However, when I load the "cdparanoia" package, k3b stops complaining and does what I expect it to do - evidently a packaging error in k3b

Added a Requires for cdparanoia, can you please test ?

(In reply to comment #2)
> Added a Requires for cdparanoia, can you please test ?

The fixed package seems to be in cauldron but this bug report is against Mageia 1. Can you also provide an update candidate in updates_testing and once you think the update is ready send it to qa-bugs@ml.mageia.org?

Uploaded to updates_testing. Please install and let us know.

so reassign to QA

CC: (none) => juan.baptiste
Assignee: juan.baptiste => qa-bugs

I uninstalled k3b and cdparanoia (with rpm -e --nodeps), then installed
k3b, which pulled in cdparanoia.  Quick testing shows it can read the
audiocd.

Testing complete on i586.

One thing I noticed was that the device notifier only gave the option of
playing the audiocd with amarok.  It should also show the option of
opening k3b with the cd.  However testing with the Core Release version
shows it has the same problem, so this is not a regression.

When pushed to updates, cdparanoia will have to be linked to core update,
along with libcdda0.

CC: (none) => davidwhodgins

I am unable to replicate the original problem on x86_64
I am using k3b-2.0.2-2.mga1
On inserting an audio CD and selecting Tools>CopyMedium>Start the CD copy progresses normally without any request for libcdparanoia.
I have checked and do not have cdparanoia already installed.
In k3b setup cdparanoia does not appear in the list of required programmes.

Looking at the k3b web site it says cdparanoia-devel is a requirement but does not list cdparanoia itself.

cdparanoia is not a requirement in Mandriva.

I'm confused.  Does k3b need cdparanoia or not?

CC: (none) => derekjenn

> I'm confused.
You are not alone

> Does k3b need cdparanoia or not?
Evidence seems to say yes (in case you copy an audio CD): If I start trying to copy an audio CD without having cdparanoia installed I immediately get the "Could not load libcdparanoia" error message (see comment 1). But if I have cdparanoia installed prior to calling k3b, k3b starts happily reading my audio CD.

At the time I had filed the bug (+ the work-around to manually load cdparanoia), things then went perfectly: after reading the original audio CD, k3b asked for an empty CD and then wrote it.

In the meantime, there is additional regression: after completion of the read of the original CD, k3b still asks for an empty CD to be inserted, but then does not see the empty CD and is stuck with the "mount an empty CD" dialog - no way to work around this new problem. Disgusted.

(using k3b-2.0.2-2.mga1, i586)

Being disgusted won't help to fix the problem or do anything towards inspiring the maintainer, who I am sure is trying hard on your behalf.

Looking at the K3B website it might be missing HAL support.

If you insert a blank CD and then start K3B, does it see the blank CD?

Created attachment 929 [details]
console output from k3b

> Being disgusted won't help to fix the problem or do anything towards inspiring
> the maintainer, who I am sure is trying hard on your behalf.

You got that wrong - that has nothing to do with claiming features of personal interest or with assessing the work of a maintainer. The purpose of my post was, in response to the post of Derek, to state that (a) cdparanoia appears to be required, (b) the problem is still present as originally described and (c) that now there is an element of regression.

Sorry if 9 months later my frustratrion seeped in - I sincerely regret if disgust is too strong a word (I am not a native English speaker) - what other word do you suggest? Submitting a bug, commenting it, suggesting a workaround - trying to put in evidence facts that help a maintainer to fix the problem takes hours and hours: that is not at all the negative attitude you appear to resent. Nevertheless: using Mageia to copy an audio CD should not be a problem today. 

Back to facts: what can we constructively do to provide evidence as a base to coming nearer to a solution?

> If you insert a blank CD and then start K3B, does it see the blank CD?
Yes:
- I get the "Available devices" popup after inserting the empty disk 
- the "Disk copy" page of k3b displays that there is an empty CD and shows its size
- once the reading of the original is accomplished and I am prompted to insert an empty disk, I get again a "Available devices" popup; should not k3b grab that event?

