I'm sorry, but there isn't anything we can without more information. For the crash, we would need a backtrace. For the saving issue we need the complete steps to reproduce the issue.

Thank you for your reply.
I didn't collect the backtrace.
The one think I can say immediately is that ALL the troubles I have found with krita 4.x are _only_ happening when I use any krita 4.x *appimage* with my Wacom Cintiq display tablet. That is regardless of my system being kubuntu 16.04 or kubuntu 18.04.

To make it as clear as possible:

1. It is not possible to use a Wacom Cintiq 27QHD Graphic Tablet in (K)ubuntu 16.04 or 18.04 with any appimage version of Krita >= v4.
2. If Krita v4.x is installed using the kritalime ppa, then it is possible to work, only with a couple of very minor issues, but with no significant issues.

So: As soon as I finally was able to use the kritalime ppa and installed krita 4 from there, I confirmed that none of the the worst issues I have been experiencing with the appimage versions happened. But because I have been wasting so much time in trying to eliminate any possible causes from my side, and in the process nearly drying out my adrenal glands of any cortisol they might have left, with no help, you would understand what a relieve I had when I finally found a way to work with krita >= 4 and the Wacom Cintiq tablet.

Whether this makes any sense for a bug report I don't know any more, but whenever I try any appimage of 4.x.x, it will crash and/or provoke strange issue as reported in this thread, as soon as I get my bare fingers on the touchscreen. So: yes, I can reproduce the thing. The question is, I would have liked a bit of guidance as to how to proceed, as in another bug report I asked, but unfortunately you guys are overwhelmed with so much work for a small but brave team of talented devs, that I can't expect you to prioritize this request.

I do understand that I might be the only person on the planet that spent a few thousand euros on a Wacom Cintiq 27QHD Graphic Tablet to use it with Krita exclusively, and in Linux, although I do have a Windows partition with Photoshop in it. But I rather used Krita v3.x than Photoshop: it's been more than one year since I haven't logged on to Windows and I still think it's not worth doing that.

So fixing whatever it is in appimages that doesn't like Wacom Cintiq tablets for only one user doesn't make any sense, especially when the installed versions from kritalime ppa do work reasonable well.

Believe me, I am not going to boot any time soon my Windows partition, let alone switching to my (working) Photoshop CC2015.  And to anyone interested and wondering: Yes, any cent you would invest in a Wacom Cintiq is absolutely the best investment you would have done in many years! You won't regret. You have to try some day. In Krita (from ppa)!

Dear Bug Submitter,

This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days, the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information.

For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging

If you have already provided the requested information, please set the bug status as REPORTED so that the KDE team knows that the bug is ready to be confirmed.

Thank you for helping us make KDE software even better for everyone!

I have been trying various ways to reproduce the "Could not save (...) Failed to save the annotations for layer...", without success. The problem is that the original conditions are not met any more: The problem happened with Krita v4.1.0 and in the meantime I had to upgrade to v4.1.2, so this might mean either the problem was related to v4.1.0 OR it takes some conditions that I haven't managed to reproduce (yet).
HOWEVER, I did manage to reproduce the second part of the problem and which is already described in bug #392251, and I have captured backtrace for that. I am attaching all that in this thread and also in bug #392251, because I cannot exclude that there might be something useful for BOTH bugs, i.e. something that you can see "should not be happening" that might help in either bug.

So please don't close this bug until you have checked the backtrace and confirmed that it doesn't help THIS particular bug. If that is the case, since it can't be reproduced easily, this might need to be closed without a conclusion.  Promised Information in my next post below.

Created attachment 115312 [details]
Hangup backtrace files - three cases

Attached three backtrace files corresponding to three procedures, two of which caused hangup/crash, one that did not, as a control.

Steps to reproduce and references to the files:
Experiment 1
- 1A: (file: Exp1A_Krita_Hangup_Backtrace.txt)
1. Preparation: Run the following in the terminal:
   xsetwacom -v --set 'Wacom Cintiq 27QHD touch Finger touch' gesture off
2. Run (with dbg) krita-4.1.3-x86_64.appimage (or krita-4.1.1-x86_64.appimage).
3. Open some krita document, example: teste.kra (created originally with krita v3.3)
4. Draw a couple of color lines using the Stylus (this is optional, problems happens regardless).
5. Touch the tablet touchscreen with a bare finger. At this exact moment, krita crashes/hangs(?). The GUI becomes irresponsive and doesn't refresh any longer.  Collect the backtrace (needs to do <Ctrl><C> to get the gdb prompt) and close gdb to kill the hanged krita.
- 1B: (file: Exp1B_Krita_NOHangup_Backtrace.txt)
6. Open (with dbg) krita v4.1.2 (installed from kritalime ppa) in /usr/bin/krit. NOPROBLEM 
7. Open a different krita document created with krita 4.1.1 from kritalime, which has reference images embedded in it.
8. Touchscreen works normally, as this is not an appimage. Draw for a while.
9. Add a reference image to the existing ones.
10. Draw for a while.
11. Save the file: <Ctrl><S> => No problem. [ here it would have given the Error.]
12. Collect the backtrace "Exp1b" and quit krita

