Description: It just happened to me that LibreOfficeDev version 7.4 crashed. When I opened the same file with an earlier version, this one told me that LibreOfficeWriter has just crashed and offered me the opportunity to send a crash report. It might be the case that the older version of LibreOffice associates its own version number to the crash report. I can’t check it. But if this really happens, crash reports might become misleading for developers! Steps to Reproduce: 1.Open a document with which LO Writer crashes 2.Open the same document again with another version of LibreOffice 3. Actual Results: The other version of LibreOffice tells me that it just has crashed. It also offers me to send a crash report. If it associates its own version number to the crash report, you'll get one under the wrong headline! Expected Results: In case of a crash report, as a user I would expect to see: • an indication in which part of the program the error occurred and what type of bug it is (e.g. insufficient memory, Null pointer, … or whatever else can be found in the crash dump, only the first few lines to just see what type of bug it is), • an indication what information will be sent in the crash report, • a progress indicator showing the percentage to be transferred and how many bytes this will be when completed. • a unique issue number related to the uploaded crash information, which is shown in one of these bug reporting windows, • I would expect that LibreOfficeWriter keeps track on these issue numbers and dates and I would expect LibreOffice to have a function, which – when invoked – asks te LibreOffice bug tracing system which of these issues are resolved and telling me, if I should install a new version of LO because of this progress, • I would expect a small dialogue where I would be asked a few questions: ◦ What did you do just before the crash happened: in this case: open a file, ◦ If it has to do with files: questions on their size (in number of pages, storage, …) ◦ If the bug happened with more than one file if the respective characteristics (yes, it happened with more than one large file, but it did not happen with one with 1200 pages instead of ca 670 and it also happened on a 167 pages document: this one showed up for a short moment, then it crashed. A ~-lock<filename.odt># locking file was left over after each such crash.. ◦ If it did not happen, if they are loaded into an earlier version – which one? ◦ If it was possible to deliver an example document of up to ten pages, which reproducibly shows the bug. For more details on my expectations see my attachment to another crash report: https://bugs.documentfoundation.org/attachment.cgi?id=178959 Reproducible: Always User Profile Reset: No Additional Info: This is the version which caused a crash: Version: 7.4.0.0.alpha0+ / LibreOffice Community Build ID: 59a7c40255b836ed75e64686fabb3e9938b755f0 CPU threads: 4; OS: Linux 5.4; UI render: default; VCL: gtk3 Locale: de-DE (de_DE.UTF-8); UI: en-US Calc: threaded I opened again in Version: 6.4.4.2 Build-ID: 3d775be2011f3886db32dfd395a6a6d1ca2630ff CPU-Threads: 4; BS: Linux 5.4; UI-Render: Standard; VCL: gtk3; Gebietsschema: de-DE (de_DE.UTF-8); UI-Sprache: de-DE Calc: threaded
(In reply to Adalbert Hanßen from comment #0) > If it associates its own version number to the crash report, you'll get one > under the wrong headline! > > Expected Results: > In case of a crash report, as a user I would expect to see: > • ... You have good point there ... but you put all these *different* issues into a single report, making is completely unmanageable. Every task should have an own issue, to be able to mark it "resolved" independently: e.g., "an indication in which part of the program the error occurred and..." is completely independent from "LibreOfficeWriter keeps track on these issue numbers and dates and ...". Please split it into separate issues. It doesn't matter if the tracker gets tens more issues; what matters is each one is manageable.
(In reply to Mike Kaganski from comment #1) > (In reply to Adalbert Hanßen from comment #0) > > If it associates its own version number to the crash report, you'll get one > > under the wrong headline! > > > > Expected Results: > > In case of a crash report, as a user I would expect to see: > > • ... > > You have good point there ... > > Please split it into separate issues. It doesn't matter if the tracker gets > tens more issues; what matters is each one is manageable. This is the split off part about bug reporting. The crash report which triggered it is merely the motivation to show possible causes of trouble by uploaded bug reports and it also suggests an enhancement how to improve this process. I had however forgotten to set the flag "enhancement" which I have set now.
