SUMMARY The user is led to believe a file transfer was successful when Dolphin crashes while the transfer is in progress. STEPS TO REPRODUCE 1. Open Dolphin 2. Start a file transfer: copy a big folder or file and paste into another place 3. Crash Dolphin[¹] while the transfer is in progress OBSERVED RESULT 1. A notification appears saying the transfer was successful (but it wasn't) EXPECTED RESULT 1. The file transfer continues after crash recovery OR the user is informed that the transfer failed SOFTWARE/OS VERSIONS Operating System: Kubuntu 18.10 PPA: kubuntu-ppa/backports KDE Plasma Version: 5.14.3 Qt Version: 5.11.1 KDE Frameworks Version: 5.51.0 Kernel Version: 4.18.0-11-generic OS Type: 64-bit Plasma Wayland session ADDITIONAL INFORMATION [¹] Not a reproducible crash, but I'll describe what I was doing if it helps. I was copying a lot of folders and only after the third crash I realized that things were not getting copied 100%. I made a checksum to confirm. 1. Open two Dolphin instances with split panels 2. Start a virtual machine with accelerated graphics 3. Start intensive operations that cap (like 100%) your disk I/O and CPU (e.g. compiling) 4. Start a Dolphin file transfer and that instance will crash
(In reply to wg9rffujwz8y276u from comment #0) > EXPECTED RESULT > 1. The file transfer continues after crash recovery OR the user is informed > that the transfer failed Dolphin shouldn't crash in the first place. If there is a crash, we need a backtrace to figure out what's going on. > 1. Open two Dolphin instances with split panels > 2. Start a virtual machine with accelerated graphics > 3. Start intensive operations that cap (like 100%) your disk I/O and CPU > (e.g. compiling) > 4. Start a Dolphin file transfer and that instance will crash So you can reproduce this crash? If yes, please post a backtrace (see https://community.kde.org/Dolphin/FAQ/Crashes)
Like I mentioned I can't reproduce the crash, and that's not the main issue I'm reporting. Even if nothing can be done about the transfer when Dolphin crashes (as much as we don't want that to ever happen), I think the user should be notified that the transfer was never completed. This could lead to data loss.
Created attachment 116398 [details] Crash 1
Created attachment 116399 [details] Crash 2
I managed to find a way to crash. It's basically what I've said before (just saturate I/O) but using a slow 5400rpm HDD makes it reproducible. Crash 1 was rsync in the background + Dolphin making a transfer. When you try to access the slow disk (HDD) it crashes. And that failed transfer was reported as a sucess after the crash. Crash 2 is just rsync from my ssd to a slow HDD. Once you try to browse that disk on Dolphin it crashes. I hope it helps.
The valgrind log seems to point to wayland. Are you using wayland? What about the backtrace from Dr Konqi? Do you get one?
Created attachment 116493 [details] New log (veracrypt HDD) > Plasma Wayland session > What about the backtrace from Dr Konqi? Do you get one? Nope. Dolphin starts fading and crashes.
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 mark the bug 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!
For admin purposes, this is fixed recently with 2beb1a0ad23177f7dc2e5ee622bed3a70f671278 in plasma-workspace. It was an issue in kuiserver.