Experiment 2
(file: Exp2_Krita_Hangup_Backtrace.txt)
1. If preparation of Experiment 1, step 1 was done, no action needed here. Else, do that.
2. Run krita (dbg) krita-4.1.3-x86_64.appimage (or krita-4.1.1-x86_64.appimage).
3. Open some krita document, example: teste.kra (created originally with krita v3.3)
4. Draw a couple of color lines using the Stylus (this is optional, problems happens regardless).
5. Touch the tablet touchscreen with a bare finger. It did not crash/hang and zoom gesture was recognized. Draw and do touch gestures, all seem OK. Close the document, but let Krita open.
--> Problem will happen now:
6. Open another document. (in our case a trivial krita document (with only few lines) created with krita v4.0.4).
7. Repeat steps 4 and 5. Krita crashes/hangs as soon as touchscreen is touched with bare finger.
Collect the backtrace (needs to do <Ctrl><C> to get the gdb prompt) and close gdb to kill the hanged krita.

Changing the status to REPORTED

Sorry for the late reply. Apparently I missed a lot of bug mail last autumn.

All your backtraces are in Qt's xcb implementation. There are two sides to this: the first is that Qt has rewritten its xcb implementation for Qt 5.12, and the second Ubuntu might have problems with some library versions. The backtraces don't show a problem in Krita's code.

I run Kubuntu 18.04 on my Wacom Mobile Studio, which pretty much similar to running a cintiq attached to a linux computer, and I don't get these problems.

We currently have appimage builds of Krita 4.2 pre-alpha, with Qt 5.12, which means you could give those a try and check whether this works better on your system: https://binary-factory.kde.org/job/Krita_Nightly_Appimage_Build/

Sadly, this project behaves like a one man's show, despite of what other enthusiastic contexts say, where even big names like Intel or Google are mentioned.  The truth is, when the leader is down (for understandable and worrying reasons), there is a huge impact. Not a criticism, it's hard reality, as has been rather rudely expressed several times in comments in bug reports.

Some hard facts that don't play nicely about krita's "team" response to this serious issue:

1. Despite overwhelming demonstrations that this problem will (and I bet that it does) affect others in a valid use case, there has been no noticeable interest in this bug.

2. Not only there is lack of user's interest or, more likely, lack of knowledge and stomack for entering the ugly world of bugs reports. The problem extends more importantly to krita's team.  "Doesn't affect me? Then too bad."

Fact, there are not enough people to even look at what has cost me dozens of hours of troubleshooting and reporting, late at night, making huge efforts to painfully demonstrate a real problem in real case scenario.  It is announced in some places that there is a team, but there is only one person looking at this bug report.  With indeed only one person looking at this, it is impossible to manage the workload, and one could hardly blame such overwhelmed person, only wish him to take care of his health seriously. This is absolutely insane.

3. I have lost the count of the times I have been told that this is not a krita's code problem.
Frankly, it's beyond irritating that such statement would be thrown in a bug report. Who is interested in fingerpointing? How old are we, 6?  Who's problem it is that your product doesn't work in a VALID use case? Not a fantasy, not an edge case. A normal case. If krita doesn't recognize the strokes of my toothbrush on my monitor, I won't blame you or open a ticket on Oral-B's customer support or on the supermarket where I bought it for 2.35 Euros.

I can't say this clearer or frequently enough: I do not care "who" forgot a semicolon or who wrote an idiotic API or 3rd party program or which criminal manufactures a disastrous piece of hardware.  As long as that thing is required and supported by your product, then you own that problem. Your product doesn't work in a __valid_use_case__.  If I wouldn't have proven a *valid use case*, I wouldn't be wasting my time here, late at night and with bad health.

Why is it so hard to say: "yes, it is _our_ problem, because it affects _our_ product and _our_ users." ? I will never get it.  Have you ever had a problem with your new car? would you accept they do nothing because they blame China, which produces such bad aluminium that your engine cracked? What about your stinky smoking "non-polluting" Volkswagen?

Whether the root cause is bad Ubuntu, bad KDE, bad Qt, Wacom, Nvidia, bad Goverment, the NSA or climate changes: the only and absolute workable truth is that krita is _the_thing_ that doesn't work in a well documented, normal _valid_use_case_scenario_, demonstrated to exhaustion, literally.  I don't see any interest to admit it and _document_ it for the existing and potential users.  It seems it isn't interesting except to myself.

No matter how moronic Linux, Ubuntu, Debian or Qt might be and how clean and elegant krita's code might be, it is krita's team lack of taking ownership of well demonstrated, well described valid failure modes and not providing vital information in the right places, what deeply frustrates any good intentions and all hard efforts I have made to help.  No, it has nothing to do with work load, not after nearly a year of so much discussion and information offered from my side in the same topic.

