsing some sections in Synaptic, looking for lightweight
> > webbrowsers and image gallery creation. During my browse I found some
> > strange package placements:
> >
> > w3m, a text based webbrowser, is placed in the section Word Processing
> > bluefish, a text editor, is placed in World Wide Web (multiverse)
> > florence, a virtual keyboard for X, is placed in World Wide Web
> > (multiverse)
> >
> > All on Ubuntu 20.04(.3).
>
> Weird. I installed w3m (on 21.04) to take a look.
>
> $ cat /var/lib/dpkg/status | grep 'Package: w3m' -A 3
> Package: w3m
> Status: install ok installed
> Priority: optional
> Section: web
> $ apt-cache show w3m | grep Section
> Section: text
>
> It looks some kind of 'disagreement' between dpkg and apt while Synaptic
> is an interface to browse apt info.

Maybe because it had changed in the past?
One can understand why one could mark it "text" in the first place as
it is a text based browser not realizing that text is "for text
processors".
But in 0.5.2-3 it had the change of "Set Section to web."
That version is old (pre-precise), so it has been that way (web) for
quite some time.

Therefore when you check the sources "web" is what you get just as
with the /var/lib/dpkg/status check mentioned above.
$ apt source w3m
$ grep Section w3m-0.5.3/debian/control
Section: web

Also it is Section web on lauchpad for example at
https://launchpad.net/ubuntu/+source/w3m/0.5.3-37

But OTOH the cache data is on "text" - and that is what apt/synaptic will use:
$ apt-cache show w3m | grep Section
Section: text

It seems some data holds to the old "text" entry ever since.

Overview:
web: /var/lib/dpkg/status, launchpad, source
text: apt-cache, synaptic

Is there a step needed/missed to make the apt/repository data refresh
the section?

> --
> Gunnar Hjalmarsson
> https://launchpad.net/~gunnarhj
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Major changes in sorting method of glibc on Ubuntu 18.04 after upgrading to Ubuntu 18.04.5

[Christian Ehrhardt]

On Thu, Feb 25, 2021 at 6:47 PM Ганиев Ильяс  wrote:
>
> I had an Ubuntu 18.04.0 with Postgres 9.6.
>
> Before Ubuntu upgrade following command result was:
>
> vodka@ubuntu140:~$  ( echo "1-1"; echo "11" ) | LC_COLLATE=en_US.UTF-8 sort
> 1-1
> 11
>
> But after upgrading Ubuntu 18.04 to latest release via apt upgrade (18.04.5) 
> result of the command above totally changed:
>
> vodka@ubuntu140:~$ ( echo "1-1"; echo "11" ) | LC_COLLATE=en_US.UTF-8 sort
> 11
> 1-1
>
> Due to this our production Postgres database started work very slowly and we 
> had a long downtime for REINDEX. Sorting method is very important for 
> Postgres database: https://wiki.postgresql.org/wiki/Locale_data_changes
>
> Please, read "Testing collation" part.
>
> Is it normal behavior for Ubuntu? Why glibc totally changed within LTS 
> release without 'major upgrade'? I did not expect this...

Hi,
this has been filed and discussed here:
https://bugs.launchpad.net/ubuntu/+source/base-files/+bug/1916935
I'd ask you to read (and write if you have more to add) there for
further progress on this.

> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: adduser race conditions

[Christian Ehrhardt]

On Fri, Feb 19, 2021 at 6:19 PM Brent W. Baccala  wrote:
>
> Hi -
>
> I've found at least one race condition in the adduser script on Ubuntu 18.
>
> The program searches for an available UID (and GID), then adds the new user 
> with the selected UID/GID numbers without any synchronization, so two 
> near-simultaneous invocations of the program can produce two users with 
> identical GIDs (what I saw), or, presumably, identical UIDs.
>
> Is there a bug/issue tracker somewhere for this program?

Hi Brent,
the upstream for this program is Debian [1][2] for general bugs and/or
feature requests you'd use those.
While the Ubuntu bug tracker [3] would mostly be for issues due to the
Ubuntu Delta or to ask for backports of fixes into active Ubuntu
Releases as part of the SRU process [4].

Do these pointers help enough to enable you to take the next step
reporting [5][6] this?

