Re: jail on ZFS - Unable to mount devfs
hi patrick, If I remember correctly there was no password file for in the jail. I think you have to rerun a certain command. Of course I do not remember the command :( The command should create the master password database. using the ServiceJail model, after populating the jail skeleton and running mergemaster, the two commands i run are, /usr/bin/cap_mkdb /j/jSKEL/etc/login.conf /usr/sbin/pwd_mkdb -d /j/jSKEL/etc -p /j/j/etc/master.passwd which should take care of that. Also you have to run within in the jail newaliases to create the aliases file, do a touch /etc/fstab to stop complaints about unable to read mountpoints. hm. i did not do that this time around. i'd built my jail-world with *both* NO_MAILWRAPPER=true NO_SENDMAIL=true, so i mayhave caused myself a problem. rather than cp'ing bins, tobe safe, i'll just rebuild world ... and see in a bit if that helps. thanks. Furthermore I am not sure that you can run a jail on a zfs file system. The setup I have is that I run my jails on ufs and have a zfs filesystem available within the jail. ?? if that's true, then that renders the rest moot -- and i have a problem. atm, i have cat /etc/fstab /dev/mirror/gm0s1a /bootdirufs rw1 1 /dev/mirror/gm0s1b noneswapsw0 0 /dev/acd0 /cdrom cd9660 ro,noauto 0 0 /j/jMROOT /j/jTESTnullfs ro0 0 /j/s/jTEST /j/jTEST/s nullfs rw0 0 zfs list NAME USED AVAIL REFER MOUNTPOINT z5.23G 213G 250M /z z/data 20K 213G20K /data z/home 28.5K 213G 28.5K /home z/j23K 213G23K /j z/tmp 406K 213G 406K /tmp z/usr4.88G 213G 4.88G /usr z/var 105M 213G 105M /var where z/j is a zfs mount. i *can* access the jail, and do just about 'all' i need to in the jail (builds, exec, etc). but do *not* yet know if, by running the jail on zfs space whehter i've compromised anything. do you have a reference for your comment? or, perhaps, someone else can comment, as well? thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: jail on ZFS - Unable to mount devfs
hi wesley, I have a jail running in a ZFS environment. [EMAIL PROTECTED] ~ % jls JID IP Address Hostname Path 3 192.168.1.100 asterisk /u/jails/asterisk [EMAIL PROTECTED] ~ % mount | grep data data on /u (zfs, NFS exported, local, noatime) [EMAIL PROTECTED] ~ % mount | grep devfs devfs on /dev (devfs, local) devfs on /u/jails/asterisk/dev (devfs, local) [EMAIL PROTECTED] ~ % here's what i have jls JID IP Address Hostname Path 1 10.0.0.200 jTEST.internal.net /j/jTEST mount z on / (zfs, local) devfs on /dev (devfs, local) /dev/mirror/gm0s1a on /bootdir (ufs, local, soft-updates) z/data on /data (zfs, local) z/home on /home (zfs, local) z/j on /j (zfs, local) z/tmp on /tmp (zfs, local) z/usr on /usr (zfs, local) z/var on /var (zfs, local) /j/jMROOT on /j/jTEST (nullfs, local, read-only) /j/s/jTEST on /j/jTEST/s (nullfs, local) devfs on /j/jTEST/dev (devfs, local) which, i think?, says the same. argh. confused. following too many threads with partial solutions ... can you share your setup-a-jail-on-a-zfs'd-host steps? thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
amd64 native boot loader?
hi, i've FBSD/amd64 62Rp9 installed. kernel world are my own builds from latest cvsup. on boot I see: FreeBSD/i386 bootstrap loader odd. i'd expect a native loader ... checking in, /usr/src/sys/boot ls Makefile alpha/arm/ efi/ forth/ia64/ pc98/ sparc64/ READMEarc/ common/ ficl/ i386/ ofw/ powerpc/ other arches seem to be there ... just not amd64. where's the src for the amd64? thanks. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ports/115885: misc/help2man: help2man ignores installed gettext
[EMAIL PROTECTED] wrote:
Synopsis: misc/help2man incorrectly insists on p5-* dependency when already
installed as CPAN module
State-Changed-From-To: open-closed
State-Changed-By: arved
State-Changed-When: Mon Sep 3 08:36:07 UTC 2007
State-Changed-Why:
It is unecessary to reopen another bug report, since you neither provide a
patch nor convinced us that this should be a supported configuration.
