[Ovirt-devel] [PATCH] * ovirt-logrotate (EXITVALUE): Propagate failure to exit, too.

[Perry N. Myers]

Jim Meyering wrote:
> "Perry N. Myers" <pmyers at redhat.com> wrote:
>> Jim Meyering wrote:
>>> Here's an even tinier change:
>>> Without this, a failing ovirt-logrotate would exit successfully.
>>>
>>> >From 2691a4203489e250fa1e8d64e3c52072f25929a3 Mon Sep 17 00:00:00 2001
>>> From: Jim Meyering <meyering at redhat.com>
>>> Date: Mon, 8 Sep 2008 11:43:06 +0200
>>> Subject: [PATCH] * ovirt-logrotate (EXITVALUE): Propagate failure to exit, too.
>>>
>>> ---
>>>  logrotate/ovirt-logrotate |    3 +--
>>>  1 files changed, 1 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/logrotate/ovirt-logrotate b/logrotate/ovirt-logrotate
>>> index f0a2dde..a6a394d 100644
>>> --- a/logrotate/ovirt-logrotate
>>> +++ b/logrotate/ovirt-logrotate
>>> @@ -6,5 +6,4 @@ EXITVALUE=$?
>>>  if [ $EXITVALUE != 0 ]; then
>>>      /usr/bin/logger -t logrotate "ALERT exited abnormally with [$EXITVALUE]"
>>>  fi
>>> -exit 0
>>> -
>>> +exit $EXITVALUE
>> What happens if a logrotate script returns a non-zero value?  Is there
>> something that we need to catch and handle here?
> 
> I noticed the problem via inspection and fixed it on principle ;-)
> In case someone runs ovirt-logrotate ... || notify-me-of-failure
> 
> However, now I see that it is invoked solely via cron.  In that case,
> the failure will be recorded via logger, but the exit status will be ignored.

That's what I thought...  Still, there's not reason to artificially return 
zero in all cases.  So I'd say go ahead and push this patch.  ACK

Perry

-- 
|=-        Red Hat, Engineering, Emerging Technologies, Boston        -=|
|=-                     Email: pmyers at redhat.com                      -=|
|=-         Office: +1 412 474 3552   Mobile: +1 703 362 9622         -=|
|=- GnuPG: E65E4F3D 88F9 F1C9 C2F3 1303 01FE 817C C5D2 8B91 E65E 4F3D -=|