CONCURRENCY=startpar doesn't work

Bug #96851 reported by Evan Carroll
4
Affects Status Importance Assigned to Milestone
upstart
Invalid
Undecided
Unassigned
sysvinit (Ubuntu)
Fix Released
Medium
Unassigned

Bug Description

apparently Ubuntu uses upstart, however /etc/init.d/rc has an option to enable concurrency
CONCURRENCY=startpar as mentioned in the comment.

"=startpar" is *only* valid with sysvinit, and will result in a loop of errors on boot, and a brick of the system, (because even in non-multiuser run levels root is -l, which is a dumb idea you would think there would be a way to unlock root when not in run level 3-5)

Tags: upstart

Related branches

Revision history for this message
Mike Dahlgren (dahlgren) wrote : Re: /etc/init.d/rc has critical typo.

If I understand this correctly the default option is set to CONCURRENCY=none, and the user has the option to set it to "startpar". You say this is a critical issue because someone could change it to a broken state. By that token I believe many of the configuration files on the computer could easily be considered issues, for they often have suggestions in there comments that would break someone's computer. I do not see this as an issue.

        ~ Thanks

Revision history for this message
Evan Carroll (evancarroll) wrote :

Thats because you're an argumentative moron. It's a critical issue because it's a conf option that will mudbrick the computer, *and* because the comments lead the user into believing that an invalid option, is in fact a valid one. If upstart doesn't support CONCURRENCY, there is no reason to have a "default" value, or a comment to begin with.

If there are other things in conf files that could coerce a user into mudbricking his box, and that serve no legitimate purpose, please report them, for they are also bugs in the most conservative meaning of the word.

Evan

P.S. Thanks for the ad absurdum.

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

Upstart has nothing to do with whether this works or not; this is likely just a general bug with that script.

Changed in upstart:
status: Unconfirmed → Rejected
Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

You haven't provided evidence of the "typo", which suggests a simple mistake in the file -- could you explain what it is?

Revision history for this message
Evan Carroll (evancarroll) wrote : Re: [Bug 96851] Re: /etc/init.d/rc has critical typo.

# Specify method used to enable concurrent init.d scripts.
# Valid options are 'none', 'shell' and 'startpar'
CONCURRENCY=none

On 3/28/07, Scott James Remnant <email address hidden> wrote:
> You haven't provided evidence of the "typo", which suggests a simple
> mistake in the file -- could you explain what it is?
>
> --
> /etc/init.d/rc has critical typo.
> https://launchpad.net/bugs/96851
>

--
Evan Carroll
System Lord of the Internets
<email address hidden>
832-445-8877

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote : Re: /etc/init.d/rc has critical typo.

Sorry ...

I still don't see the typo there, all of those words appear to be spelled correctly.

Revision history for this message
Mike Dahlgren (dahlgren) wrote :

The "problem" that Evan is describing is that the comment for CONCURRENCY, has three options listed, and only two of them work. If you boot with CONCURRENCY=startpar, it will probably throw an error and boot to a command prompt. Although this option doesn't currently "work" it doesn't mean that it is not valid option, and I don't think the comment should be changed.

Also if Ubuntu was to go through and try to "fix" every config file that a user could possible break, then there would be no resources left to actually work on things that do matter.

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

ah yes, we don't ship startpar

description: updated
Changed in sysvinit:
importance: Undecided → Medium
status: Unconfirmed → Confirmed
Revision history for this message
Evan Carroll (evancarroll) wrote : Re: [Bug 96851] Re: /etc/init.d/rc has critical typo.

Jesus Christ, launchpad is full of oafs. A typo can encompass more
than a spelling error; but, if you insist on being hard to work with,
"invalid" is obviously spelt wrong.

Evan.

On 3/28/07, Scott James Remnant <email address hidden> wrote:
> Sorry ...
>
> I still don't see the typo there, all of those words appear to be
> spelled correctly.
>
> --
> /etc/init.d/rc has critical typo.
> https://launchpad.net/bugs/96851
>

--
Evan Carroll
System Lord of the Internets
<email address hidden>
832-445-8877

Revision history for this message
Evan Carroll (evancarroll) wrote : Re: [Bug 96851] Re: CONCURRENCY=startpar doesn't work

Thank you Scott James Remnant, for helping establish validity. Your
cooperation is much valued by the community, now go forth and rub off
on your peers.

Evan.

On 3/28/07, Scott James Remnant <email address hidden> wrote:
> ah yes, we don't ship startpar
>
> ** Summary changed:
>
> - /etc/init.d/rc has critical typo.
> + CONCURRENCY=startpar doesn't work
>
> ** Description changed:
>
> - Binary package hint: base-files
> -
> apparently Ubuntu uses upstart, however /etc/init.d/rc has an option to enable concurrency
> CONCURRENCY=startpar as mentioned in the comment.
>
> "=startpar" is *only* valid with sysvinit, and will result in a loop of
> errors on boot, and a brick of the system, (because even in non-
> multiuser run levels root is -l, which is a dumb idea you would think
> there would be a way to unlock root when not in run level 3-5)
>
> ** Changed in: sysvinit (Ubuntu)
> Importance: Undecided => Medium
> Status: Unconfirmed => Confirmed
>
> --
> CONCURRENCY=startpar doesn't work
> https://launchpad.net/bugs/96851
>

--
Evan Carroll
System Lord of the Internets
<email address hidden>
832-445-8877

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

Evan, if you wish to continue to participate in the Ubuntu Community, please read the following very carefully:

http://www.ubuntu.com/community/conduct

It is the Code of Conduct by which members of this community rule themselves; if you are unable to do so, I will have to ask that you are banned from this bug tracking system since your comments are extraordinarily rude.

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :
Changed in sysvinit:
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.