again, WHAT supported configuration are you talking about here?
there's a properly installed perl-module. expecting freebsd ports of
perl programs to correctly use perl-compliant module installs is,
somehow, unreasonable?
here's what i know/see.
a freebsd port (help2man)
(a) looks for Locale::gettext
(b) finds Locale::gettext
(c) uses Locale::gettext in the build
THEN,
*fails* to recognize the in-place dependency IT ALREADY LOOKED FOR
USED in updates of ports ...
BTW. Don't miss the helpful answer by Mel on freebsd-question.
http://www.freebsd.org/cgi/query-pr.cgi?pr=116026
Mel wrote:
Except for apps in the ports, because cpan/cpanp can not register *port*
dependencies.
RUN_DEPENDS does not mean check if this file exists, it means register a
dependency on this port. What it does is:
1) check if the port is installed and if not install it.
2) add a line @pkgdep portname-version to $PKGDB/portname-version/+CONTENTS
3) add a line @DEPORIGIN:category/origin to $PKGDB/portname-version/+CONTENTS
4) add a line portname-version to $PKGDB/depname-version/+REQUIRED_BY
If you wanna get rid of this problem, you could use an AFTER_INSTALL hack in
pkgtools.conf to undo 2)-4). Messy, but works.
Much cleaner is to just use the ports system p5-*. It does it's job very well
that's not been my experience -- but, that's not the argument here.
i think it's not unreasonable to expect that a perl-module is a
perl-module is a perl-module.
if it's properly installed -- and, in this case, it is, just like ALL
my other perl-modules, then a port should treate it
consistently/correctly.
just like all my other ports do -- just NOT help2man.
as for working around the problem ... the SIMPLEST solution is not to
bother with ports system in this case, and do a manual install of
help2man.
but my goal was to be helpful report a problem.
and since you're familiar with cpan/cpanp I'm sure you'll have no problem
installing modules that might not exist in the freebsd ports into a directory
*outside* the ports hierarchy (/usr/local) and add this to the module search
path.
when help2man is installed via ports it *USES* the cpan-installed perl-module.
cpanp -u gettext Locale::gettext
module_info Locale::gettext
Can't create Module::Info object for module 'Locale::gettext' at
/usr/local/bin/module_info line 96.
cd /usr/local/ports
rm -rf gettext-*
wget http://search.cpan.org/CPAN/authors/id/P/PV/PVANDRY/gettext-1.05.tar.gz
tar zxvf gettext-1.05.tar.gz
# #21625: Compilation under freebsd
# http://rt.cpan.org/Public/Bug/Display.html?id=21625
http://rt.cpan.org/Ticket/Attachment/243467/103618/gettext-freebsd.patch
http://rt.cpan.org/Ticket/Attachment/243467/103621/gettext-MANIFEST
cd /usr/local/ports/gettext-1.05
patch -p0 ../P_gettext-freebsd.patch
cp ../P_gettext-MANIFEST MANIFEST
perl Makefile.PL
make install
module_info Locale::gettext
Name:Locale::gettext
Version: 1.05
Directory: /usr/local/lib/perl5/site_perl/5.8.8/mach
File:/usr/local/lib/perl5/site_perl/5.8.8/mach/Locale/gettext.pm
Core module: no
now, building help2man
% portinstall help2man
[Updating the pkgdb format:bdb_btree in /var/db/pkg ... - 176
packages found (-1 +0) (...) done]
[Gathering depends for misc/help2man .. done]
--- Installing 'help2man-1.36.4_1' from a port (misc/help2man)
--- Building '/usr/ports/misc/help2man'
...
