Ihor Radchenko <[email protected]> writes: > "Peter J. Jones" <[email protected]> writes: > >>> Are you using some kind of non-standard auto-revert package or settings >>> for auto-revert-mode? >> >> No, just standard auto-revert-mode. >> >> As I mentioned previously, I run multiple Emacs daemons. This happens >> when I change an org file in one daemon instance and auto-revert-mode >> detects the change and reverts the buffer in another daemon instance. > > Strange. > > I actually tried to run secondary Emacs and modify an Org file from > there, but I was unable to reproduce - the Org file gets updated by > auto-revert-mode, but no cache issues occur. > > Are you able to reproduce the problem reliably? May you create a > reproducer starting from emacs -Q? See > https://orgmode.org/manual/Feedback.html#Feedback
I am going through the mailing list and closing old reports. This one is about multiple emacs' instances creating an error. So far a reproducer with "emacs -Q" has not been provided. Peter J. Jones experienced this bug on org-mode version 9.6.4 and emacs 28.2 back on <2023-06-15 Thu>. Since nix was used it might be possible to recreate this environment even today. It would be very useful to know if this bug can be reproduced on the latest version of emacs and org-mode. Emacs has a variable `create-lockfiles' that affects this behaviour. Have you modified that? Some information about file locking is provided in these info manuals: (emacs) Interlocking (elisp) File Locks It's also possible this has something to do with org-mode's use of `inhibit-modification-hooks' or `with-silent-modifications'. I also wonder if this has anything to do with the org-cache being written to disk (see variable `org-element-cache-persistent'). If I receive confirmation that this bug is still present then I will reopen this report and investigate it further. Canceled.
