On 2024/10/01 13:06, Stefan Sperling wrote:
> I don't believe 7.6 -stable commits are being done yet so we
> need to wait a bit longer. No worries.
Right. Hopefully soon, release is in about a week and the Ms are
building up already ;)
Tuesday, October 01, 2024
Re: dhcpd with failover
On Tue, Oct 01, 2024 at 12:34:31PM +0300, Kapetanakis Giannis wrote:
>
> On 01/10/2024 08:36, Otto Moerbeek wrote:
> > On Mon, Sep 30, 2024 at 10:50:06PM +0300, Kapetanakis Giannis wrote:
> >
> >> How did you solve the problem of initial sync?
> >>
> >> Suppose one server goes down for maintenance. When it comes up it does not
> >> know what new leases have been given by the other server which was all the
> >> time up.
> >>
> >> There are chances that later it gives an IP already given by the other dhcp
> >> server. The leases (while down) were not synced.
> >>
> >> I don't think there is an option for initial sync (like pfsync) and even if
> >> you scp the lease file you can't be 100% sure since you might miss something
> >> while copying and starting service.
> >>
> >> This is the primary reason I've sticked with isc-dhcpd and failover peer.
> >>
> >> G
> >>
> > IIRC on dhcpd statup all leases from the lease db are advertized to
> > the other dhcpds.
> >
> > -Otto
>
> Didn't know about this but I don't think it solves the problem.
> The starting server is behind and missing entries from the lease file.
>
> Does the starting server also request a full lease advertisement from the other server?
>
> What happens with the other server (which is NOT restarting) and how will it notify the starting server that there are new leases (without a restart)?
>
> Unless a server startup triggers full advertizement on all sides.
>
> G
>
Look in src/usr.sbin/dhcpd/sync.c, on receiving a lease, dhcpd checks
if the other deamon has old info:
/*
* our partner sent us a lease
* that is older than what we have,
* so re-educate them with what we
* know is newer.
*/
So if the starting dhcpd daemon sends out info that is known to be
outdated, it wil be corrected by the others.
I'm not 100% sure that will cover all cases though.
-Otto
>
> On 01/10/2024 08:36, Otto Moerbeek wrote:
> > On Mon, Sep 30, 2024 at 10:50:06PM +0300, Kapetanakis Giannis wrote:
> >
> >> How did you solve the problem of initial sync?
> >>
> >> Suppose one server goes down for maintenance. When it comes up it does not
> >> know what new leases have been given by the other server which was all the
> >> time up.
> >>
> >> There are chances that later it gives an IP already given by the other dhcp
> >> server. The leases (while down) were not synced.
> >>
> >> I don't think there is an option for initial sync (like pfsync) and even if
> >> you scp the lease file you can't be 100% sure since you might miss something
> >> while copying and starting service.
> >>
> >> This is the primary reason I've sticked with isc-dhcpd and failover peer.
> >>
> >> G
> >>
> > IIRC on dhcpd statup all leases from the lease db are advertized to
> > the other dhcpds.
> >
> > -Otto
>
> Didn't know about this but I don't think it solves the problem.
> The starting server is behind and missing entries from the lease file.
>
> Does the starting server also request a full lease advertisement from the other server?
>
> What happens with the other server (which is NOT restarting) and how will it notify the starting server that there are new leases (without a restart)?
>
> Unless a server startup triggers full advertizement on all sides.
>
> G
>
Look in src/usr.sbin/dhcpd/sync.c, on receiving a lease, dhcpd checks
if the other deamon has old info:
/*
* our partner sent us a lease
* that is older than what we have,
* so re-educate them with what we
* know is newer.
*/
So if the starting dhcpd daemon sends out info that is known to be
outdated, it wil be corrected by the others.
I'm not 100% sure that will cover all cases though.
-Otto
Re: Maintainer Update NSH (2x bug fixes)
On Tue, Oct 01, 2024 at 11:44:21AM +0100, Tom Smyth wrote:
> Thanks Stefan, and all for your help in getting this released,
> if this can be fed into the stable of supported versions of OpenBSD
> (at your convenience) I would appreciate it
Yes, it can.
The -stable packages will switch to 7.6 soon, 7.5 is over.
I don't believe 7.6 -stable commits are being done yet so we
need to wait a bit longer. No worries.
