I was checking out my ntp to make sure it still worked and it does not so I did the following after checking the forums, bugzilla, and the mail list. Well 4.2.0-r2 is still broken. I downgraded to the previous version and it works. I then went back to 4.2.0-r2 and tried the -U ntp fix for /etc/conf.d/ntpd mentioned in the mail list but it did not work. ntp complained about invalid option. I don't have any restricts in /etc/ntp.conf as they are all commented out. In short, it's broken badly so I'm downgrading again and masking it. Others have had to downgrade also.
no real information here the 4.2.x ntpd's use -u for setting users while 4.1.x uses -U
Well, it may work for you but it sure doesn't work for many of us - check the mail list archives. I went through the forums, the mail list archives and the bugs for a long time last night reading all the suggested fixes, how-tos, etc. Tried all the things suggested their and finally ended up downgrading - as did many of the posters. It's worked flawlessly since. Yup, I'm aware that 4.2 use -u vs -U. One poster was able to get 4.2 to work by using -U in the /etc/conf.d/ntpd.conf file. That didn't work here. The system is a 2.6.7-r9 gentoo-dev-sources kernel. I also have a 2.4.24 xfs-sources system that ntp 4.2 works on. However, this system is being upgraded to 2.6.7-r11 gentoo-dev-sources so we'll see if ntp breaks on this system. If it works maybe that will give a clue - if not that may be a clue, too. I'd suggest leaving the bug open especially in light of the problems others have had.
there is no such file as /etc/conf.d/ntpd.conf so i really dont know what you're talking about there the custom options go in /etc/conf.d/ntpd i'm not going to re-open the bug until i see what you're talking about; people complaining that '-U' causes ntpd to not start is not a valid bug ... they need to update their config files
My apologies on the filename. I shouldn't do that from memory. It was the stock /etc/conf.d/ntpd that was installed by the 4.2 ebuild. I didn't touch it. My point is this - according to the mail list archive/forum a lot of people had problems with 4.2 and to try and make it work there were a lot of things tried. I tried all the ones that applied and ended up doing what many did - downgrade. I did downgrade - works, upgrade again just in case I'd messed something up - no work, downgrade again and it works. As for the -U I did NOT use that in the 4.2 version. I fully realize it's a pre 4.2 item. My config file had -u ntp:ntp in it when I installed 4.2 - that's what etc-update put in it and it didn't work with that. My comment about that was one person made his 4.2 work by going back to the -U option. Why I don't know. It will be interesting to see what my other system with 2.4.x and ntp 4.2 does when it's upgraded to 2.6.7. I spent a lot of time digging throught the archives, forums, etc. and many people had problems with 4.2 - enough to indicate to me that there is a problem or problems and although it doesn't affect all it effects enough people that it's not just someones system that's broken. As for what I'm talking about check the mail list archives (gentoo-user) and search on ntp.
i'm not exactly inclined to go searching for 'bug reports' since i really have no idea *what* i'm searching for bugzilla is for people to bring their info here and report a bug, not to redirect developers to some other place why dont you just post your `emerge info` and a summary of the trouble you're seeing
Whoops! I thought I'd posted all the info in this bug but I posted it on the mail list, did some more work and just though I put it in the bug. My apologies as I thought I had included the errors in the bugreport but I didn't and you can't read minds! Sorry about that and my sincere apologies. Let's start over with this - I'll give you the info I should have given you in the first place. When I use ntp 4.2 I get this error with ntptrace, ntpq Servname not supported for ai_socktype The /var/log/ntpd.log shows this when I start ntpd. 16 Jul 01:01:25 ntpd[7971]: ntpd exiting on signal 15 16 Jul 01:01:28 ntpd[8013]: parent died before