RE: 2.6.28.4 regression: mmap fails if mlockall used

[Hugh Dickins]

On Tue, 10 Feb 2009, Doug Bazarnic wrote:
> I can confirm the mlock.c patch works on 2.6.28.4.  2.6.28.3 works fine.
> This issue happens on both Centos 5.2 x86_64 and RHEL 5.3 x86_64.
> 
> Without the mlock.c patch, ntpd fails to start on 2.6.28.4:
> Feb 10 22:03:19 testbox ntpd[4030]: kernel time sync status 0040
> Feb 10 22:03:19 testbox ntpd[4030]: getaddrinfo: "127.0.0.1" invalid host
> address, ignored
> Feb 10 22:03:19 testbox ntpd[4030]: getaddrinfo: "::1" invalid host address,
> ignored
> Feb 10 22:03:19 testbox ntpd[4030]: getaddrinfo: "127.127.1.0" invalid host
> address, ignored
> Feb 10 22:03:19 testbox ntpd[4030]: Cannot find user `ntp'
> Feb 10 22:03:21 testbox ntpd_initres[4034]: parent died before we finished,
> exiting
> 
> Fyi.. gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)
> 
> Thanks to Sami for mentioning ntpd failures, as I was going nuts trying to
> figure out why my ntpd.conf file wasn't working anymore.  
> 
> Thanks for the patch as well.

Thanks a lot for testing and confirming:
you should find the fix is in 2.6.28.5, due out maybe tomorrow.

Rafael, you did already close Bug #12669 - now please cancel my
suggestion that it remain open until the fix has been confirmed ;)

Hugh
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/