Hello Adalbert, a new major major release of LibreOffice is available since this bug was reported. Could you please try to reproduce it with the latest version of LibreOffice from https://www.libreoffice.org/download/libreoffice-fresh/ ? I have set the bug's status to 'NEEDINFO'. Please change it back to 'UNCONFIRMED' if the bug is still present in the latest version. Please test only with older versions, that are still under development
Created attachment 189305 [details] the file which crashed LO Writer when double-clicked upon in the file browser (In reply to Dieter from comment #3) > Hello Adalbert, a new major major release of LibreOffice is available since > this bug was reported. Could you please try to reproduce it with the latest > version of LibreOffice from > https://www.libreoffice.org/download/libreoffice-fresh/ ? > I have set the bug's status to 'NEEDINFO'. Please change it back to > 'UNCONFIRMED' if the bug is still present in the latest version. > > Please test only with older versions, that are still under development I tried it with Version: 7.6.0.3 (X86_64) / LibreOffice Community Build ID: 69edd8b8ebc41d00b4de3915dc82f8f0fc3b6265 CPU threads: 4; OS: Linux 5.4; UI render: default; VCL: gtk3 Locale: de-DE (de_DE.UTF-8); UI: de-DE Calc: threaded and I tried to load a file which caused the described error by double-clicking on a file which I had put aside when I reported the bug. The outcome was another crash. During the bug reporting sequence there was a link to https://crashreport.libreoffice.org/stats/crash_details/02b0d723-3657-438b-856b-758278c6cbbc. I followed the link and I observed that it contains information which probably will enable someone to pin this bug. Strangely (after LO Writer was running) I was able to explicitly load the file into it with the open file dialog by explicitly telling the file to open. Strangely I was unable to open it by the open file>Recent dialog: nothing happened. Even more strangely: After trying it several times, now I can open this file by all the described ways. Very strange!
[Automated Action] NeedInfo-To-Unconfirmed
(In reply to Adalbert Hanßen from comment #4) > Created attachment 189305 [details] > the file which crashed LO Writer when double-clicked upon in the file browser I could open it without any problems Version: 7.6.1.2 (X86_64) / LibreOffice Community Build ID: f5defcebd022c5bc36bbb79be232cb6926d8f674 CPU threads: 4; OS: Windows 10.0 Build 19045; UI render: Skia/Raster; VCL: win Locale: de-DE (de_DE); UI: en-GB Calc: CL threaded Perhaps somebody else can help.
(In reply to Dieter from comment #6) > (In reply to Adalbert Hanßen from comment #4) > > Created attachment 189305 [details] > > the file which crashed LO Writer when double-clicked upon in the file browser > > I could open it without any problems > > Version: 7.6.1.2 (X86_64) / LibreOffice Community > Build ID: f5defcebd022c5bc36bbb79be232cb6926d8f674 > CPU threads: 4; OS: Windows 10.0 Build 19045; UI render: Skia/Raster; VCL: > win > Locale: de-DE (de_DE); UI: en-GB > Calc: CL threaded > > Perhaps somebody else can help. Dieter, I just tried to load the file into Version: 24.2.0.0.alpha0+ (X86_64) / LibreOffice Community Build ID: b3fdd999f87312447d03915585812b3a5cd48141 CPU threads: 4; OS: Linux 5.4; UI render: default; VCL: gtk3 Locale: de-DE (de_DE.UTF-8); UI: en-US Calc: threaded and it did not crash. I tried it by going through the normal load file process and I did another trial via Fils>Recent Documents. I did yet another test by double-clicking on the set aside file which I had uploaded when reporting the bug. All three trials did not crash with this version. It looks like the bug is fixed in the development version under my operating conditions.
(In reply to Adalbert Hanßen from comment #7) > It looks like the bug is fixed in the development version under my operating > conditions. Adalbert, thank you for retesting. => RESOLVED WORKSFORME