[1]: https://bugs.debian.org/cgi-bin/pkgreport.cgi?repeatmerged=yes=adduser
[2]: https://salsa.debian.org/debian/adduser
[3]: https://bugs.launchpad.net/ubuntu/+source/adduser
[4]: https://wiki.ubuntu.com/StableReleaseUpdates
[5]: https://www.debian.org/Bugs/Reporting
[6]: https://ubuntu.com/server/docs/reporting-bugs


> agape
> brent
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Issue: pppd and IPv6 interface identifiers

[Christian Ehrhardt]

On Fri, Jan 15, 2021 at 5:54 PM Alexander Velkov
 wrote:
>
> Hi,
>
> I want to point out an issue with the latest pppd version in Ubuntu 20.04. 
> There is also a patch attached in the issue which should disable the 
> 'ipv6cp-accept-local' option by default.

Hi Alexander,
indeed there is a chance for a merge with this and other fixes.
I'm unsure after all the time the other 2.4.7 behavior is in the field
how SRU-able that will be.
But I filed bug [1] for your problem and pinged the Desktop Team (who
usually looks after ppp) on IRC about it.

Kind Regards,
Christian

[1]: https://bugs.launchpad.net/ubuntu/+source/ppp/+bug/1912168

> https://github.com/paulusmack/ppp/issues/243
>
> Regards
> Alex
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: gambas3 3.14.3-2ubuntu3.1

[Christian Ehrhardt]

On Tue, Aug 11, 2020 at 3:59 AM Admin  wrote:

> Hi,
>
> Can you help me get the above package fixed. Installing *Gambas* results
> in a non functioning program. The problem seems to be that the list of
> files is incomplete.
>

Hi,
Due to the work for the bug report at [1] this actually works fine in the
coming 20.10 release without the massive listing of to-be-installed
packages.
Dependencies there are: gambas3 -> gambas3-ide -> gambas3-gb-form-print

With the package in Universe I'm not sure anyone is planning/working on an
SRU [2] for it, but the bug I linked should give you the right pointers if
you (or anyone else) want to do so.

[1]: https://bugs.launchpad.net/ubuntu/+source/gambas3/+bug/1876450
[2]: https://wiki.ubuntu.com/StableReleaseUpdates



> The command "*sudo apt-get -y install gambas3*" works but when you try to
> run the program it results in "*gbr3: unable to load component:
> gb.form.print*".
>
> The solution is to run the following command: -
>
> *sudo apt-get -y install gambas3 gambas3-gb-form-print
> gambas3-gb-gtk-opengl gambas3-gb-gtk3 gambas3-gb-gui-opengl
> gambas3-gb-gui-qt gambas3-gb-gui-qt-webkit gambas3-gb-gui-trayicon
> gambas3-gb-libxml gambas3-gb-option gambas3-gb-report gambas3-gb-sdl
> gambas3-gb-sdl-sound gambas3-gb-v4l gambas3-templates gambas3-gb-qt5 *
> *gambas3-gb-qt5-ext  **gambas3-gb-qt5**-opengl  **gambas3-gb-qt5-webkit*
>
> Gambas will now run.
>
> Thanks for any help you can provide.
>
> Charlie Ogier
> Admin Gambas.One <https://www.gambas.one/>
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>


-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Package version naming

[Christian Ehrhardt]

On Tue, May 19, 2020 at 3:25 AM Robin Hansson  wrote:
>
> Hello Ubuntu Developers,
>
> I wonder what the meaning is of of the package naming like ghc 
> 8.8.1+dfsg1+is+8.6.5+dfsg1-3
> When running "ghc --version" it says "version 8.6.5". Does the initial 
> "8.8.1" mean some stuff from 8.8.1 is back-ported? What about "dfsg1" and 
> "dfsg1-3"
> ?

