DevHeads.net

Postings by Gerhard Schneider

Re: [CentOS] vmplayer for linux 12.5.7 on CentOS 7.4

<a href="https://communities.vmware.com/message/2686431" title="https://communities.vmware.com/message/2686431">https://communities.vmware.com/message/2686431</a>

is working for me.

GS

Spurious Segmentation Fault after upgrade to 7.4

After upgrading approx.

Accountsservice in CentOS 7.3

Is there any known problem with accountsservice-0.6.35-12 with NFS users?
After logging in via lightdm they are not "remembered" and there is no
entry created to /var/lib/AccountsService/users

Downgrading to accountsservice-0.6.35-9 resolved that issue

Gerhard Schneider

7.3: Gnome workspaces mirrored

An user is using 2 displays and configured Gnome to "Static Workspaces
only on primary display on" using gnome-tweak-tool

Till 7.2 it was working as expected. After upgrading to 7.3 the content
on the primary screen gets mirrored to all workspaces.

When switching to "Static Workspaces only on primary display off"
everything is working as expected on 7.3, too.

Any ideas?

Thank you in advance!

Gerhard Schneider

Re: [CentOS] Some thoughts about EL 6

Not from my side. I'm CentOS user - only testing 6.0

The tape issue has been discussed on the linux-scsi mailing list 6
months ago - including some fix - and there was a bug report for the
Ubuntu 2.6.32 kernel.

The slab issue has been discussed in the rhelv6-list in February.

Sorry, but the large servers I used for testing are back in business
with CentOS 5.6 after playing around for one week (it's my way to check
that issues are not hardware related).

Gerhard Schneider

Some thoughts about EL 6

Many people seem to wait for the announcement of CentOS 6.0, so I want
to share some test results I did with SL 6.0..

The actual 6.0 kernel can NOT allocate tape buffers when the server is
heavily loaded at least on some LSILogic hardware.
There is a big problem of slab buffer increase that can cause
reboot/freeze of the server under load. Reported by many and verified by
me :-(

So perhaps it's a good idea to wait for 6.1 for mission critical servers..

Gerhard Schneider

WD RE4-GP Dropped From Raid

Did you check if you already have the G05 firmware on all RE4-GP?
The G04 firmware is not suitable for RAID.

<a href="http://forum.qnap.com/viewtopic.php?f=182&amp;t=25057" title="http://forum.qnap.com/viewtopic.php?f=182&amp;t=25057">http://forum.qnap.com/viewtopic.php?f=182&amp;t=25057</a>

GS

Re: [CentOS] Any update on 5.6 / 6?

Perhaps I'm a little bit pragmatic, but..

I will upgrade to 5.6 when it's out - BUT I'm not really angry when
CentOS 6 would be CentOS 6.1 - if you want to be beta tester for a large
distibution you still can shift to Ubuntu..

GS

Re: [CentOS] Some RAID-6 observations ... RHEL-6 vs CentOS-5.5

Sorry, I don't use CentOS 6 now so I cannot check, but:

Could it be that RHEL6 changed the default superblock mdadm is creating?
CentOS 5 is creating a 0.9 superblock. Some "other OS" are creating 1.2
per default. But you can change that on the command line.

If that's the case it's not a bug but a feature..

GS

No kernel-debuginfo for actual kernel?

After a crash I tried
wget
<a href="http://debuginfo.centos.org/5/x86_64/kernel-debuginfo-2.6.18-194.32.1.el5.x86_64.rpm" title="http://debuginfo.centos.org/5/x86_64/kernel-debuginfo-2.6.18-194.32.1.el5.x86_64.rpm">http://debuginfo.centos.org/5/x86_64/kernel-debuginfo-2.6.18-194.32.1.el...</a>
as usual.

But for 32.1 there is no kernel-debuginfo.

Is there any reason for that?

Thank you in advance!

Gerhard Schneider

Re: [CentOS] Intel NIC

The problem with VMWare Server is that it is a discontinued product for
longer time and they don't provide us with a suitable replacement.

So our institute had to switch to VMWare workstation that can be run as
a server, too. I'm running CentOS 5.5 as the Host OS and 5.5 and
RHEL6beta as guests. We didn't try out RHEL6 (waiting for CentOS6 :-)

GS

2.6.18-194.26.1.el5 NFS4 race conditions?

Are there any known race conditions with 2.6.18-194.26.1.el5 x86_64 and
NFS4?

Today a user managed to shut down two identical Intel servers by copying
a simple text file via NFS4.

Reverting back to 17.1 made the error disappear.

On other non-identical hardware the error was not reproducible (so it
must be a combination of hardware and this special kernel revision).

Anybody interested in a kernel dump? It's only 16GB. I could install
kernel-debuginfo and try to get something out of the dump..

Gerhard Schneider

Ac1dB1tch3z Vs Linux Kernel x86_64 0day

Are there any 64bit CentOS5 kernels available that are immune against
the exploit mentioned in the subject? Turning off 32bit support is no
option to me..

Gerhard Schneider

P.S.: Source code can be found at
<a href="http://seclists.org/fulldisclosure/2010/Sep/268" title="http://seclists.org/fulldisclosure/2010/Sep/268">http://seclists.org/fulldisclosure/2010/Sep/268</a> and is working "well" on
2.6.18-194.11.3.el5.centos.plus