4. No matter how overwhelming the work load of a one man's show might be, this 8 months late "response" is inequivocably poor and frustrating because you have known the issue and stated that you know the solution since very long time already (https://bugs.kde.org/show_bug.cgi?id=392251#c24) and yet no follow up action was taken to at least inform the users, other than "we need to think" in an obscure comment deep down in a long thread of some bug report more than half a year ago.

Please no blame on overwhelming work load and loneliness. That has been demonstrated countless times, not in the kindest manners, which I reckon why, having been myself victim of burnout. So please, don't.  I have been in worse situations than that, truly alone and without any resources at all, no Google, no Intel, no money. Only pressure. One myocardial infarct later, I should know. Believe me man, you don't want to be there. I pray that you won't.

If you are the only one reading and taking care of these reports and krita's problems, I sincerely wish that you rather focus more on your own health and sanity, and shutdown this nonsense bug report system.

But first!! please stop encouraging people to test krita and report bugs.   That also should help others to keep their sanity and stop wasting their valuable night hours and brain power in useless contributions like those I have made.

Sorry, your wall-of-text approach of admonishing us for being bad at figuring out how you messed up your system means we cannot do anything with your reports. 

I am closing all reports that you made that are still open. Please don't reopen them.

Please refrain from making further reports; you are apparently incapable of normal interaction.

_
  

Argumentum ad hominem


"A fallacious argumentative strategy whereby genuine discussion of the topic at hand is avoided by instead attacking the character, motive, or other attribute of the person making the argument, (...), rather than attacking the substance of the argument itself." (source: wikipedia).


Q.E.D.


None of what you say challenges the substance of the argument itself. Your intention is solely to offend and punish me because don't like the overwhelming argumentation I presented.

According to you, my "offense" seems to be verbosity.  But instead of addressing that as a criticism negative or constructively, you opt to demonstrate irrational hatred and distorted reality. In your own words, which unfortunately disprove and discredit yourself:

"how you messed up your system"

...which is a lie, without any substance and disproved largely, as you have several times demonstrated it with your own words. It only intends to hurt me, as a person: "That guy messed up his system. What a ..."  It is not substantiated in any reasoning that leads unequivocally to that.   Or: can you explain, why is that a logical conclusion?

  Ergo: Ad hominem.

"I am closing all reports that you made that are still open."

...just to "hurt me", not because they were fixed.  So "let's punish him by hurting our project and the users!".

Brilliant strategy.  I will wait until you realize you are not "hurting" me.  You are hurting your project, your users and your own image and reputation as leader of a relevant project for a big community.  Not me.   I am still waiting: don't you see it yet? I have time. What about now?

  Ergo: Ad hominem.


The problem does exist, not in "my system".  It is a bug affecting your product and your users, like it or not.  Users less technically savvy and less resourceful than myself.  Forget about me as a person.  I am not relevant to you.  Let alone the multiple experiments and evidence that can be read by everybody.  Just tell them to read your words in your comment here: https://bugs.kde.org/show_bug.cgi?id=392251#c24 . Among many others.

But you are trying to deflect from that.  By insult.  By diminishing me as an incompetent user that "messes up his system", although you have said the exact opposite many times.

Deflection does works well as a strategy, but only with non intelligent, ignorant people, as any politician will tell you.   But you have revealed your true position, although without justification or honest reasoning: "This is not my bug, it's somebody else's, I won't fix it."   Good for you.   Bad for krita.

Not being right, makes you explode, so:  Ad hominem.  "Let's insult this guy to scare others that might come like him in the future".

Bad news. You don't get to dictate what I do, what I say or where I say it.  Anything.  That is not how the real world works.   bugs.kde.org is not your private property, neither the Internet.

Equally grotesque would be me ordering you to refrain from insulting me personally and to stop believing you can give me orders or instructions, or order you to refrain from using fallacies to "argument" against concrete, objective technical issues.  I would be delusional if I would think that would work.

I will keep demonstrating that I am verbose with very good reasons, if and whenever I want, and you can keep demonstrating what you have demonstrated so concise but eloquently (thanks for that, really).

If you would step down of such arrogant position for a little while you might find that there might be others reading your "argumentation" and find it a bit dictatorial and too obviously conveniently deviating the conversation from technically demonstrated issues.   But is that how you solve issues in life? Insulting those who disagree with you or that you have offended and bullied previously?


I have no idea what makes you think you command masses, but whatever it is, .... Well, with ignorant, low intelligence audience, fallacy, manipulation and bullying do work like a charm, I give you that. Those eventually end up following and obeying.  With intelligent, highly educated people, doesn't work.

So, no, I won't "obey your commands" or be intimidated by lies, attempt of bullying, threats and personal insult.

Ad hominem.  Shame on you.

Is your next move to elevate the insulting to major cursing or lies to deflect the substance of the issue?  Really you need help, at various levels.   An apology wouldn't hurt you and could give better results, improving a little bit the shot you gave on your own foot and in krita project today.

A person like that, who doesn't accept any criticism and uses bullying, fallacy, and personal attack as means of terminating a technical disagreement, is seriously incapable of normal, intelligent interaction with highly educated, intelligent audience.

Q.E.D.




_