Hi Robin,
multiple questions in one :-)

- most of the time +is or +really indicate a problem that forced the
people to go back
  e.g. they went to 8.8.1, but later found a big issue
  So they had to go back to 8.6.5 but since upgrades only work with
increasing values a version string like the one you see is
constructed.
  See 
https://www.debian.org/doc/debian-policy/ch-controlfields.html#epochs-should-be-used-sparingly

- dfsg -> removing file forbidden by policy
  See 
https://wiki.debian.org/DebianMentorsFaq#What_does_.2BIBw-dfsg.2BIB0_or_.2BIBw-ds.2BIB0_in_the_version_string_mean.3F

In general if in doubt, check the changelog of the package which is
accessible in your case with:
  $ apt changelog ghc

There you will for this example find:
  * 8.8.1+dfsg1-1~exp4 was mistakenly targeted to unstable.
   - use 8.6.5 repacked tarball to fix the issue

I hope that helps,
Christian

> Best,
> Robin
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss



-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Update qemu

[Christian Ehrhardt]

On Fri, Dec 27, 2019 at 3:54 PM Daniel Llewellyn 
wrote:

>
>
> On Mon, 23 Dec 2019 at 15:23, Chaython Meredith 
> wrote:
>
>> Why has qemu not been updated?
>> 4.2 out now. 4.1 since November.
>>
>
Hi,
we knew we wanted to go to 4.2 for 20.04
Extra merges of interim versions are quite some work without much gain and
in addition 4.1 had some issues discussed in regard to block handling that
I didn't want to trigger.

Furthermore history has told us that merging qemu along the libvirt release
that follows it (containing fixups as needed) is the best approach to not
leave the current dev version of Ubuntu in a bad state.
That would be libvirt 6.0 which isn't released yet and therefore holding
things back a bit (plus the actual time to do the merge work and plenty of
testing).



> Stable releases of Ubuntu maintain packages in the repositories for
> security-patches only; bumping versions beyond this is not a standard modus
> operandi. To bump the versions in stable releases we follow the Stable
> Release Updates documentation and rules:
> https://wiki.ubuntu.com/StableReleaseUpdates. You will need to justify
> why QEMU should be version bumped from 4.1 to 4.2 in the currently
> available Ubuntu releases. If 4.2 is available in Debian Unstable then
> Ubuntu 20.04 will include that version, or a later version if it arrives
> before the repository sync, when it is released in April.
>
> Dan.
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>


-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: The Kitty package in 19.10 is old

[Christian Ehrhardt]

On Wed, Jan 1, 2020 at 1:18 PM Behrang Saeedzadeh 
wrote:

> Hi,
>
> The current Kitty package in 19.10 is 0.14.3-1, released on 2019-07-29.
> There have been 5 releases since then with numerous new features and bug
> fixes, with latest release on 2019-12-21:
> https://sw.kovidgoyal.net/kitty/changelog.html
>

Hi,
the Kitty in 19.10 (sounds odd) is on 14.3 which is the version that was
released upstream at 29th July.
Version 14.4 was released on 31st August which was after the Feature Freeze
of 19.10 on 22nd of August.

Is it possible to publish the updates to the 19.10 repos?
>

Not as easily as you might think.
The next version of Ubuntu is 20.04 and already on Kitty 15.0 just missing
the most recent release 15.1 which probably is due to most  people being on
christmas break.
For 19.10 updates would have to follow the SRU policy [1] which means
individual, testable, backportable bug fixes.
Furthermore Kitty is "only" on universe [2] support so you'd also need to
spend some effort yourself or find a volunteer driving this.

[1]: https://wiki.ubuntu.com/StableReleaseUpdates
[2]:
https://help.ubuntu.com/community/Repositories/Ubuntu#The_Four_Main_Repositories

Thanks in advance.
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>


-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: strongswan 5.8.1 for ubuntu 18.04 LTS

[Christian Ehrhardt]

On Thu, Nov 14, 2019 at 11:55 PM Jean-Sébastien Frerot
 wrote:
>
> Hi, I would need to use strongswan 5.8.1 on 18.04 LTS with hwe (kernel 5+) 
> which supports xfrm interfaces. Do you guys think it would be possible to 
> have an official package for the latest version of strongswan?

Hi Jean-Sébastien,
there is the ongoing tradeoff between new and stable for any
Distribution and to some extent this is a question about that.
The SRU policy defines the general process of this tradeoff including
rare acknowledged exceptions.
Some packages get upstream stable releases by those exceptions, and
others like the kernel that you have already identified have -HWE
(Hardware Enablement) version that allow you to get newer code.
But strongswan has no such exception.
18.04 Bionic comes with Strongswan 5.6.2 and following [1] will only
get bug-fixes onto 5.6.2 instead of a move to a totally new version as
that would put way too much regression risk on every other user of it
in 18.04.

