Control: severity -1 minor Control: tag -1 pending On Tue, 08 Dec 2020 at 13:33:43 +0100, Jonas Smedegaard wrote: >> Added some thoughts at https://bugs.debian.org/976734#10 (leaving the >> decision whether to merge these to you). > > I prefer to track these issues separately: The confusing but harmless > chatter here, and the confusing messages for a fatal error at 976734.
Fair enough, changing -1's severity accordingly then. Quoting the above with the tentative fix: | As for "accept: Invalid argument", this comes from `/usr/libexec/lacme/webserver` , | a single instance of which is used for the entire command so it's not | specific to a particular order. I agree with you that it's not really | useful, but the web server is (intentionally) pretty dumb and will only | fail due to socket errors which don't really have a corresponding | user-facing command. This particular error comes from the accept(2) | syscall failing and setting errno to EINVAL. I *think* it's due to a | race condition: lacme(8) shuts down the webserver's listening socket | *before* terminating the child, causing any blocking accept(2) call to | fail, and the scheduler might allow the error to be spit out before | signal handling kicks in. Need to do more tests but I think simply | shutting down the socket last would fix this. In any case, it's indeed harmless since it happens during the cleanup phase just before lacme(8) exits, but it's indeed confusing and maybe even a little scary. > Would be helpful if you documented that detail. > > Maybe related to that mention that you make no promises for future IPC > stability (if that's the case) - still I think documenting _past_ > stability is helpful. Will add something to that effect. Thanks! -- Guilhem.
signature.asc
Description: PGP signature