> Thanks Stefan, and all for your help in getting this released,
> if this can be fed into the stable of supported versions of OpenBSD
> (at your convenience) I would appreciate it
Yes, it can.
The -stable packages will switch to 7.6 soon, 7.5 is over.
I don't believe 7.6 -stable commits are being done yet so we
need to wait a bit longer. No worries.
Re: Maintainer Update NSH (2x bug fixes)
Thanks Stefan, and all for your help in getting this released,
if this can be fed into the stable of supported versions of OpenBSD
(at your convenience) I would appreciate it
Thanks
Tom Smyth
On Mon, 30 Sept 2024 at 08:16, Stefan Sperling <stsp@stsp.name> wrote:
>
> On Wed, Sep 18, 2024 at 02:01:24PM +0100, Tom Smyth wrote:
> > Folks
> > Sorry for this, but we came across 2 bugs that were quite annoying and
> > wanted to push this release and update nsh in ports
> > please find attached diff ...
> >
> > Bug fixes:
> >
> > (major) stsp@ fixed "no autoconf4" disabling dhcp on an interface
> > where there was a dhclient.leases.if file in /var/db/ this would
> > prevent static ips being configured on an interface (found by tom)
> > (major) stsp@ fixed nsh so that nsh now checks what Rdomain it was
> > invoked in so that users can switch rdomains when accessing nsh from a
> > ssh client where ssh server is running in a non default rdomain ( like
> > a management network interface in an ISP) found by Tom
> > (minor) show diff-config now does not show striped blank lines in
> > rendering nshrc after boot up (found by tom and stsp@)
> >
> > New features (minor)
> >
> > add logger command for entering messages into the .. by @smytht in #161
>
> Committed now that the ports tree has been unlocked.
>
> I ran 'make makesum' to update the distinfo file as well:
>
> blob - a7e985fb742b98e95d500633060e431d93eb95cc
> file + shells/nsh/distinfo
> --- shells/nsh/distinfo
> +++ shells/nsh/distinfo
> @@ -1,2 +1,2 @@
> -SHA256 (nsh-1.2.2.tar.gz) = EfR0Yya3F7Ej6nfPd+WlqLT+Z8A8ZUb6fcWb1pQamw8=
> -SIZE (nsh-1.2.2.tar.gz) = 228756
> +SHA256 (nsh-1.2.3.tar.gz) = OYBae9a1FwxJkJcG12LD1S7sIteLTaxUuhZIeVvVca0=
> +SIZE (nsh-1.2.3.tar.gz) = 229518
>
> > Index: Makefile
> > ===================================================================
> > RCS file: /cvs/ports/shells/nsh/Makefile,v
> > diff -u -p -r1.52 Makefile
> > --- Makefile 6 Sep 2024 15:45:12 -0000 1.52
> > +++ Makefile 18 Sep 2024 11:58:29 -0000
> > @@ -2,7 +2,7 @@ COMMENT = network switch style shell
> >
> > CATEGORIES = shells net
> > HOMEPAGE = https://www.nmedia.net/nsh/
> > -V = 1.2.2
> > +V = 1.2.3
> > DISTNAME = nsh-${V}
> > SITES = https://github.com/yellowman/nsh/releases/download/v${V}/
> >
>
--
Kindest regards,
Tom Smyth.
if this can be fed into the stable of supported versions of OpenBSD
(at your convenience) I would appreciate it
Thanks
Tom Smyth
On Mon, 30 Sept 2024 at 08:16, Stefan Sperling <stsp@stsp.name> wrote:
>
> On Wed, Sep 18, 2024 at 02:01:24PM +0100, Tom Smyth wrote:
> > Folks
> > Sorry for this, but we came across 2 bugs that were quite annoying and
> > wanted to push this release and update nsh in ports
> > please find attached diff ...
> >
> > Bug fixes:
> >
> > (major) stsp@ fixed "no autoconf4" disabling dhcp on an interface
> > where there was a dhclient.leases.if file in /var/db/ this would
> > prevent static ips being configured on an interface (found by tom)
> > (major) stsp@ fixed nsh so that nsh now checks what Rdomain it was
> > invoked in so that users can switch rdomains when accessing nsh from a
> > ssh client where ssh server is running in a non default rdomain ( like
> > a management network interface in an ISP) found by Tom
> > (minor) show diff-config now does not show striped blank lines in
> > rendering nshrc after boot up (found by tom and stsp@)
> >
> > New features (minor)
> >
> > add logger command for entering messages into the .. by @smytht in #161
>
> Committed now that the ports tree has been unlocked.