That said I'm working on 5.8.1 for Ubuntu 20.04 which also surely will
have kernel >= 5.3.

IMHO you have two ways for this now:
a) use 20.04 which seems to have all you need
b) put the 5.8.1 version I'll put into 20.04 soon into a PPA [2] of
your own and build it there for your Bionic use

I doubt there is enough need/gain to maintain an official backport in
[3] and even less for a permanent SRU exception to backport full
versions.
But the community is free to convince me otherwise :-)

[1]: https://wiki.ubuntu.com/StableReleaseUpdates
[2]: https://launchpad.net/+tour/ppa
[3]: https://help.ubuntu.com/community/UbuntuBackports

> Thank you.
>
> --
> Jean-Sébastien Frerot
> jean-sebast...@frerot.me
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at: 
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

-- 
Christian Ehrhardt
Staff Engineer, Ubuntu Server
Canonical Ltd

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: questions re: package qemu-system-sparc

[Christian Ehrhardt]

On Thu, Apr 11, 2019 at 8:32 PM  wrote:
>
> On 4/9/2019 at 1:59 PM, "Dan Streetman"  wrote:
>
> On Tue, Apr 9, 2019 at 1:33 PM  wrote:
> >
> > [Emailing here because it's the published maintainer address for Ubuntu 
> > package qemu-system-sparc. Please redirect me if I should look elsewhere.]
> >
> > From what I've been able to find within Ubuntu 18.10, the latest available 
> > version of qemu-system-sparc (and qemu in general) is 2.12.0 (Debian 
> > 1:2.12+dfsg-3ubuntu8.6).
> >
> > However the upstream QEMU project released v3.0.0 on August 14 2018, thus 
> > I'd hoped it would make it into Ubuntu 18.10. Is the delay simply due to a 
> > lack of resources?
>
> the next ubuntu release, disco, has qemu 3.1 included:
> $ rmadison -s disco -a source qemu
> qemu | 1:3.1+dfsg-2ubuntu3 | disco | source
>
> It is scheduled for release soon:
> https://wiki.ubuntu.com/DiscoDingo/ReleaseSchedule
>
>
> Thank you for the fast reply, and for the pointer to rmadison.--

Hi as well,
for the general selection e.g. to predict the upcoming versions you
can always check the qemu planning pages like [1][2].
And compare it to the timeline of the Ubuntu Releases [3][4].
Ubuntu is on regular schedules since almost forever so we can use that reliably.
Furthermore these days qemu is more and more on a regular release
pattern as well with years bumping the major number and a .0 version
in ~April.
Followed by a .1 version in August and most likely a .2 in December -
3.x was special, but it is usually ~4 months each release.
Combining the two and considering the constraints like Ubuntu Entering
Feature freeze we can try to forecast future versions will map like
this:

a) Ubuntu XX.04 => Qemu XX-16.Y (y being the last .y that qemu
released the year before)
b) Ubuntu XX.10 => Qemu XX-15.1 (If they make early August and no
hickups happen, otherwise .0 of April)

You already see that it is more of a stretch for the Ubuntu .10 releases.
Sometimes - actually most of the times looking at the past - you'd
want a libvirt release that is after the Qemu release to have it
handle the known hickups/changes well.
That means for the August Qemu you would often need the September
libvirt (monthly releases) and that is unfortunately too deep into
Ubuntu Feature Freeze - this is a case by case decision every cycle.

Following the SRU process [5] the qemu version in that Ubuntu release
will then stick and only be patched by individual bugfixes and
security updates.
In general, but even more so for LTS releases of Ubuntu, we also parse
the following qemu upstream development a few times to add further
patches for stabilization before releasing and then being under the
SRU process.

And if you want to look back in history rmadison (thanks ddstreet) or
just the Launchpads overview page of qemu [6] can tell you which qemu
is in which Ubuntu release.
Both of these techniques work for any package in Ubuntu by the way.

Finally there is also the Ubuntu Cloud Archive [7] which not always,
but often will carry the qemu/libvirt/... of the current Ubuntu
release - like 19.04 carrying qemu 3.1 which is in UCA-Stein - but as
backport to the latest LTS which is 18.04 at the moment.
This is mostly meant and supported for the context of a OpenStack
deployment, but depending on what you need/use it for that can be a
source for "new qemu on LTS" for you.

