Koozali.org: home of the SME Server
Contribs.org Forums => Koozali SME Server 10.x => Topic started by: p-jones on December 10, 2022, 03:11:44 AM
-
Anyone else got a bricked DHCP server after this weeks updates ??
-
what do you mean by brick ? I you say by phone to your car specialist my car does not work, he will need more details to help you.
what is your server mode ?
any contribs ?
what exact updates ?
just did the update on few servers with or without network contribs and dhcpd was working after
-
https://bugs.koozali.org/show_bug.cgi?id=12257
-
see Bug for proposed resolution, further testing then update expected
-
e-smith-base update is in smeupdates-testing new bridge-interface package is in /smetest for same see Bu8 12271 and 12257
-
I got that DHCP failed to start after a server update this morning.
Here's the dhcpd.log file entry:
Dec 21 08:49:43 mail dhcpd: Internet Systems Consortium DHCP Server 4.2.5
Dec 21 08:49:43 mail dhcpd: Copyright 2004-2013 Internet Systems Consortium.
Dec 21 08:49:43 mail dhcpd: All rights reserved.
Dec 21 08:49:43 mail dhcpd: For info, please visit https://www.isc.org/software/dhcp/
Dec 21 08:49:43 mail dhcpd: WARNING: Host declarations are global. They are not limited to the scope you declared them in.
Dec 21 08:49:43 mail dhcpd: Not searching LDAP since ldap-server, ldap-port and ldap-base-dn were not specified in the config file
Dec 21 08:49:43 mail dhcpd: Wrote 0 deleted host decls to leases file.
Dec 21 08:49:43 mail dhcpd: Wrote 0 new dynamic host decls to leases file.
Dec 21 08:49:43 mail dhcpd: Wrote 141 leases to leases file.
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd: No subnet declaration for eth0 (no IPv4 addresses).
Dec 21 08:49:43 mail dhcpd: ** Ignoring requests on eth0. If this is not what
Dec 21 08:49:43 mail dhcpd: you want, please write a subnet declaration
Dec 21 08:49:43 mail dhcpd: in your dhcpd.conf file for the network segment
Dec 21 08:49:43 mail dhcpd: to which interface eth0 is attached. **
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd: Not configured to listen on any interfaces!
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd: This version of ISC DHCP is based on the release available
Dec 21 08:49:43 mail dhcpd: on ftp.isc.org. Features have been added and other changes
Dec 21 08:49:43 mail dhcpd: have been made to the base software release in order to make
Dec 21 08:49:43 mail dhcpd: it work better with this distribution.
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd: Please report for this software via the CentOS Bugs Database:
Dec 21 08:49:43 mail dhcpd: http://bugs.centos.org/
Dec 21 08:49:43 mail dhcpd:
Dec 21 08:49:43 mail dhcpd: exiting.
What do I need to do to resolve this?
-
If no bridge interface
http://mirror.canada.pialasse.com/releases/10/smeupdates-testing/x86_64/RPMS/e-smith-base-5.8.1-23.el7.sme.x86_64.rpm
Then add a comment to the bug that it did the job,
https://bugs.koozali.org/show_bug.cgi?id=12257
and if have the bridge interface installed as well
http://mirror.canada.pialasse.com/releases/10/smetest/x86_64/RPMS/smeserver-bridge-interface-0.2-8.el7.sme.noarch.rpm
and its bug to also add comments
https://bugs.koozali.org/show_bug.cgi?id=12271
When they are both verified as working will be moved to updates and contribs
-
Thank you! It’s a challenge to get this installed because it’s also our gateway server. Had to copy over in thumb drive. Also had to use rpm —force -i {file} because it said the config files clashed. I’m typing on my phone to enter this. When it’s all back online again, I’ll post more details.
-
I removed the vpnserver and bridge interface, restarted
I then downloaded and copied onto the server e-smith-base-5.8.1-23.el7.sme.x86_64.rpm via thumb drive because the network is broken
rpm --force -i e-smith-base-5.8.1-23.el7.sme.x86_64.rpm
signal-event e-smith-base-update
signal-event post-upgrade; signal-event rebootsystemctl status dhcpd says that it's running
but I can't get an address on my workstations
I also can't access the outside world.
I've also tried to login as admin and run the server config, and after reboot no change.
It seems that I've screwed up my network config somehow. Is there a way to roll back an update?
-
suspect the removal of sofether may be the issue..
is system server/gateway or server only
what eth port are you now using? swap the eth cables around
-
Thanks @TerryF!
That was indeed the problem - I swapped ethernet cables and it seems to be working now.
My server is setup as server/gateway. I removed the softether VPN thinking that it would simplify the problem, but what it did was swap the ethernet configs.
I had to re-run the admin > configure server option and pay close attention to the mac addresses of the ethernet cards, then problem sorted.
-
I applied the 2 patches that were pushed out today and the DHCP server came to life again.
Later in the day I shut it down. Bought it back up, did a reconfigure & restart and now the DHCP doesnt work again.
adjusting non-supervised dhcpd (start)
Dec 22 22:07:49 server2 esmith::event[10642]: Job for dhcpd.service failed because the control process exited with error code. See "systemctl status dhcpd.service" and "journalctl -xe" for details.
Dec 22 22:07:49 server2 esmith::event[10642]: serviceControl: Couldn't system( /usr/bin/systemctl start dhcpd.service): No such file or directory
Dec 22 22:07:49 server2 esmith::event[10642]: adjusting non-supervised dhcpd (sigterm)
Dec 22 22:07:49 server2 esmith::event[10642]: Job for dhcpd.service failed because the control process exited with error code. See "systemctl status dhcpd.service" and "journalctl -xe" for details.
Dec 22 22:07:49 server2 esmith::event[10642]: serviceControl: Couldn't system( /usr/bin/systemctl restart dhcpd.service): No such file or directory
Dec 22 22:07:49 server2 esmith::event[10642]: adjusting non-supervised nmbd (start)
Dec 22 22:07:49 server2 esmith::event[10642]: adjusting non-supervised nmbd (sigterm)
Dec 22 22:07:50 server2 esmith::event[10642]: adjusting non-supervised smbd (start)
Dec 22 22:07:50 server2 esmith::event[10642]: adjusting non-supervised smbd (sigterm)
Dec 22 22:07:51 server2 esmith::event[10642]: adjust-services=action|Event|workgroup-update|Action|adjust-services|Start|1671700068 975008|End|1671700071 60935|Elapsed|2.085927
-
two patches? so you were also running the bridge interface?
-
Yes. The Open VPN Bridge
-
I should also add that I could not shut it down from the GUI management screen. I had to shut it down from the command line.
That is the reason for the reconfig/reboot and the rest is as above.
-
in interim, from cli restart the service or try anyway
# systemctl restart dhcpd.service
doing some digging, looking into a remote system with the bridge interface that has been updated..just need to be careful as it is remote and I get to it using the softher contrib :-) if i bork it have to drive up to and sort locally..and that aint happening until sunday :-(
Added: what shows up with # cat /usr/lib/systemd/system/dhcpd.service.d/50koozali.conf
-
Time for the finger pointing - me
Did not correctly update a line in the bridge-interface template patch for the
/etc/e-smith/templates/usr/lib/systemd/system/dhcpd.service.d/50koozali.conf/41service
*insert rude and graphic words here
working on it
-
in interim, from cli restart the service or try anyway
# systemctl restart dhcpd.service
Worked 4 me.
-
see Bug 12286 seems to be good, dhcp doesn't die in ways others were..
https://bugs.koozali.org/show_bug.cgi?id=12286#c5