If I run k3b from a console, there are several error messages whose importance or relevance wrt this topic I cannot judge (they appear once I insert the CD to be copied and hit "Start") see attachment 929 [details]

> 
> Sorry if 9 months later my frustratrion seeped in - I sincerely regret if
> disgust is too strong a word (I am not a native English speaker) - what other
> word do you suggest? Submitting a bug, commenting it, suggesting a workaround -

You aren't taking into account that we are a new distro, we are still organizing our selves and many many many packages didn't have a maintainer set until less than a week. This means that many bugs went unnoticed until someone took maintainership of those packages as I did, last week. Now that the package has a maintainer there's someone working actively on fixing it ;)

> Back to facts: what can we constructively do to provide evidence as a base to
> coming nearer to a solution?
> 
> > If you insert a blank CD and then start K3B, does it see the blank CD?
> Yes:
> - I get the "Available devices" popup after inserting the empty disk 
> - the "Disk copy" page of k3b displays that there is an empty CD and shows its
> size
> - once the reading of the original is accomplished and I am prompted to insert
> an empty disk, I get again a "Available devices" popup; should not k3b grab
> that event?
> 

This seems to me to be completely unrelated to this bug report, if this is the case please open a new bug report and post/attach all the needed info to debug it.

Thanks.

I can confirm the bug where a disk is not recognised unless it is in the drive when the program is started. That appears to be the HAL which is missing. 

It is not just when a copy is being made and new CD inserted. It also doesn't recognise a disk is present if it is inserted after the program has started. It *must* be present before it is started, which is a bit of a problem considering it's function.

Start k3b, insert audio cd, click copy. Nothing recognised as being in the drive.

Insert audio CD, start k3b, click copy. It recognises the CD and even lists it by title.

Juergen, did you create a bug for that?

Hardware: i586 => All

(In reply to comment #13)
> I can confirm the bug where a disk is not recognised unless it is in the drive
> when the program is started. That appears to be the HAL which is missing. 
> 
> It is not just when a copy is being made and new CD inserted. It also doesn't
> recognise a disk is present if it is inserted after the program has started. It
> *must* be present before it is started, which is a bit of a problem considering
> it's function.
> 
> Start k3b, insert audio cd, click copy. Nothing recognised as being in the
> drive.
> 
> Insert audio CD, start k3b, click copy. It recognises the CD and even lists it
> by title.
> 
> Juergen, did you create a bug for that?

So indeed this is another issue, please open another bug report for this.

For the original issue, can someone confirm if it is fixed ?

> Juergen, did you create a bug for that?
Not yet.

Subjectively: I like to work in a peaceful environment and doubted whether arguments should not calm down a bit - I think they have (-.

Objectively: (1) I wanted feedback - like yours - to avoid submitting a bug on a local problem; and (2) I remember a (now resolved, have to dig for its number) bug on urpmi not grabbing the disk it expects to be inserted - wondering whether this is not a very similar problem - or just a common issue of integration between HAL and applications? (3) I am reluctant to increase my "zoo of open bugs", think twice before opening a new one.

But I agree: this appears to be of sufficient importance and solid - I will submit the bug (essentially use your wording) - on what, on HAL? I know very little about HAL, can submit a bug but not substantially contribute to its discussion.

Testing x86_64:

Before
------

Installed from core/release. It pulled in cdrdao from Tainted.

(medium "Core Release")
  k3b                            2.0.2        2.mga1        x86_64  
  lib64k3bdevice6                2.0.2        2.mga1        x86_64  
  lib64k3blib6                   2.0.2        2.mga1        x86_64  
  normalize                      0.7.7        9.mga1        x86_64  
(medium "Tainted Release")
  cdrdao                         1.2.3        5.mga1.taint> x86_64 

I don't get the error message about cdparanoia.

$ rpm -qa | grep cdparan
gstreamer0.10-cdparanoia-0.10.32-3.mga1

$ urpmq -i gstreamer0.10-cdparanoia

Source RPM  : gstreamer0.10-plugins-base-0.10.32-3.mga1.src.rpm

When I simulate the copy, it first creates an image and then asks for a blank disk. As I don't have a blank CD I tried with a blank DVD instead.

K3b did then recognise a blank DVD had been inserted which is strange.

Checking that again, it seems it is working correctly too! There was an unfortunate communication error between QA via ssh and QA's other half who is sat at the computer, disk in hand!

$ ps aux | grep -i hal
17        1795  0.0  0.1  43816  2344 ?        Ssl  07:38   0:00 hald
root      2017  0.0  0.0  22464   800 ?        S    07:38   0:00 hald-runner
root      2126  0.0  0.0  24612   564 ?        S    07:38   0:00 hald-addon-input: Listening on /dev/input/event0 /dev/input/event2 /dev/input/event3 /dev/input/event4
root      2152  0.0  0.0  24584   908 ?        S    07:38   0:05 hald-addon-storage: polling /dev/sr0 (every 2 sec)
root      2161  0.0  0.0  24592   252 ?        S    07:38   0:00 /usr/lib64/hal/hald-addon-cpufreq
17        2162  0.0  0.0  20052   316 ?        S    07:38   0:00 hald-addon-acpi: listening on acpid socket /var/run/acpid.socket


So, to summarise, I am not able to reproduce either bug.


