Koozali.org: home of the SME Server
Obsolete Releases => SME Server 9.x => Topic started by: turandot on April 14, 2016, 08:05:57 PM
-
I was suffering from the problems described here: https://forums.contribs.org/index.php/topic,52402.0.html
To overcome the problems, uninstall the patches. This is achieved some commands on a root console. How to do that: https://access.redhat.com/solutions/64069
First check the ID number of the patch:
yum history
Then uninstall the related patches by
yum history undo <ID number>
To be 100% safe, I also applied
signal-event post-upgradeand
signal-event reboot
I hope this helps other users, I had to find it out :lol:
turandot
EDIT: added YET to subject.
-
As announced on all mailing list at contribs :
Several people are reporting an issue with the various samba updates
released yesterday (2016-04-13) to fix the mediatic badlock
vulnerability. After installing those updates, Windows clients members
of your domain will loose the trust relationship, making any domain
login impossible.
Please, do not apply the update for now, we need some time to
investigate and work with our upstream vendor to get this fixed.
You can follow the bug here:
https://bugs.contribs.org/show_bug.cgi?id=9448 and here:
https://bugzilla.redhat.com/show_bug.cgi?id=1326918
-
Hi,
thanks for the information!
that cost me today about five hours work.
yum history undo <ID number>etc, saved My life
The next time I read here first :-?
-
Our wiki holds TONS of information :wink:
https://wiki.contribs.org/Useful_Commands#YUM.27ing_and_repositories (https://wiki.contribs.org/Useful_Commands#YUM.27ing_and_repositories)
-
Our wiki holds TONS of information :wink:
https://wiki.contribs.org/Useful_Commands#YUM.27ing_and_repositories (https://wiki.contribs.org/Useful_Commands#YUM.27ing_and_repositories)
Yes, the wiki is great! But my "five hours work" I need before I read this thread and before I see the need of this command.....
Can someone tell us if the problem is solved?
-
ONLY if you follow upstream, for there is where the issue resides.
-
I set up a new 9.1 server over the weekend, still testing it and ran into this problem. Running "yum downgrade samba3x\* libsmbclient" does nothing since there is nothing to roll back too. This server isn't in use yet. Should I just wait for the fix from Redhat or remove the new version and manually download/install the previous one?
-
9.1 is not affected until you apply yum updates (in which case you should be able to rollback). We're still waiting for a fix from upstream. They already have a fix currently being tested, but not released yet
-
9.1 is not affected until you apply yum updates (in which case you should be able to rollback). We're still waiting for a fix from upstream. They already have a fix currently being tested, but not released yet
Thanks! For reasons I couldn't determine, the same rollback command worked the next day. Before that it would return a "nothing to do" message. The Win 8.1 and 10 clients I have joined to it login in fine now.
-
I am checking the Redhat bug tracker https://bugzilla.redhat.com/show_bug.cgi?id=1326918 once a while. The bug is neither fixed upstream within CentOS, nor with SME yet?
turandot
-
We're still waiting for Red Hat to release the fix
-
We're still waiting for Red Hat to release the fix
Andreas Schneider 2016-04-19 01:52:27 EDT
If you require a hotfix until we are able to roll out an update please talk to your Red Hat support contact!
No, we're waiting for them to extract a bit more cash from lots of desperate customers = ! :-o
-
Anyone applied recent 6.8 upstream samba packages? Are these fixed by any chance?
-
I hope this helps other users, I had to find it out :lol:
No need to find this out. Simply read the wiki https://wiki.contribs.org/Useful_Commands#YUM.27ing_and_repositories
-
The new updates show versions samba x86_64 3.6.23-35.el6_8
Version that first caused the issues was samba-3.6.23-30.el6_7.x86_64
BUT, Red Hat bug 1326918, https://bugzilla.redhat.com/show_bug.cgi?id=1326918 that covers the problem HAS NOT been resolved, so I would assume the bug is still evident, even in the latets samba update.
-
Heres a fix from git.samba if you want to delve into the code: https://git.samba.org/?p=asn/samba.git;a=commitdiff;h=82fa625540abf8b8ec23d43c41e2ca906a9928a5;hp=ea6f2386611d0a4edd65962a59b3448be976c1bb
and the original bug is still active at CentOS
https://bugs.centos.org/view.php?id=10716
and
https://bugs.centos.org/view.php?id=10740
-
I've installed samba updates from EL6.8 and it seems to be working (but I only have a single Win7 client)
-
I did some tests..
on an old VM with SME 9.1 and samba-3.6.23-24.el6_7.x86_64 I enabled the domain controller feature, joined a virtual w2003 machine, logged successfully with 2 users..
I then updated SME and now I have samba-3.6.23-35.el6_8.x86_64
I'm still able to login with the users I tested before and with some other users..
IMO this seems resolved
Dani: should we update bug 9448?
-
After updating on a real machine and real clients, I don't face any issues as with samba-3.6.23-30.el6_7.x86_64, so to me it looks as solved. No complaints for the last 24 hours.
-
After updating on a real machine and real clients, I don't face any issues as with samba-3.6.23-30.el6_7.x86_64, so to me it looks as solved. No complaints for the last 24 hours.
yum update today. => SME Server 9.1, samba-3.6.23-35.el6_8
yes, the same goes for me too. Domain login and BackupPC now working again with the current version of Samba.
But I am surprised that so little response to this serious problem follows ... ???
-
3.6.23-35 fixes this https://rhn.redhat.com/errata/RHBA-2016-0992.html
-
Works fine with me as well. I think that this story can be closed now.
turandot