=== help2man-1.36.4_1 depends on file:
/usr/local/lib/perl5/site_perl/5.8.8/mach/Locale/gettext.pm - found
See? HERE it knows the dependency, AND finds it
NOT SURPRISING, as,
grep -i gettext /usr/ports/misc/help2man/Makefile
BUILD_DEPENDS=
${SITE_PERL}/${PERL_ARCH}/Locale/gettext.pm:${PORTSDIR}/devel/p5-Locale-gettext
RUN_DEPENDS=
${SITE_PERL}/${PERL_ARCH}/Locale/gettext.pm:${PORTSDIR}/devel/p5-Locale-gettext
USE_GETTEXT=yes
demonstrates that it, in fact, DOES look for a normal perl-module
**BEFORE** looking for the p5-Locale-gettext port-install!
...
=== Compressing manual pages for help2man-1.36.4_1
=== Registering installation for help2man-1.36.4_1
=== Cleaning for help2man-1.36.4_1
[Updating the pkgdb format:bdb_btree in /var/db/pkg ... - 177
packages found (-0 +1) . done]
and the build continues to a cussessful conclusion.
pkg_info | grep help2man
help2man-1.36.4_1 Automatically generating simple manual pages from
program o
pkg_info -rR help2man-1.36.4_1
Re: ports/115885: misc/help2man: help2man ignores installed gettext
note(s) to self: help2man port is borked. every other port with perl-module dependencies is fine ... freebsd folks not interested in fix. do a manual install instead. problem solved. outa here. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: ports/115885: misc/help2man: help2man ignores installed gettext perl mod; insists on stale p5-* dependency
hi,
You have perl modules installed, that have no corresponding installed
FreeBSD port.
i do not use BSDPAN -- it's sloppy about its dependency mgmt.
as i do on every other os/platform, i use ONLY native cpan/cpanp.
i have dozens of cpan-installed perl-modules. cpan/cpanp manage the
dependencies just fine.
the problem is in the case of 'help2man'.
the port-install of help2man *DOES* use the cpan-installed gettext
perl-module correctly,
pkg_info | grep help2man
help2man-1.36.4_1 Automatically generating simple manual pages from
pkg_delete help2man-1.36.4_1
portinstall help2man
--- Installing 'help2man-1.36.4_1' from a port (misc/help2man)
--- Building '/usr/ports/misc/help2man'
...
=== help2man-1.36.4_1 depends on file: /usr/local/bin/perl5.8.8 - found
=== Applying FreeBSD patches for help2man-1.36.4_1
=== help2man-1.36.4_1 depends on file:
/usr/local/lib/perl5/site_perl/5.8.8/mach/Locale/gettext.pm - found
...
*BUT*, in subsequent port updates/upgrades, it looks for the
p5-Locale-gettext* module.
looking in source,
cd /usr/ports/misc/help2man
grep p5-Locale-gettext *
Makefile:BUILD_DEPENDS=
${SITE_PERL}/${PERL_ARCH}/Locale/gettext.pm:${PORTSDIR}/devel/p5-Locale-gettext
Makefile:RUN_DEPENDS=
${SITE_PERL}/${PERL_ARCH}/Locale/gettext.pm:${PORTSDIR}/devel/p5-Locale-gettext
and verifying in my system,
module_info Locale::gettext
Name:Locale::gettext
Version: 1.05
Directory: /usr/local/lib/perl5/site_perl/5.8.8/mach
File:/usr/local/lib/perl5/site_perl/5.8.8/mach/Locale/gettext.pm
Core module: no
there's no legitimate reason why it should NOT be looking for the
*correctly installed* gettext dependency in site_perl path ...
but, the fact remains that it isn't.