P.S. and for your particular question why qemu 3.0 isn't in Ubuntu
18.10 - which is a good question - this is one of the cases were
"early August" didn't work well for us.
This upstream release was just before the feature freeze (9 days), but
3.0 had too many changes that required extra work and we'd have wanted
the libvirt of September which was too late.
I didn't want to shove that onto Ubuntu users unprepared - so we
decided to stick with qemu 2.12 at that cycle.

P.P.S. Our planning on all of that is rather open and public, you'll
find a trello card like [8] for every cycle, the next one being [9].
There tasks, thoughts and references for the next releases virt-stack
will accumulate and be handled eventually.

[1]: https://wiki.qemu.org/Planning/2.12
[2]: https://wiki.qemu.org/Planning/3.1
[3]: https://wiki.ubuntu.com/BionicBeaver/ReleaseSchedule
[4]: https://wiki.ubuntu.com/DiscoDingo/ReleaseSchedule
[5]: https://wiki.ubuntu.com/StableReleaseUpdates
[6]: https://launchpad.net/ubuntu/+source/qemu
[7]: https://wiki.ubuntu.com/OpenStack/CloudArchive
[8]: https://trello.com/c/NigLYgS2
[9]: https://trello.com/c/r2zfPLnO

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: ZRAM & read_ahead_kb

[Christian Ehrhardt]

Hi Fred,
not really the wrong, but also not fully the right place.
As the name implies the list is more for discussions/announcements - but
you already have a very specific solution in mind.
So I'd recommend opening a but against the zram-config at
https://bugs.launchpad.net/ubuntu/+source/zram-config

OTOH if I remember correctly device specific readaheads are usually a
device specific udev rule, but I'd let that discussion better be part of
the discussion in the LP bug.

Also you actually might want to read-ahead if there is an efficiency bonus
in bulk uncompressing it.
And I'm not sure if it even uses "normal" page cache on this sort of
backing device.
So if you have any data to back up the need for it please add it to the bug.
​
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: ppa-purge

[Christian Ehrhardt]

On Mon, Nov 28, 2016 at 10:13 PM, Jarno Suni <j_s...@yahoo.co.uk> wrote:

> Would it be better to call it remove-apt-repository?
>

Just my opinion, but there are so many people already used and guides
written with ppa-purge that I'd not "rename" it.
I'd consider it a softer suggestion to:
- add an additional link under the new name to the ppa-purge executable.
- make it part of the ppa-purge package so everyone picks up the new name
- if you want a package "named" the new way consider a (virtual) package
that depends on ppa-purge (kind of the first step to renaming anyway)

In general renaming would be some extra effort see
https://wiki.debian.org/RenamingPackages for more details.

Well I'm not the maintainer there, so really only personal opinion of one
used to the old name :-)


Nor any feedback. https://answers.launchpad.net/ppa-purge/+question/404025
>

While not the best expert on this I think I could answer the question you
had there.
TL;DR - Target reference in your case is "master"


-- 
Christian Ehrhardt
Software Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Bug in ubuntu package for Mailman

[Christian Ehrhardt]

On Thu, Nov 17, 2016 at 5:18 PM, Philip Colmer <philip.col...@linaro.org>
wrote:

> It looks as if https://bugs.launchpad.net/mailman/+bug/1251495 has not
> been applied to the ubuntu package for Mailman. I have checked both
> 1:2.1.16-2ubuntu0.2 and 1:2.1.16-2, and Tagger.py has the bug in it.
> Hand-altering the file to implement the fix does correct the problem.
>

Hi Phillip,
thanks for your report.
The old bug unfortunately doesn't have a lot of information.
Would that be this Diff that is missing?:

@@ -71,7 +73,7 @@
 if hits:
 msgdata['topichits'] = hits.keys()
 change_header('X-Topics', NLTAB.join(hits.keys()),
-  mlist, msg, msgdata, Delete=False)
+  mlist, msg, msgdata, delete=False)

I wasn't sure so I didn't want to flag the bug improperly - but If so
please just add this snippet (or whatever else is the diff that is missing)
to the bug and set it to "new" again.
That should let it enter the normal pipeline to get handled and cleaned up.