>
> I ran 'make makesum' to update the distinfo file as well:
>
> blob - a7e985fb742b98e95d500633060e431d93eb95cc
> file + shells/nsh/distinfo
> --- shells/nsh/distinfo
> +++ shells/nsh/distinfo
> @@ -1,2 +1,2 @@
> -SHA256 (nsh-1.2.2.tar.gz) = EfR0Yya3F7Ej6nfPd+WlqLT+Z8A8ZUb6fcWb1pQamw8=
> -SIZE (nsh-1.2.2.tar.gz) = 228756
> +SHA256 (nsh-1.2.3.tar.gz) = OYBae9a1FwxJkJcG12LD1S7sIteLTaxUuhZIeVvVca0=
> +SIZE (nsh-1.2.3.tar.gz) = 229518
>
> > Index: Makefile
> > ===================================================================
> > RCS file: /cvs/ports/shells/nsh/Makefile,v
> > diff -u -p -r1.52 Makefile
> > --- Makefile 6 Sep 2024 15:45:12 -0000 1.52
> > +++ Makefile 18 Sep 2024 11:58:29 -0000
> > @@ -2,7 +2,7 @@ COMMENT = network switch style shell
> >
> > CATEGORIES = shells net
> > HOMEPAGE = https://www.nmedia.net/nsh/
> > -V = 1.2.2
> > +V = 1.2.3
> > DISTNAME = nsh-${V}
> > SITES = https://github.com/yellowman/nsh/releases/download/v${V}/
> >
>
--
Kindest regards,
Tom Smyth.
Re: dhcpd with failover
On 01/10/2024 08:36, Otto Moerbeek wrote:
> On Mon, Sep 30, 2024 at 10:50:06PM +0300, Kapetanakis Giannis wrote:
>
>> How did you solve the problem of initial sync?
>>
>> Suppose one server goes down for maintenance. When it comes up it does not
>> know what new leases have been given by the other server which was all the
>> time up.
>>
>> There are chances that later it gives an IP already given by the other dhcp
>> server. The leases (while down) were not synced.
>>
>> I don't think there is an option for initial sync (like pfsync) and even if
>> you scp the lease file you can't be 100% sure since you might miss something
>> while copying and starting service.
>>
>> This is the primary reason I've sticked with isc-dhcpd and failover peer.
>>
>> G
>>
> IIRC on dhcpd statup all leases from the lease db are advertized to
> the other dhcpds.
>
> -Otto
Didn't know about this but I don't think it solves the problem.
The starting server is behind and missing entries from the lease file.
Does the starting server also request a full lease advertisement from the other server?
What happens with the other server (which is NOT restarting) and how will it notify the starting server that there are new leases (without a restart)?
Unless a server startup triggers full advertizement on all sides.
G
> On Mon, Sep 30, 2024 at 10:50:06PM +0300, Kapetanakis Giannis wrote:
>
>> How did you solve the problem of initial sync?
>>
>> Suppose one server goes down for maintenance. When it comes up it does not
>> know what new leases have been given by the other server which was all the
>> time up.
>>
>> There are chances that later it gives an IP already given by the other dhcp
>> server. The leases (while down) were not synced.
>>
>> I don't think there is an option for initial sync (like pfsync) and even if
>> you scp the lease file you can't be 100% sure since you might miss something
>> while copying and starting service.
>>
>> This is the primary reason I've sticked with isc-dhcpd and failover peer.
>>
>> G
>>
> IIRC on dhcpd statup all leases from the lease db are advertized to
> the other dhcpds.
>
> -Otto
Didn't know about this but I don't think it solves the problem.
The starting server is behind and missing entries from the lease file.
Does the starting server also request a full lease advertisement from the other server?
What happens with the other server (which is NOT restarting) and how will it notify the starting server that there are new leases (without a restart)?
Unless a server startup triggers full advertizement on all sides.
G
Re: proposal to remove sysutils/fleetctl
On Tue, Oct 01, 2024 at 01:49:39AM -0700, Nick Owens wrote:
> coreos has been defunct for many years, and fleet even moreso since it
> was deprecated in favor of helm and kubernetes.