After
-----

Installed from core/updates_testing.

# urpmi k3b
To satisfy dependencies, the following packages are going to be installed:
   Package                        Version      Release       Arch   
(medium "Core Release")
  cdparanoia                     10.2         7.mga1        x86_64  
(medium "Core Updates Testing")
  k3b                            2.0.2        3.mga1        x86_64  


Everything works as before.


As Juergen confirmed in comment 8 that this bug is fixed by the cdparanoia require I will validate.

Juergen, if you still experience the problem with your disk not being recognised please create a new bug, but I am unable to reproduce, likely because HAL is running on my system.



Advisory
--------
K3b would complain it was missing cdparanoia when attempting to copy an audio CD. A require has been added for cdparanoia to correct this.
--------

Source RPM: k3b-2.0.2-3.mga1.src.rpm


Could sysadmin please push from core/updates_testing to core/updates.

Thankyou!

Keywords: (none) => validated_update
CC: (none) => sysadmin-bugs

Sysadmin please also link cdparanoia from core/release to core/updates when this is pushed.

Thankyou.

I did some more checking to get a solid base for the description of a bug to file on HAL - very few solid facts to go upon. Here is what I see on my system.

1. When I insert a cdrom (empty or not), most times I get a device notifier popup window, but not always: when I commented the other day, I remembered that fact, but could not reproduce it a single time - today it is the other way round. I should check how this is related to what happend since re-booting

2. When I first insert a cdrom and then - even without getting the popup window - start k3b, k3b correctly reports this cdrom (after hitting the "Copy Medium" button in the k3b main window).

3. The same thing if I first launch k3b, then hit the "Copy Medium" button and - getting the "Please insert a non-empty medium" button - insert my cdrom: the label "Please insert..." toggles to a button a description of the disk I just inserted - that is opposed to what Claire has reported.

4.Scenario described the other day: if after 3 (with "use ...") I pursue and hit Start, everything runs smoothly until I get the "Please insert an empty CD medium" popup: k3b remains stuck, does not see the empty medium (but - the first time today - I get the "Available Media" popup for the empty cdrom.

Weird. Enough for a rant - but is it enough for somebody to start debugging?

(2), (3), and (4) appear to be totally reproducebable on my system (meaning I do not remember a single occasion where the system behaved differently), (1) is as random as can be.

I know, #2417  is not an ideal place for this discussion - but where should we have it in preparation of a bug not yet defined? suggestions?

I could submit a bug report on strictly (4) - there is agreement that this is reproducable. The input from Claire could be added as additional info, along with a reference to the (re-opened) bug #145 which is in support of what Claire is seeing - on my system the problem addressed by #145 has gone away.

Just to document what is running here:
<harms@pcjuergen ~> ps -A | grep hal
  940 ?        00:00:00 hald
 1012 ?        00:00:00 hald-runner
 1059 ?        00:00:01 hald-addon-inpu
 1068 ?        00:00:00 hald-addon-acpi
 1283 ?        00:00:07 hald-addon-stor

Apropos: does HAL log events somewhere? there is nothing in /var/log/syslog

(In reply to comment #18)
> 
> I know, #2417  is not an ideal place for this discussion - but where should we
> have it in preparation of a bug not yet defined? suggestions?
> 

In it's own bug report, what makes you think that another bug not related to the problem is the best place to discuss it? it doesn't matter if the info to debug your problem is complete or not at this moment, this has nothing to do with this bug report. Please don't continue to pollute this bug report with stuff that has nothing to do with it.

Open a new bug report, put all of this there and wait and see what happens.

I will not submit a badly prepared bug that has little chance to get resolved - the definition of a bug defines a strict scope for the problems to be solved. End of polution.

As you wish, but as the maintainer of the package I ask you to NOT continue posting unrelated info on this bug. The correct way is to open a NEW bug report and discuss it there, no matter is complete or valid etc.

Please be respectful of the way we handle bug reports if you want us to be able to fix them effectively.

Sysadmin please see comment 16 for advisory and SRPM and comment 17 for required linking.

Thankyou

Update pushed.

Status: NEW => RESOLVED
CC: (none) => tmb
Resolution: (none) => FIXED