On Thu, Sep 8, 2016 at 7:27 AM, maderios <mader...@gmail.com> wrote:
> On 09/08/2016 01:31 PM, Tom Horsley wrote:
>
>> I also often wonder what the heck it is doing during that
>> long pause. I see the CPU hit 100% while it is doing it,
>> but have no idea what might be going on in there.
>
> You'll find answer doing
> cat /var/log/dnf.log
Unlikely.
So I just did a single rpm update (btrfs-progs, less than 1MiB) and
dnf hung for about 2 minutes between
Cleanup : btrfs-progs-4.7.1-1.fc26.x86_64
and
Verifying : btrfs-progs-4.7.2-1.fc26.x86_64
I attached strace to the dnf pid and all I got was
wait4(20314,
It stayed there for the entire 1-2 minute hang and then I got a spew
of stuff, but also that line became
wait4(20314, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 20314
--- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=20314,
si_uid=0, si_status=0, si_utime=0, si_stime=2} ---
close(48) = 0
unlink("/var/tmp/rpm-tmp.vAGW3L") = 0
[...snip 1000+ lines...]
So it might not actually be dnf itself, maybe this calls for -ff -o to
get all child pids also straced and written to their own files. It may
be dnf is hung up waiting on something else, I have no idea.
But this is sorta new. There's always been a hang without a status of
what it's doing, but it's only been particularly long for bigger
updates. For a single rpm to have this long of a hang is unusual in my
experience.
--
Chris Murphy
--
users mailing list
users@lists.fedoraproject.org
To unsubscribe or change subscription options:
https://lists.fedoraproject.org/admin/lists/users@lists.fedoraproject.org
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org