>
> i speak with some authority as a former coreos employee.
>
> can we remove the sysutils/fleetctl port?
I think so. If anyone wants to do it,
ok tb
Or I'll do it next week.
> coreos has been defunct for many years, and fleet even moreso since it
> was deprecated in favor of helm and kubernetes.
>
> i speak with some authority as a former coreos employee.
>
> can we remove the sysutils/fleetctl port?
I think so. If anyone wants to do it,
ok tb
Or I'll do it next week.
proposal to remove sysutils/fleetctl
coreos has been defunct for many years, and fleet even moreso since it
was deprecated in favor of helm and kubernetes.
i speak with some authority as a former coreos employee.
can we remove the sysutils/fleetctl port?
cheers,
nick
was deprecated in favor of helm and kubernetes.
i speak with some authority as a former coreos employee.
can we remove the sysutils/fleetctl port?
cheers,
nick
Re: OpenBSD nano r6s
On 2024-09-30, Pierre Dupond <76nemo76@gmx.ch> wrote:
> The display is certainly on a serial console but there is no serial console
> on the nano pi r6s (in fact it seems to be only headers on mainboard and
> one has to solder pins).
That's not entirely unusual on this class of machine. At least it looks
easy enough to solder and I guess you could probably run some jumper
cables via the antenna hole. If you can't or don't want to solder it
yourself then I bet if you take the board and some pin strip to a small
mobile phone repair shop they'd be able to do it for you.
> I have then try to boot OpenBSD on a radxa rock pi 5b. On this machine,
> the serial console is available on the GPIO. The device boots but after a while
> the connection is broken (even if the serial console and its speed was specified
> before booting OpenBSD). After this text, you will see the log of the boot
> obtained with the minicom program (the only one which is working).
> With kermit or "cu", the boot is similar but it is not possible
> to type character on keyboard.
>
> Do you have any idea on how to solve this problem? My tests were
> done with OpenBSD 7.5, I should do them again with OpenBSD 7.6.
> sd0: 14804MB, 512 bytes/sector, 30318592 sectors
> softraid0 at root
> scsibus1 at softraid0: 256 targets
> root on rd0a swap on rd0b dump on rd0b
> WARNING: bad clock chip time
> WARNING: CHECK AND RESET THE DATE!
> ŠŠ3ÞÿÚKXø\]‚òŠÿáøáøáøáøáøáøáøáøáøáøáøáøáøáøáøá
At this point output moves from the kernel to userland. That looks like
the port has changed speed. I would first try 1500000 as it's fairly
common on the embedded ARM boards.
--
Please keep replies on the mailing list.
> The display is certainly on a serial console but there is no serial console
> on the nano pi r6s (in fact it seems to be only headers on mainboard and
> one has to solder pins).
That's not entirely unusual on this class of machine. At least it looks
easy enough to solder and I guess you could probably run some jumper
cables via the antenna hole. If you can't or don't want to solder it
yourself then I bet if you take the board and some pin strip to a small
mobile phone repair shop they'd be able to do it for you.
> I have then try to boot OpenBSD on a radxa rock pi 5b. On this machine,
> the serial console is available on the GPIO. The device boots but after a while
> the connection is broken (even if the serial console and its speed was specified
> before booting OpenBSD). After this text, you will see the log of the boot
> obtained with the minicom program (the only one which is working).
> With kermit or "cu", the boot is similar but it is not possible
> to type character on keyboard.
>
> Do you have any idea on how to solve this problem? My tests were
> done with OpenBSD 7.5, I should do them again with OpenBSD 7.6.
> sd0: 14804MB, 512 bytes/sector, 30318592 sectors
> softraid0 at root
> scsibus1 at softraid0: 256 targets
> root on rd0a swap on rd0b dump on rd0b
> WARNING: bad clock chip time
> WARNING: CHECK AND RESET THE DATE!
> ŠŠ3ÞÿÚKXø\]‚òŠÿáøáøáøáøáøáøáøáøáøáøáøáøáøáøáøá
At this point output moves from the kernel to userland. That looks like
the port has changed speed. I would first try 1500000 as it's fairly
common on the embedded ARM boards.
--
Please keep replies on the mailing list.