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: request add db45/db46 support to mail/exim port
I have now fixed it so that the exim port doesn't need to set WITH_BDB_VER?=1 anymore. Instead setting either WITH_BDB or WITH_BDB_VER 1 will make the port use Mk/bsd.database.mk to choose the version of BDB to use, otherwise it will default to the system BDB. The updated patch is in PR 115427: http://www.freebsd.org/cgi/query-pr.cgi?pr=115427 great! Thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
how 2 addressport map outbound traffic to multiple/different IPs on a single intfc?
hi, i'm trying to do 1:1 (nat?) *outbound* address mapping using pf, but NOT 'whole server' binat, but rather a single addressport. i'm close, but no cigar ... any suggestions? details follow ... thanks! i've a block of 8 static IPs, x.x.x.1 - x.x.x.8. freebsd6.2-Rp5+pf are installed as my edge router/firewall. the router has a SINGLE external interface, sis0, mapped at bootup to a single pppoe-generated interface, tun0. ifconfig shows that tun0 is assigned the 'primary' IP of x.x.x.1. so most of my LAN-WAN traffic travels out, appearing to originate at x.x.x.1. i've two mail servers on my LAN, at private addresses z.z.z.100 z.z.z.200. i've NAT+rdr set up to address-map WAN to LAN addresses for the two servers. nat on $ext_if from $int_if:network to any - ($ext_if) rdr on $ext_if proto tcp from any to x.x.x.2 port 25 - z.z.z.102 port 25 rdr on $ext_if proto tcp from any to x.x.x.3 port 25 - z.z.z.103 port 25 where $ext_if == tun0. now as to OUTBOUND mapping ... i want to make sure that traffic: FROM internal server @ z.z.z.102:25 exits $ext_if, 'seen' as SRC_ADDR=x.x.x.2 FROM internal server @ z.z.z.103:25 exits $ext_if, 'seen' as SRC_ADDR=x.x.x.3 reading: PF: Network Address Translation (NAT) http://cvs.openbsd.org/faq/pf/nat.html and, PF: Address Pools and Load Balancing http://cvs.openbsd.org/faq/pf/pools.html (1) 'binat' is not what i want, as i want to ONLY map a single addr for a single port -- NOT the whole server in a 1:1 mapping for all ports and, (2) from the POOLS discussion, 'simple' outbound NAT: nat on $ext_if from z.z.z.102 port 25 to any - ($ext_if) nat on $ext_if from z.z.z.102 port 25 to any - ($ext_if) doesn't do it either - i dno't think -- as $ext_if picks up the primary IP assigned via the pppoe startup, x.x.x.1. so, i think i'm in the right ballpark with *nat of some sort, but how do i get this done correctly? cheers! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
hi, though the prob's been fixed, just to ack/comment ... the issue 4 me was that pf itself was not starting, not that it had started but the rules were not loaded, or some such ... Reloading the rules is supposed to allow pf to pick-up new interfaces, which is why it's done after ppp is started. I was wondering if Volker had a valid reason for thinking there is a problem, or whether he was speculating from incomplete knowledge. As regards pf not starting, in another sub-thread you seem to be saying that the underlying problem was an irregularity in ppp.conf. ppp.conf is not read until *after* pf starts-up, so can't explain pf's not starting. argh. well, i'm awash in subtleties. atm, i'm choosing to not 'look a gift horse in the mouth', and be happy that it's up-n-running/working again. as for /understanding/ why, that'll require reading beer. which is why weekends were invented ;-) thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
hi, Funny thing is, I doubt I'd have noticed it without your blank line! heh. well, glad i could help! i live to serve ;-) cheers! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
hi, Hello, it's your niggly proofreader :-D (and fellow Stephenson fan) !! If you really have that blank line before 'add default HISADDR' above, then it marks the end of your default section. The 'add default' and the two lines following will not be executed. I expect you'll want the 'add default' line as the last in your ppp1: section anyway; the other two could go in either, but I'd opt for the default block myself. i had not realized that blank lines were 'read for real'. it's now been removed ... and i've moved the 'add default' to the connection ... I'm again unsure whether it's related to your pf problem, rusted-on ipfw here, but my connections tend to work better with a default route .. recently converted to pf, and been pretty pleased/impressed with it so far. a few gotchas, mainly due to not (yet) having read the /right/ man page, bu i'm makin progress! now, to clean up a bit more, and see if all's better ... thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
well, per Ian's catch/suggestion, removing the 'blank line' from my ppp.conf, and moving, add default HISADDR to the ppp1: connection stanza seems to have done the trick! pf loads properly on reboot. swithc it back, and it does not. so, guessing, it's the lack of a default root as a result of the blank line. the gotcha here was that, according to my notes, i *HAD* checked/ensure that my default routes were correctly initialized (with netstat -nr), but, apparently, BEFORE i'd naively/mistakenly added that blank line. woohoo! and, thanks all for the add'l comments -- good pointers on anchors and pf operation in general. archiving this thread! :-) cheers. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
hi, Have you any particular reason to think that this is really a problem? Given that /etc/rc.d/ppp automatically reloads the pf rules after the tun device is created. though the prob's been fixed, just to ack/comment ... the issue 4 me was that pf itself was not starting, not that it had started but the rules were not loaded, or some such ... thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
fbsd 6.2 pf starts -- but not on boot
hi,
i've fbsd 6.2R/p5, with pf compiled into a custom kernel.