-- 
Christian Ehrhardt
Software Engineer, Ubuntu Server
Canonical Ltd
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: unable to install libxrandr-dev

[Christian Ehrhardt]

On Mon, Nov 7, 2016 at 12:53 AM, Bryan White  wrote:

>
> The following packages have unmet dependencies:
>  libxrandr-dev : Depends: libxrandr2 (= 2:1.4.2-1) but 2:1.5.0-1~trusty1
> is to be installed
> E: Unable to correct problems, you have held broken packages.
>
> How do I fix this?


Hi Bryan,
just checked and was able to install as expected.
Actually for you libxrandr-dev should be on 1.5.0 from trusty-updated as
well.
The change is out for a while (2016-07-29), but to be sure you could run
'apt-get update' once more.
Then check why your system isn't going for libxrandr-dev 2:1.5.0-1~trusty1
with apt-cache policy.

IMHO it should these days look like:

root@trusty-tests:~# apt-cache policy libxrandr-dev
libxrandr-dev:
  Installed: (none)
  Candidate: 2:1.5.0-1~trusty1
  Version table:
 2:1.5.0-1~trusty1 0
500 http://archive.ubuntu.com/ubuntu/ trusty-updates/main amd64
Packages
 2:1.4.2-1 0
500 http://archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages
root@trusty-tests:~# apt-cache policy libxrandr2
libxrandr2:
  Installed: (none)
  Candidate: 2:1.5.0-1~trusty1
  Version table:
 2:1.5.0-1~trusty1 0
500 http://archive.ubuntu.com/ubuntu/ trusty-updates/main amd64
Packages
 2:1.4.2-1 0
500 http://archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages

Kind Regards,
Christian
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: canonical livepatch

[Christian Ehrhardt]

Hi,
just checked, with the same kernel mine still looks today like yours did
initially.
If run the status command with --verbose it will list the status it is in,
which might help seeing whats going on on your system.
​
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Kernel issues since -22

[Christian Ehrhardt]

Hi,
I happened to own a W520 together with hundreds of coworkers in the past.
There seems to be a reoccurring HW defect on some of the boards regarding
voltage control.
About 10% of our devices (which already is a lot statistically) had the
issue you describe.

After a while we realized weird stuff, like if you once tried to boot
without AC plugged you are likely go into the black-boot issues and have to
reboot quite some times until it works again.
Since this matches closely what you describe in your state with -22 it
might be the same.

Back then, when it still bootet "sometimes" I was able to insert a few
things in the boot process and we realized that it still boots in the
background, but the graphic device wasn't seen on the pci bus.
Neither by X.org nor by a lspci we inserted to write to disk. Now without a
graphic devices it couldn't show us much :-(

Given the raciness this issue had for us it is quite possible that slight
changes in the kernel initialization made it close to 100% to occur now.
The issue was bound to HW, so no reinstall or anything solved it for us -
one had a mainboard replacement which fixed it, but since nobody knows for
sure what it is I wouldn't count a 100% on that.


Christian Ehrhardt
Software Engineer, Ubuntu Server
Canonical Ltd

On Wed, Jun 22, 2016 at 10:06 PM, Nish Aravamudan <
nish.aravamu...@canonical.com> wrote:

> On 22.06.2016 [12:43:46 -0700], Dale Amon wrote:
> > I have not got a whole lot of information because this happens
> > so early in the boot. Basically, no kernel since -22 works on
> > my Lenovo W520. If I were to hazard a guess, I would say it has
> > to do with some race condition in the systemctl set up. Even
> > in -22 I often must boot several times before it succeeds, although
> > the failures in it seem to be different.
>
> That seems buggy in and of itself. You're sure the hardware is stable?
>
> > Has anyone else reported similar show-stopper issues in
> > recent kernels?
>
> I would probably check the existing bugs and maybe file one.
>
> -Nish
>
> --
> Ubuntu-devel-discuss mailing list
> Ubuntu-devel-discuss@lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
>
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Fwd: Xenial - OpenvSwitch with DPDK binding to 10G NIC, not working...

[Christian Ehrhardt]

Hi Thiago,
comments inline below ...

