You can test the above ISO, the new one will have only really few changes.
my server must have a upgrade first from XCP-NG 8.1 to 8.2. Thats done now
I am copying my Server now.
and i hope ich can use he upgrade soon
5 posts were split to a new topic: Ntopng 4.2 not starting
now i let run first my nomal update
than building a copy of my VM
what is to do ?
yum --enablerepo=nethserver-testing update nethserver-subscription\*
looking for upgrades in the software-center and let it run ?
first try was not working, no upgrades found …
I try it again no function
so i am trying this
[root@neth ~]# yum repolist Loaded plugins: changelog, fastestmirror, nethserver_events Determining fastest mirrors epel/x86_64/metalink | 31 kB 00:00:00 * epel: mirror.23media.com ce-base/7/x86_64/signature | 811 B 00:00:00 ce-base/7/x86_64/signature | 3.6 kB 00:00:00 !!! ce-extras/7/x86_64/signature | 811 B 00:00:00 ce-extras/7/x86_64/signature | 2.9 kB 00:00:00 !!! ce-sclo-rh | 3.0 kB 00:00:00 ce-sclo-sclo | 3.0 kB 00:00:00 ce-updates/7/x86_64/signature | 811 B 00:00:00 ce-updates/7/x86_64/signature | 2.9 kB 00:00:00 !!! epel | 4.7 kB 00:00:00 nethforge/7/x86_64/signature | 836 B 00:00:00 nethforge/7/x86_64/signature | 3.6 kB 00:00:00 !!! nethserver-base/7/x86_64/signature | 836 B 00:00:00 nethserver-base/7/x86_64/signature | 2.9 kB 00:00:00 !!! nethserver-updates/7/x86_64/signature | 836 B 00:00:00 nethserver-updates/7/x86_64/signature | 3.6 kB 00:00:00 !!! (1/14): ce-base/7/x86_64/group_gz | 153 kB 00:00:01 (2/14): ce-extras/7/x86_64/primary_db | 206 kB 00:00:01 (3/14): ce-sclo-sclo/7/x86_64/primary_db | 296 kB 00:00:01 (4/14): epel/x86_64/group_gz | 95 kB 00:00:00 (5/14): ce-sclo-rh/7/x86_64/primary_db | 2.9 MB 00:00:04 (6/14): nethforge/7/x86_64/group_gz | 1.9 kB 00:00:00 (7/14): epel/x86_64/updateinfo | 1.0 MB 00:00:02 (8/14): ce-updates/7/x86_64/primary_db | 4.5 MB 00:00:03 (9/14): nethforge/7/x86_64/primary_db | 103 kB 00:00:00 (10/14): epel/x86_64/primary_db | 6.9 MB 00:00:03 (11/14): nethserver-updates/7/x86_64/group_gz | 22 kB 00:00:00 (12/14): nethserver-base/7/x86_64/primary_db | 376 kB 00:00:00 (13/14): nethserver-updates/7/x86_64/primary_db | 626 kB 00:00:00 (14/14): ce-base/7/x86_64/primary_db | 6.1 MB 00:00:06 repo id repo name status ce-base/7/x86_64 CE-Base-7.8.2003 10,070 ce-extras/7/x86_64 CE-Extras-7.8.2003 413 ce-sclo-rh/7/x86_64 CE-SCLo-rh-7.8.2003 6,999 ce-sclo-sclo/7/x86_64 CE-SCLo-sclo-7.8.2003 812 ce-updates/7/x86_64 CE-Updates-7.8.2003 1,134 epel/x86_64 Extra Packages for Enterprise Linux 7 - x86_64 13,473 nethforge/7/x86_64 NethForge 7 248 nethserver-base/7/x86_64 NethServer-7.8.2003 - Base 363 nethserver-updates/7/x86_64 NethServer-7.8.2003 - Updates 630 repolist: 34,142
looks like the old once?
no upgrade option
Upgrade from NethServer 7.8.2003
Upgrade will be available as usual from the Software Center during the coming week.
Thank you to the whole community
And SF wants some… coffee…
Yes it’s slow! Did you try the .torrent download? There is plenty of seeders: it should be faster hopefully!
Yes it is! about 14 - 15 MB/s
Not allowed here.
Hello everyone! Can we update to version 7.9 now? The message does not appear in the software center, nor through ssh. Thank you and greetings!
Please attach the output of
yum history pkg-info nethserver-subscription
I can’t guess the history of the server where you found the Upgrade warning but after some investigations I’m now sure I did a mistake
The effect was an unexpected disclosure of the 7.9 upgrade in the Software Center, or finding a 7.9 system at the end of a new installation where 7.8 was expected.
Luckily it has not been an issue and it is not an issue at all, now that 7.9 is officially released. However I try to give an explanation of what happened: it’s necessary and it’s fun too!
[the long story]
As it often happens, there is a chain of events and mistakes behind an issue like this. Let’s see them:
It starts from far: on Oct 9, 2019 when an additional .spec file was added to nethserver-subscription code repository. This package is responsible for showing the Software Center upgrade warning and starting the upgrade, if automatic updates are enabled.
Our release script can handle just one .spec file safely.
On Nov 13, 2020 I prepared the initial repositories for 7.9. I cloned the .rpm files with hardlinks.
I used to make copy of them, then run a deduplication after the release.
On Nov 19, 2020 I released a bogus nethserver-subscription RPM to 7.9.2009 repositories. The evidence of the error is still in the build logs. A 3.6.5 version was built and uploaded, where a 3.6.6 was expected. Why? I ran old an
releasetagscript, it was missing this safety measure.
At that point I realized the build 3.6.5 was bad. I removed the bogus RPM from 7.9.2009 and fixed the nethsever-subscription build procedure. The two .spec files are now one, so that shouldn’t happen anymore.
But, but… I forgot the hardlink, so the bogus 3.6.5 RPM was creeping into 7.8.2003 too.
In the end, if a 7.8.2003 system has already got 3.6.5 it does not hit the “bug”. If it has an older version or it is a new installation it gets nethserver-subscription 3.6.5 and the upgrade starts.
How to avoid similar problems in the future?
As said, nethserver-subscription build procedure has already been fixed and the
releasetagdoes not accept implicit arguments any more.
Furthermore I’m planning to improve the scripts that do the YUM repo managements when a new minor release arrives. I want to complete them before 7.10.
If you have any doubt or question I’m glad to provide further explanations! Thank you for reading.
PS: and the bogus .rpm? I replaced it with a backup
H David, this server was installed fresh from ISO. The announcement is no problem for me, but surprising and funny. It is also a mistake without consequences and people make mistakes. I know that best.
Updated a test-vm from 7.8 with local AD, red, green.
Installed Apps: Antivir, ntopng, collectd-chats, firewall, fileserver, guacamole, IPS, Sogo, threadshield, vpn, webproxy and filter.
Every thing seem to work. No relevant error in logs. Only a warning about guacamole:
WARNING [localhost-startStop-2] org.apache.catalina.loader.WebappClassLoaderBase.clearReferencesThreads The web application [guacamole] appears to have started a thread named [com.google.inject.internal.util.$Finalizer] but has failed to stop it. This is very likely to create a memory leak. Stack trace of thread:
@mrmarkuz can you say if it’s relevant?
Great work as always @davidep! Thank’s a lot!!
Update from 7.8 to 7.9. with modules:
… everything was fine, but Webtop was broken.
The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request GET /webtop.
Reason: Error reading from remote server“
I reverted to 7.8
That’s strange because we already have WebTop running on 7.9.
Could you please provide the errors from
In the meanwhile maybe @lucag can confirm us that WT is working good on 7.9