apport uses sys.argv instead of named arguments

Um, we can do that. But I think there's a problem with containers and compatibility in forwarding messages over the socket (think new container, old host).

I wonder if /proc/sys/kernel/core_pattern is per-namespace too. If it is, that would be great. If not, a new host with old containers would break too: The containers would not understand the arguments. Specifically, containers need to get the host_pid in sys.argv[5], they do not care about the other arguments, so these could be -- options like that:

--pid=%p --signal=%s --core-size=%c --dump-mode=%d %P

but we could not add more options then either, because the forwarding only works with 6 arguments.

So the situation is:

- We need to keep the argument count at 5-6 no matter what, otherwise old containers in new hosts break
- In containers, we do not parse any arguments except for 5.
- The core_pattern is defined on the host.

I suggest we just split the command-line by "," as well. So if we have

--pid=%p --signal=%s --core-size=%c --dump-mode=%d %P

we can extend that to

--pid=%p --signal=%s --core-size=%c --dump-mode=%d,foo-bar=%d %P

or we just drop the -- alltogether:

pid=%p signal=%s core-size=%c dump-mode=%d,foo-bar=%? %P

That should work:

host=old, container=old => everything as before
host=old, container=new => core_pattern as before, no effective changes
host=new, container=old => new core_pattern, container just forwards to host, host does parsing
host=new, container=new => fine too

Let's just generalize that into

--version2 --version2 --variables pid=%p,.... %P

I'm sorry I think I lost something here. What does --version2 mean exactly?

It's just a silly fill up argument so we end up with 5/6 arguments to make old clients in containers happy. And well, it easily allows you to distinguish the new commandline from the old positional style :)

WiP merge proposal in https://code.launchpad.net/~juliank/apport/named-arguments/+merge/337590

OK, so that's not all: core_pattern itself only stores 128 bytes.

Just to summarize the situation: It seems that if you have apport installed inside an LXC container, that apport creates reports, rather than doing the forwarding dance. Unfortunately, the core_pattern does not know about name spaces.

So if we change the host to use a new pattern with named arguments, older apport in containers is broken - and needs to be SRUed. So this needs some sort of flag day transition. To minimize the impact of the transition, a good strategy seems to be the following:

1. Upload apport that understands named arguments and SRU it everywhere
2. Wait some time
3. Change core_pattern in devel
4. Wait some time
5. Change core_pattern everywhere else

Right?

Given what Tyler said about "magic arg ordering being dangerous" I wonder if we could instead do that is a security update simultaneously for all releases thereby avoiding the multiple uploads and flag day transition.

If we do that, I'd say we built them as security updates, copy them to proposed, and then once they are fit for release, copy them to the -security in addition to -updates; so we get the usual SRU checks.

I still think that at least rolling out support first via an SRU makes sense.

This bug was fixed in the package apport - 2.20.10-0ubuntu23

---------------
apport (2.20.10-0ubuntu23) disco; urgency=medium

  * Fix python coding style issue introduced in previous upload.

 -- Julian Andres Klode <email address hidden> Tue, 05 Mar 2019 10:35:17 +0100

I'll go ahead and sponsor these today.

Updated debdiff and made sure autopkgtest were successful on both xenial and bionic

Sponsored for bionic.

Also sponsored for xenial.

Hello Brian, or anyone else affected,

Accepted apport into bionic-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/apport/2.20.9-0ubuntu7.18 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation on how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, what testing has been performed on the package and change the tag from verification-needed-bionic to verification-done-bionic. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-bionic. In either case, without details of your testing we will not be able to proceed.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance for helping!

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

Hello Brian, or anyone else affected,

Accepted apport into xenial-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/apport/2.20.1-0ubuntu2.25 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation on how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, what testing has been performed on the package and change the tag from verification-needed-xenial to verification-done-xenial. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-xenial. In either case, without details of your testing we will not be able to proceed.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance for helping!

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

All autopkgtests for the newly accepted apport (2.20.9-0ubuntu7.18) for bionic have finished running.
The following regressions have been reported in tests triggered by the package:

apport/2.20.9-0ubuntu7.18 (amd64)

Please visit the excuses page listed below and investigate the failures, proceeding afterwards as per the StableReleaseUpdates policy regarding autopkgtest regressions [1].

https://people.canonical.com/~ubuntu-archive/proposed-migration/bionic/update_excuses.html#apport

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

Thank you!

Verification completed on both bionic and xenial using proposed pockets

This bug was fixed in the package apport - 2.20.1-0ubuntu2.25

---------------
apport (2.20.1-0ubuntu2.25) xenial; urgency=medium

  * data/apport: Introduce support for non-positional arguments so we can
    easily extend core_pattern in the future (LP: #1732962)

 -- Matthieu Clemenceau <email address hidden> Fri, 21 Aug 2020 10:24:13 -0500

The verification of the Stable Release Update for apport has completed successfully and the package is now being released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.

This bug was fixed in the package apport - 2.20.9-0ubuntu7.18

---------------
apport (2.20.9-0ubuntu7.18) bionic; urgency=medium

  * data/apport: Introduce support for non-positional arguments so we can
    easily extend core_pattern in the future (LP: #1732962)

 -- Matthieu Clemenceau <email address hidden> Fri, 21 Aug 2020 10:12:55 -0500