on boot, pf is, apparently, not starting.
but, if i exec
/etc/rc.d/pf start
immediately after boot to prompt is done, then all's OK.
the only related (?) messages -- error or otherwise -- i've found are
on startup.
any ideas/suggestions as to what might be the prob? and/or how to troubleshoot?
thanks!
for reference, from console output @ startup,
...
sis0: link state changed to UP
sis1: link state changed to UP
lo0: flags=8049UP,LOOPBACK,RUNNING,MULTICAST mtu 16384
inet6 fe80::1%lo0 prefixlen 64 sscopeid 0x5
inet6 ::1 prefisxlen 128
inet2 127.0.0.1 netma:sk 0xff00
sis0: flags=8843lUP,BROADCAST,RUiNNING,SIMPLEX,MUnLTICAST mtu 149k2
options=48V LAN_MTU,POLLINGs
inet 10.0.0.10 netmask 0xfaf00 broadcastt 10.0.0.255
ether 00:00:12:d4:15:88
media:t Ethernet autoseolect (100baseTX full-duplex)
status: active
sis1: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1492
options=48VLAN_MTU,POLLING
ether 00:00:12:d4:15:89
media: Ethernet autoselect (100baseTX full-duplex)
status: active
Starting pflog.
pflog0: promiscuous mode enabled
Enabling pf.
Jun 4 13:38:11 pflogd[479]: [priv]: msg PRIV_OPEN_LOG received
pfctl: DIOCSETSTATUSIF
pf enabled
Starting ppp.
add net default: gateway 10.0.0.10
Additional routing options:.
Starting devd.
Mounting NFS file systems:.
...
and, further,
% cat /etc/rc.conf
ifconfig_sis1=mtu 1492 polling
ifconfig_sis0=inet 10.0.0.10 netmask 255.255.255.0 mtu 1492 polling
hostname=router.mydomain.com
defaultrouter=10.0.0.10
# PPP
ppp_enable=YES
ppp_mode=ddial
ppp_nat=NO
ppp_profile=ppp`
# PF
pf_enable=YES
pf_flags=
pf_rules=/usr/local/etc/pf/pf.conf
pflog_enable=YES
pflog_flags=
pflog_logfile=/var/log/pflog
# SPAMD
obspamd_enable=YES
obspamd_flags=-v -l 127.0.0.1 -h mail.mydomain.com -n GATEWAY
obspamlogd_enable=YES
obspamlogd_flags=
# MISC
inetd_enable=YES
firewall_enable=NO
nfs_client_enable=YES
usbd_enable=YES
snmpd_enable=NO
webmin_enable=NO
pcscd_enable=NO
sshd_enable=NO
ntpdate_enable=NO
ntpd_enable=YES
# BIND9
named_enable=YES
named_chrootdir=/var/chroot/named
named_flags=-c /etc/named.conf
named_pidfile=/var/run/named.pid
named_program=/usr/local/sbin/named
named_uid=bind
# RBLDNSd
rbldnsd_enable=YES
rbldnsd_flags=${rbldnsd_flags:--4 -u rbldns:rbldns -r
/var/chroot/rbldnsd -b 127.0.0.1/530 -t 900
my.dnsbl:ip4set:dnsbl/mx_local_black.txt}
# SENDMAIL
mta_start_script=/etc/rc.sendmail
sendmail_enable=YES
sendmail_flags=-L sm-mta -bd -q30m
sendmail_pidfile=/var/run/sendmail.pid
sendmail_procname=/usr/sbin/mailwrapper
sendmail_msp_queue_enable=YES
sendmail_msp_queue_flags=-L sm-msp-queue -Ac -q30m
sendmail_outbound_enable=YES
sendmail_outbound_flags=-L sm-queue -q30m
sendmail_submit_enable=YES
sendmail_submit_flags=-L sm-mta -bd -q30m -ODaemonPortOptions=Addr=localhost
# DHCP Services
dhcpd_enable=YES
dhcpd_chroot_enable=YES
dhcpd_chuser_enable=YES
dhcpd_devfs_enable=YES
dhcpd_conf=/usr/local/etc/dhcpd.conf
dhcpd_flags=-q -early_chroot
dhcpd_ifaces=sis0
dhcpd_jail_enable=YES
dhcpd_rootdir=/var/chroot/dhcpd
dhcpd_withgroup=dhcpd
dhcpd_withumask=022
dhcpd_withuser=dhcpd
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
On 6/4/07, Volker [EMAIL PROTECTED] wrote: without seeing your pf.conf ruleset, happy to send/post if required/helpful ... I guess you're using a ppp connection to your upstream provider and firewalling on the tunX interface (using tun0 as $ext_if). you're absolutely correct here. As FreeBSD boots up, this interface does not yet exist when pf is loaded. clear. As soon as ppp is loaded and interface tun0 has been created, pf will happily load your ruleset. aha. does that suggest that i'm simply not waiting long enough? your following comments suggest otherwise, iiuc, that i need to proactively _do_ something different ... The solution is to either have pf rules loaded late (later than ppp is started) clearly, simply including pf-related items in rc.conf after pppoe-related items is not sufficient. i'll take a look at rcorder ... which i wasn't aware of at all. thanks! or use anchors and load ext rules into the anchor when the ppp interface is up. i hadn't thought of using anchors in this fashion. i'm off to google, but any good examples you can reference? The easier is to have the rules loading late (check using rcorder) but this may also fail if something goes wrong with ppp. i /thought/ i'd dealt with the intfc/ppo/pf ordering issue, configuring, cat /etc/ppp/ppp.linkup ppp1: ! sh -c /sbin/pfctl -ef /usr/local/etc/pf/pf.conf !bg sh -c echo `/bin/date` `/etc/bin/ip` ppp.linkup /etc/ppp/log cat /etc/ppp/ppp.linkdown ppp1: !bg route delete HISADDR ppp1 !bg pfctl -F all -d cat /etc/ppp/ppp.conf default: set device PPPoE:sis1: set speed sync set ctsrts off set dial set login set cd 10 set timeout 0 set redial 0 0 enable lqr set lqrperiod 20 set log Phase tun command add default HISADDR enable tcpmssfixup disable dns ppp1: set authname [EMAIL PROTECTED] set authkey set MRU 1492 set MTU 1492 are these NOT supposed to address/solve the problem? or are the configs wrong? Mikhail Goriachev [EMAIL PROTECTED] Just a shot in the dark. You are probably putting hostnames in your pf.conf instead of IPs. PF starts before Bind. So it can't resolve hostnames in the rules and hence doesn't start. heh. a good call, but, i'd already made THAT mistake a month or so ago. ;-) thanks though! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: fbsd 6.2 pf starts -- but not on boot
hi, I really don't know whether this might be related to your problem, but my proofreading eye was distracted by this in your rc.conf: # PPP ppp_enable=YES ppp_mode=ddial ppp_nat=NO ppp_profile=ppp` What rc would make of that backtick inside quotes, I know not .. wow! good eye. checked, and that's some cp-n-paste weirdness, or my fat thumbs. the line actually is, ppp_profile=ppp1 thanks for the check! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PPPoE equivalent?
hi, I'm not sure there is, I've had a similar problem like this before, and I got around it by writing a simple script that would try and ping a local site 4 times, and if no responses got back it would killall ppp and delete the default routes and tell ppp to reconnect. It worked quite well when put into cron to run at 1 minute intervals. i'm getting all this set up as well, and following this thread ... (1) are not ppp.linkup ppp.linkdown supposed to be the 'place' to deal with this? (2) could you share your simple script? thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: PPPoE equivalent?
