On Wed, Jun 18, 2014 at 5:43 PM, Colin Walters <walt...@verbum.org> wrote:
> On Wed, Jun 18, 2014, at 03:24 PM, Adam Miller wrote: > > Hello all, > > I noticed that a few of my RHEL6 machines were not reporting what > I expected from them in the output of needs-restarting from yum-utils > (version yum-utils-1.1.30-17.el6_5) so I started digging into the > needs-restarting code and I noticed that it's looking for 'fd:' in a line > of output in /proc/$pid/smaps which is a device field (if I understand > the kernel docs correctly). > > I was curious if there is any reason that's looking for a device field > specifically and not just a '/' character? I don't have a "fd:" in any > of /proc/*/smaps on a number of my machines > > > I do on RHEL7 at least, but your patch looks right to me regardless. > Actually isn't this fixed by > http://yum.baseurl.org/gitweb?p=yum-utils.git;a=commit;h=81f18617756866700c86ae34989edc7d0e71478d > ? > <http://yum.baseurl.org/gitweb?p=yum-utils.git;a=commit;h=81f18617756866700c86ae34989edc7d0e71478d> > > _______________________________________________ > Yum-devel mailing list > Yum-devel@lists.baseurl.org > http://lists.baseurl.org/mailman/listinfo/yum-devel > > It does appear to be fixed upstream and in Fedora. I suppose I'll file and issue ticket with Red Hat about the version that's in RHEL6. Also, I'm looking at what's upstream and what's in Fedora and it see the following: if slash == -1 or line.find('00:') != -1: # if we don't have a '/' or if we fine 00: in the file then it's not _REALLY_ a file I can't find any documentation to back up the claim of "00:" in the smaps denoting the 'truthiness' of a file. I was hoping I was just missing it and someone could point me in the right direction. Thanks again, -AdamM
_______________________________________________ Yum-devel mailing list Yum-devel@lists.baseurl.org http://lists.baseurl.org/mailman/listinfo/yum-devel
