Joey Hess wrote:
> Package: coreutils
> Version: 8.1-1
>
> I run tail -q --follow=name --retry -n 2 , tailing all log files
> in /var/log.
>
> Starting I think with version 8.0, this has began to die approximatly
> once a week.
Thanks for the report.
The "tail -F may abort" bug has just been fixed upstream,
along with another bug in the vicinity.
See the details on this thread:
http://thread.gmane.org/gmane.comp.gnu.coreutils.bugs/19382
Here are the NEWS entries:
tail -F (inotify-enabled) would abort when a tailed file is repeated
renamed-aside and then recreated.
[bug introduced in coreutils-7.5]
tail -F (inotify-enabled) could fail to follow renamed files.
E.g., given a "tail -F a b" process, running "mv a b" would
make tail stop tracking additions to "b".
[bug introduced in coreutils-7.5]
If you can evoke a failure also with "tail -f"
(as mentioned in the subject), please let us know,
since that would probably be due to a different problem.