Christian Ehrhardt
Software Engineer, Ubuntu Server
Canonical Ltd

From: Martinx - ジェームズ <thiagocmarti...@gmail.com>
Date: Mon, Mar 7, 2016 at 2:28 PM
Subject: Re: Xenial - OpenvSwitch with DPDK binding to 10G NIC, not
working...
To: Ubuntu Core developers <ubuntu-devel-discuss@lists.ubuntu.com>


On 7 March 2016 at 16:40, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:

> Hey guys,
>
>  I'm trying to create two OVS bridges powered by DPDK, on Xenial, but I'm
> facing problems...
>
>  First, there is no documentation about how to do this... So far, I found
> the following:
>
>
>  - This link doesn't how to bind real NIC:
>
> https://software.intel.com/en-us/articles/using-open-vswitch-with-dpdk-on-ubuntu
>
>  - This one isn't specific to Ubuntu OVS + DPDK packages:
>  https://github.com/openvswitch/ovs/blob/master/INSTALL.DPDK.md
>
>  - Similar with above but, very old... Also not related to Ubuntu packages:
>  https://feiskyer.github.io/2015/03/01/open-vswitch-over-dpdk-on-ubuntu/
>
>
Yeah docs are often outdated, I have realized that as well.
Not because people are lazy, but because the environment is changing so
fast.
It is already planned to do a blog post or wiki or such along 16.04 release
to guide users of the very updated dpdk / openvsiwtch-dpdk.
But for now we are working on so many issues that things are changing too
fast to pre-write that document.

To add a bit I also found this one useful:
http://www.ran-lifshitz.com/2015/08/28/open-vswitch-netdev-dpdk-with-vhost-user-support-update/
But be aware of
https://bugs.launchpad.net/ubuntu/+source/dpdk/+bug/1546565
if you want to pass it to guests.
At least there the log entries will be pretty obvious with "permission
denied" on the vhost-user sockets.



>  So, to begin with, I start with Intel.com guide, it worked! But without
> real NIC.
>
>  Then, I tried to bind it to real NIC, like this:
>
>  Added two NIC PCI id to "/etc/dpdk/interfaces" file:
>
> ---
> pci :01:00.0uio-pci-generic
> pci :01:00.1uio-pci-generic
> ---
>
>  Then, reboot, I can see both NICs using "DPDK-compatible driver" by
> running: "dpdk_nic_bind --status".
>
>  Trying to create new OVS Bridges with DPDK devices, failing:
>
> ---
> root@xenial-1:~# ovs-vsctl add-br ovsbr0
> root@xenial-1:~# ovs-vsctl set bridge ovsbr0 datapath_type=netdev
> root@xenial-1:~# ovs-vsctl add-port ovsbr0 dpdk0 -- set Interface dpdk0
> type=dpdk
> ovs-vsctl: Error detected while setting up 'dpdk0'.  See ovs-vswitchd log
> for details.
>
[...]

> 2016-03-07T19:34:07.532Z|00136|bridge|WARN|could not open network device
> dpdk0 (No such device)
>
[...]

> Port "dpdk0"
> Interface "dpdk0"
> type: dpdk
> error: "could not open network device dpdk0 (No such
> device)"
>

I had seen something similar.
The particular one you are describing here was for me a device that was
either no bound by dpdk or had no compatible driver.
But according to what I see both should be ok for you.

Often I realized that the vswitch log alone isn't too helpful.
I recommend using journalctl along to get the EAL messages of dpdk as well.

To help you I created a merged log (time goes top down) which has both logs
and the command I used for a successful ovs-dpdk port add.
I think that will help you to see where your systems behaviour starts to
differ.

=> http://paste.ubuntu.com/15326358/

In your case you might e.g. see "skipped device" there leading eventually
to your dpdk0 not found issue.

Also you might already have done that, but just not mentioned. Still to
help as much as possible - have you set DPDK_OPTS, I for example did that
before (re)starting it?
For example
echo "set DPDK_OPTS='--dpdk -c 0x1 -n 4 -m 4096'" >>
/etc/default/openvswitch-switch


[...]

I have also seen the part about vfio-pci in your mail and thank you for
reporting the bug, I'll handle work and communication of this issue in the
bug.
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss