Re: syslogd: using IPv6 as hostnames results in "IP mismatch"

Hi, > On Thu, 31 Jan 2019 08:24:38 +0100 > "O. Hartmann" said: ohartmann> validate: dgram from IP ffdff:dead:beef::, port 514, name \ ohartmann> fdff:dead:beef::; ohartmann> rejected in rule 1 due to IP mismatch. The -a option was broken. It should be fixed now. Please,

Re: syslogd: using IPv6 as hostnames results in "IP mismatch"

Hi, > On Thu, 31 Jan 2019 08:24:38 +0100 > "O. Hartmann" said: ohartmann> validate: dgram from IP ffdff:dead:beef::, port 514, name \ ohartmann> fdff:dead:beef::; ohartmann> rejected in rule 1 due to IP mismatch. The -a option was broken. It should be fixed now. Please try

Re: syslogd: using IPv6 as hostnames results in "IP mismatch"

Hi, > On Thu, 31 Jan 2019 08:24:38 +0100 > "O. Hartmann" said: ohartmann> validate: dgram from IP ffdff:dead:beef::, port 514, name \ ohartmann> fdff:dead:beef::; ohartmann> rejected in rule 1 due to IP mismatch. The -a option was broken. It should be fixed now. Please try

Re: syslogd 100% cpu usage on recent FreeBSD version

On 12/24/16 13:50, Ngie Cooper (yaneurabeya) wrote: On Dec 24, 2016, at 04:14, Subbsd wrote: Probably after https://svnweb.freebsd.org/base?view=revision=310494, syslogd eat 100% cpu with follow messages: Dec 24 14:19:15 samson syslogd: select: Bad file descriptor Dec 24

Re: syslogd: select: Bad File descriptor

> On 24 Dec 2016, at 2:51 PM, Ngie Cooper (yaneurabeya) > wrote: > > >> On Dec 24, 2016, at 04:16, Daniel Braniss wrote: >> >> latest changes is causing cpu load and ‘last message repeated >> times, I guess the eggnog is affecting too

Re: syslogd: select: Bad File descriptor

> On Dec 24, 2016, at 04:16, Daniel Braniss wrote: > > latest changes is causing cpu load and ‘last message repeated times, > I guess the eggnog is affecting too early Fixed in r310504. Thanks, -Ngie signature.asc Description: Message signed with OpenPGP using

Re: syslogd 100% cpu usage on recent FreeBSD version

On 12/24/16 13:14, Subbsd wrote: Probably after https://svnweb.freebsd.org/base?view=revision=310494, syslogd eat 100% cpu with follow messages: Dec 24 14:19:15 samson syslogd: select: Bad file descriptor Dec 24 14:19:45 samson last message repeated 464140 times Dec 24 14:20:38 samson last

Re: syslogd 100% cpu usage on recent FreeBSD version

> On Dec 24, 2016, at 04:14, Subbsd wrote: > > Probably after https://svnweb.freebsd.org/base?view=revision=310494, > syslogd eat 100% cpu with follow messages: > > Dec 24 14:19:15 samson syslogd: select: Bad file descriptor > Dec 24 14:19:45 samson last message repeated

Re: syslogd no longer listens (or sends) on a network socket

On 12/19/16 22:42, Hiroki Sato wrote: Michael Butler wrote in : im> On 12/19/16 12:12, Hiroki Sato wrote: im> > Michael Butler wrote im> > in

Re: syslogd no longer listens (or sends) on a network socket

Michael Butler wrote in : im> On 12/19/16 12:12, Hiroki Sato wrote: im> > Michael Butler wrote im> > in : im> >

Re: syslogd no longer listens (or sends) on a network socket

On 12/19/16 12:12, Hiroki Sato wrote: Michael Butler wrote in : im> It appears that SVN r309925 and onward no longer opens a network im> socket unless the command-line explicitly contains "-b :syslog"

Re: syslogd no longer listens (or sends) on a network socket

Michael Butler wrote in : im> It appears that SVN r309925 and onward no longer opens a network im> socket unless the command-line explicitly contains "-b :syslog" :-( im> im> This also stops one syslog

Re: syslogd: Remote Logging busted?

On Fri, Oct 28, 2011 at 7:22 PM, Larry Rosenman l...@lerctr.org wrote: I enabled remote logging for my home subnet, and syslogd doesn't seem(!) to be logging the messages. They ARE making it to the system. Can someone look at bin/162135 which has all the details, including tcpdump to show

Re: syslogd: Remote Logging busted?

On Fri, 28 Oct 2011, Kevin Oberman wrote: On Fri, Oct 28, 2011 at 7:22 PM, Larry Rosenman l...@lerctr.org wrote: I enabled remote logging for my home subnet, and syslogd doesn't seem(!) to be logging the messages. They ARE making it to the system. Can someone look at bin/162135 which has

Re: syslogd: Remote Logging busted?

On Fri, Oct 28, 2011 at 8:37 PM, Larry Rosenman l...@lerctr.org wrote: On Fri, 28 Oct 2011, Kevin Oberman wrote: On Fri, Oct 28, 2011 at 7:22 PM, Larry Rosenman l...@lerctr.org wrote: I enabled remote logging for my home subnet, and syslogd doesn't seem(!) to be logging the messages. They

Re: syslogd: Remote Logging busted?

On Fri, 28 Oct 2011, Kevin Oberman wrote: OK. I'm baffled! I can't see anything that looks wrong, but I'll think about it a bit more. See my reply to Stas (cc'd to you). The issue is the damn cable modem is sending the packets from random source PORTS, so the -a entry needed a :* after

Re: syslogd: Too many '/' in /dev//console

On Wed, Sep 05, 2001 at 09:52:34AM +0300, Giorgos Keramidas wrote: I have it fixed now in my local CVS tree. Hopefully Kris will commit something to fix it soon :-) I fixed this a couple of hours ago. Kris PGP signature

Re: syslogd: Too many '/' in /dev//console

From: Mike Heffner [EMAIL PROTECTED] Subject: Re: syslogd: Too many '/' in /dev//console Date: Tue, Sep 04, 2001 at 07:55:33PM -0400 On 04-Sep-2001 Giorgos Keramidas wrote: | | The following patch seems to have fixed the bug for me. | Yea, Kris said he was going to fix it. This must

Re: syslogd: Too many '/' in /dev//console

From: Giorgos Keramidas [EMAIL PROTECTED] Subject: Re: syslogd: Too many '/' in /dev//console Date: Tue, Sep 04, 2001 at 06:39:36AM +0300 I'm looking at the diffs from Aug 25, so if I come up with sth by running syslogd with -d, by tomorrow I'll have spotted this in more detail - probably

Re: syslogd: Too many '/' in /dev//console

On 04-Sep-2001 Giorgos Keramidas wrote: | | The following patch seems to have fixed the bug for me. | Yea, Kris said he was going to fix it. This must be some undefined behavior because I tested the change in a test program and the two sizeofs were giving me the same result..strange ;) Mike

Re: syslogd: Too many '/' in /dev//console

On Monday, September 3, 2001, at 03:42 PM, [EMAIL PROTECTED] wrote: Between last weekend and this weekend, something changed in syslogd seems to have resulted in this boot-time error. The syslogd.c deltas from 1.82 - 1.83 look suspect since the handling of relevant variables has

RE: syslogd: Too many '/' in /dev//console

On 03-Sep-2001 [EMAIL PROTECTED] wrote: | Between last weekend and this weekend, something changed in syslogd | seems to have resulted in this boot-time error. The syslogd.c deltas | from 1.82 - 1.83 look suspect since the handling of relevant variables | has changed. This change looks

RE: syslogd: Too many '/' in /dev//console

On 04-Sep-2001 Mike Heffner wrote: | | On 03-Sep-2001 [EMAIL PROTECTED] wrote: || Between last weekend and this weekend, something changed in syslogd || seems to have resulted in this boot-time error. The syslogd.c deltas || from 1.82 - 1.83 look suspect since the handling of relevant

Re: syslogd: Too many '/' in /dev//console

On Tue, Sep 04, 2001 at 12:28:28AM -0400, Mike Heffner wrote: On 03-Sep-2001 [EMAIL PROTECTED] wrote: | Between last weekend and this weekend, something changed in syslogd | seems to have resulted in this boot-time error. The syslogd.c deltas | from 1.82 - 1.83 look suspect since the

Re: syslogd and -a

On Sun, Jul 01, 2001 at 09:20:44PM -0700, Crist J. Clark wrote: Hmmm... Looks like, # syslogd -a 192.168.1.0/29 Will work and, # syslogd -a 192.168.1.1/29 Won't. That's the standard behaviour of a netmask, isn't it? The usual way to check if host h is in network/netmask n/m is

Re: syslogd and -a

On Mon, Jul 02, 2001 at 09:38:42AM +0100, David Malone wrote: On Sun, Jul 01, 2001 at 09:20:44PM -0700, Crist J. Clark wrote: Hmmm... Looks like, # syslogd -a 192.168.1.0/29 Will work and, # syslogd -a 192.168.1.1/29 Won't. That's the standard behaviour of a netmask,

Re: syslogd and -a

On Mon, 2 Jul 2001 09:38:42 +0100 David Malone [EMAIL PROTECTED] wrote: On Sun, Jul 01, 2001 at 09:20:44PM -0700, Crist J. Clark wrote: Hmmm... Looks like, # syslogd -a 192.168.1.0/29 Will work and, # syslogd -a 192.168.1.1/29 Won't. That's the standard behaviour of

Re: syslogd and -a

On Mon, 2 Jul 2001 08:25:38 -0700 Crist J. Clark [EMAIL PROTECTED] said: cristjc That's exactly what happens in the syslogd(8) code. However, I think cristjc that should be, cristjc n = m cristjc . cristjc . cristjc . cristjc ((h m) == n) I think it should be: ((h m) == (n

Re: syslogd and -a

On Mon, Jul 02, 2001 at 09:38:42AM +0100, David Malone wrote: On Sun, Jul 01, 2001 at 09:20:44PM -0700, Crist J. Clark wrote: Hmmm... Looks like, # syslogd -a 192.168.1.0/29 Will work and, # syslogd -a 192.168.1.1/29 Won't. That's the standard behaviour of a netmask,

Re: syslogd and -a

On Mon, 2 Jul 2001 12:25:42 -0700 Kris Kennaway [EMAIL PROTECTED] said: kris This doesn't seem to work with IPv6. Isn't there a libc function kris which can be used to do this? Yup, there is no api for masking address ether libc nor standard. I'll commit the following patch for IPv6: Index:

Re: syslogd and -a

On Sun, Jul 01, 2001 at 11:41:25PM -0400, David Hill wrote: Hello - It seems the -a option for syslogd does not work 100%. I need to log from hosts from 192.168.1.1-.6 doing /usr/sbin/syslogd -a 192.168.1.1/29 does not work (nothing gets logged) but, if i do /usr/sbin/syslogd -a

Re: syslogd and -a

On 02-Jul-01 (04:20:44/GMT) Crist J. Clark wrote: It seems the -a option for syslogd does not work 100%. Hmmm... Looks like, # syslogd -a 192.168.1.0/29 Will work and, # syslogd -a 192.168.1.1/29 Won't. Under 4.3-STABLE is the same. To capure log from router I added (in rc.conf) -a

Re: syslogd(8) does not update hostname

I don't agree with this change. hostname != name-that-IP-address-resolves-to. I can see how loggin the IP address (or some manifestation thereof) to a central logger, but this is too strong. Example: My laptop has a hostname set for my home network, and I connect it to my work network and DHCP

Re: syslogd(8) does not update hostname

On Sat, Jan 20, 2001 at 11:24:16AM +0200, Mark Murray wrote: I don't agree with this change. hostname != name-that-IP-address-resolves-to. Dunno what you are talking about. That has nothing directly to do with this. No one is talking about forcing you to change your hostname. The patch just

Re: syslogd(8) does not update hostname

On Sat, Jan 20, 2001 at 11:24:16AM +0200, Mark Murray wrote: I don't agree with this change. hostname != name-that-IP-address-resolves-to. Dunno what you are talking about. That has nothing directly to do with this. No one is talking about forcing you to change your hostname. The

Re: syslogd

On Sun, 16 Jan 2000, Vladimir B. Grebeschikov wrote: On Sat, 15 Jan 2000, Doug White wrote: # log firewall messages ONLY in this file (noy in messages below) !!ipfw *.* /var/log/ipfw This is a bad example. ipfw messages come from the kernel so you can't filter those.