Yes, it puzzles me why it happens in the first place, but as it hits the XP, linux and BSD connections equally frequenly, I am assuming it is a function of hardware or IP provider. fwiw, looking at your conf, my ppp.conf contains one setting that yours does not ... enable tcpmssfixup there's a bit of discussion abt it here: http://lists.apple.com/archives/Darwin-development/2001/Aug/msg00502.html dunno if this is related -- just throwing it out there. cheers. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
setting ENV VARs in make.conf/ports.conf ?
hi,
i've installed FreeBSD v6.2-RELEASE, plus 'portconf' 'ccache' from ports.
i've setup,
/etc/make.conf
-
PERL_VER=5.8.8
PERL_VERSION=5.8.8
USE_OPENSSL_PORT=true
OPENSSLBASE=/usr/local
USE_OPENSSH_PORT=true
WITHOUT_X11=yes
CPUTYPE?=pentium-mmx
CFLAGS= -O2 -pipe
COPTFLAGS= -O2 -pipe
## Begin portconf settings
## Do not touch these lines
.if !empty(.CURDIR:M/usr/ports*) exists(/usr/local/libexec/portconf)
_PORTCONF!=/usr/local/libexec/portconf
.for i in ${_PORTCONF:S/|/ /g}
${i:S/%/ /g}
.endfor
.endif
## End portconf settings
.if !defined(NOCCACHE)
CC=/usr/local/libexec/ccache/cc
CXX=/usr/local/libexec/ccache/c++
CPP=/usr/local/libexec/ccache/cpp
.endif
-
so that i can turn OFF use of ccache for individual ports.
i find that different ports 'respond' only to specific ways of setting
the NOCCACHE var.
e.g., in,
cat /usr/local/etc/ports.conf
-
*:\
CFLAGS= -O2 -pipe -funroll-loops -s -frename-registers
-fomit-frame-pointer | \
CXXFLAGS= -O2 -pipe -funroll-loops -s -frename-registers
databases/sqlite3-threads:NOCCACHE=true | \
WITHOUT_DEBUG=true|WITH_DOCS=true|WITH_FTS1=true|WITH_TCLWRAPPER=true|BATCH=Yes
sysutils/pflogx:MAKE_ENV= NOCCACHE=true | \
WITH_EXPAT=true|BATCH=Yes
-
pflogx and sqlite3 require DIFFERENT specifications of NOCCACHE ...
swapping formate, e.g., causes the conditional .if
!defined(NOCCACHE) to fail; meaning, that the CC/CXX/CPP defs
pointing to ccahce/* are oncorrectly used.
is this expected/normal? i would've expected the same method of
defining ENV VAR knobs ...
is there a _consistent_ way of turning off NOCCACHE via the ports.conf?
thanks!
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: setting ENV VARs in make.conf/ports.conf ?
I'd direct this question to the ports@ mailing list. fair enough. thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
how to force ports to use OPENSSLBASE=/usr/local?
hi, i've an install of FreeBSD 6.2-RELEASE. base ssl is: /usr/bin/openssl version OpenSSL 0.9.7e-p1 25 Oct 2004 i've installed openssl from ports, `which openssl` version OpenSSL 0.9.8e 23 Feb 2007 i'm trying to get a ports-build of openssh-portable to link the ports-installed ssl, OpenSSL 0.9.8e. i've addedto '/etc/make.conf': +++ OPENSSLBASE=/usr/local and, am (re)building with, cd /usr/ports/security/openssh-portable make deinstall rmconfig OPENSSLBASE=/usr/local reinstall clean but, after build, i _still_ see the base-installed ssl used: ldd /usr/local/bin/ssh | grep ssl libssl.so.4 = /usr/lib/libssl.so.4 (0x283bb000) what do i need to set/change for the ports' openssl to be used -- for openssh and, eventually, all subsequent ports? thanks! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