we finished, exiting 16 Jul 10:15:25 ntpd[12260]: frequency initialized 37.072 PPM from /var/lib/ntp/ I also get two ntp processes as shown by ps ax | grep ntp as root. 18071 ? SL 0:00 /usr/bin/ntpd -p /var/run/ntpd.pid -u ntp:ntp 18076 ? S 0:00 /usr/bin/ntpd -p /var/run/ntpd.pid -u ntp:ntp All those problems go away when I go back to ntp 4.1.2. Here's the emerge info. root@strider etc # emerge --info Portage 2.0.50-r9 (default-x86-2004.0, gcc-3.3.3, glibc-2.3.3.20040420-r0, 2.6.7 -gentoo-r9) ================================================================= System uname: 2.6.7-gentoo-r9 i686 AMD Athlon(TM) MP 2000+ Gentoo Base System version 1.4.16 Autoconf: sys-devel/autoconf-2.59-r3 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-march=athlon-mp -Os -pipe" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.1 /share/config /usr/kde/3.2/share/config /usr/kde/3/share/config /usr/share/confi g /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/tex mf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi / /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-march=athlon-mp -Os -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache fixpackages nostrip sandbox" GENTOO_MIRRORS="ftp://ftp.gtlib.cc.gatech.edu/pub/gentoo http://csociety-ftp.ecn .purdue.edu/pub/gentoo/ ftp://csociety-ftp.ecn.purdue.edu/pub/gentoo/ ftp://mirr or.iawnet.sandia.gov/pub/gentoo/ ftp://ftp.ussg.iu.edu/pub/linux/gentoo" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.us.gentoo.org/gentoo-portage" USE="X aalib acl alsa apm avi berkdb caps cdr crypt cups doc dvd dvdr emacs emac s-w3 encode fbcon foomaticdb fusion gdbm gif gimp gimpprint gpm gtk gtk2 imap im lib jack jack-caps jack-tmpfs java joystick jpeg libg++ libwww mad mbox mikmod motif mpeg nas ncurses nls oggvorbis opengl oss pam passfile pdflib perl png post gres python quicktime readline rogue scanner sdl slang snmp spell ssl svga tcltk tcpd tetex truetype usb video_cards_nvidia videos x86 xatrix xfs xinerama xml2 xmms xv zlib"
i assume you're running all of these programs against localhost ? can you also post your /etc/ntp.conf file
Created attachment 35667 [details] ntp.conf file Yes, I run ntptrace or ntpq -p I'm attaching /etc/ntp.conf. I used the same file for 4.1 and 4.2. The only changes made were to add some servers. Gandalf is my 2.4.x machine. Currently I have both machines at ntp 4.1. If needed I can bring them back to ntp 4.2 to get data.
do you have ipv6 enabled in the kernel ? googling found references where this kind of thing isnt just in ntp ... http://mailman.ntp.org/pipermail/questions/2004-May/003461.html http://www.stacken.kth.se/lists/heimdal-discuss/2004-03/msg00073.html http://mail-index.netbsd.org/current-users/2003/10/02/0028.html these ones are good: http://www.fredi.de/maillist/archiv/2003/msg00720.html http://www.fredi.de/maillist/archiv/2003/msg00721.html
No, Ipv6 is not set: # CONFIG_IPV6 is not set I checked through the references you listed. What else in the kernel could cause this kind of problem? The ipv6 USE flag is NOT set.
try building your kernel with ipv6 and see if the error goes away could be that ntp handles ipv6 incorrectly when it's not present
Will do. Will that affect me elsewhere or will things still run ipv4 by default?
enabling ipv6 should not break any ipv4 apps i have ipv6 enabled on all my machines but none of them are actually hooked up to ipv6 networks ...
I did two things. 1. Went to kernel gentoo-dev-sources-2.6.7-r11. I needed to test this anyway before I switched to it from -r9. 2. Added ipv6 support to the kernel. Now ntpq -p gives me a list of hosts without error. remote refid st t when poll reach delay offset jitter ============================================================================== gandalf .INIT. 16 u - 64 0 0.000 0.000 4000.00 *ntp-4.cns.vt.ed 198.82.247.40 2 u 180 256 17 29.995 3.936 5.453 ntp3.tamu.edu 128.194.254.7 2 u 183 256 17 51.967 11.100 3.139 CYAN.SRV.CS.CMU 128.2.242.86 3 u 247 256 15 30.432 5.471 2.952 Looks like it's working now. I assume this means it needs ipv6 - is this a bug or a feature <G>.
4.2.0.20040617 has a bunch of ipv6 fixes
