>>> On 10/6/2016 at 01:54 PM, Michael MacIsaac wrote:
> Is there an option at install time to not use UUIDs with brtfs?
Yes, but I can't say what the consequences are.
When you get to the "Suggested Partitioning" screen, click on
"Expert Partitioner"
"Hard Disks"
dasda
/dev/dasda3
"Fstab Optio
Wait, maybe we don't have it fixed. We too still have the same
*root=UUID=<32-digits>* kernel parm in our /proc/cmdlines.
Is there an option at install time to not use UUIDs with brtfs? You might
think this would have been addressed in today's age of virtualization.
-Mike MacIsaac
--
Isaac
Sent: Wednesday, October 05, 2016 1:47 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] Linux system booted with old copy?
Scott,
If I were the one who fixed it, yes I would have provided more details. I did
dig up this from an e-mail:
There is still a uuid parm being passed to the
Scott,
If I were the one who fixed it, yes I would have provided more details. I
did dig up this from an e-mail:
There is still a uuid parm being passed to the kernel at boot:
Performing 'kexec -l /boot/image-3.12.59-60.45-default
--initrd=/boot/initrd-3.12.59-60.45-default
--command-line=root=UU
With no details of how? tsk tsk...
Scott Rohling
On Wed, Oct 5, 2016 at 11:50 AM, Michael MacIsaac
wrote:
> We got this fixed. Thanks for everyone's help.
>
> -Mike MacIsaac
>
> --
> For LINUX-390 subscribe / signoff / a
We got this fixed. Thanks for everyone's help.
-Mike MacIsaac
--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.mari
>>> On 10/4/2016 at 04:32 PM, Marcy Cortes
>>> wrote:
> The virtualization cookbook Redbooks says
>
> 1. Change the persistent policy:
> # echo 'persistent_policy=by-path' >> /etc/dracut.conf
> 2. Update the boot loader:
> # grub2-install
> This updates the initramfs image of the first stage b
U
Subject: Re: [LINUX-390] Linux system booted with old copy?
>>> On 10/4/2016 at 01:20 PM, Michael J Nash wrote:
> Greetings, I believe the command is grub2-install.
> There is also grub2-zipl-setup Note: This command is undocumented.
Neither of these commands will address the pr
>>> On 10/4/2016 at 01:20 PM, Michael J Nash wrote:
> Greetings, I believe the command is grub2-install.
> There is also grub2-zipl-setup Note: This command is undocumented.
Neither of these commands will address the problem Michael has experienced
since they don't change what persistent names
Greetings, I believe the command is grub2-install.
There is also grub2-zipl-setup Note: This command is undocumented.
From: Michael MacIsaac
To: LINUX-390@VM.MARIST.EDU
Date: 10/04/2016 12:58 PM
Subject:Re: Linux system booted with old copy?
Sent by:Linux on 390 Port
>>> On 10/4/2016 at 12:56 PM, Michael MacIsaac wrote:
> So how would you modify grub to use 'by-path' instead of
> 'by-id'?
You don't do it in grub2, you do it in dracut. Edit /etc/dracut.conf and add
the following line:
persistent_policy=by-path
Then re-run mkinitrd. The next time you reboo
Mark,
My colleague points out that by the time /etc/fstab is read, the root file
system is already mounted, so it's a bit of 'closing the barn door after
the horses get out'.
It used to be that the root file system was set in /etc/zipl.conf which you
could modify and run zipl to write to the boot
Mark,
Thanks, we will investigate.
-Mike
On Tue, Oct 4, 2016 at 12:17 PM, Mark Post wrote:
> >>> On 10/4/2016 at 08:30 AM, Michael MacIsaac
> wrote:
> > Thanks for the replies.
> >
> > I spoke with a colleague who suggested that btrfs may indeed be 'the
> > smoking gun'. Normally we use
>>> On 10/4/2016 at 08:30 AM, Michael MacIsaac wrote:
> Thanks for the replies.
>
> I spoke with a colleague who suggested that btrfs may indeed be 'the
> smoking gun'. Normally we use a 151 MR minidisk as the root file system.
> This virtual machine happened to also have an 1151, which probabl
Thanks for the replies.
I spoke with a colleague who suggested that btrfs may indeed be 'the
smoking gun'. Normally we use a 151 MR minidisk as the root file system.
This virtual machine happened to also have an 1151, which probably also has
a Linux system on it. He said btrfs could possibly cho
-tech diagnoses. 😷
Mike Walter
Sent from my Verizon, Samsung Galaxy smartphone
Original message
From: Mark Post Date: 10/3/16
3:36 PM (GMT-06:00) To: LINUX-390@VM.MARIST.EDU Subject:
Re: Linux system booted with old copy?
>>> On 10/3/2016 at 10:05 AM, Michael M
>>> On 10/3/2016 at 10:05 AM, Michael MacIsaac wrote:
> Recently we had a scheduled SHUTDOWN REIPL of a z/VM LPAR. After one of my
> Linux systems came back it was running with a copy of the OS and data that
> was 12 days old! (I saw the 12 day 'hole' in /var/log/messages). I have
> *never* seen
On Monday, 10/03/2016 at 02:09 GMT, Michael MacIsaac
wrote:
> Has anyone seen this or have a possible explanation? I cloned over it
and
> restored my data in order to get the system usable, so I don't have
access
> to the original system.
Would someone have dynamically linked to some new disk
Hello list,
Recently we had a scheduled SHUTDOWN REIPL of a z/VM LPAR. After one of my
Linux systems came back it was running with a copy of the OS and data that
was 12 days old! (I saw the 12 day 'hole' in /var/log/messages). I have
*never* seen this happen on zLinux nor Lintel.
It was a SLES 12
19 matches
Mail list logo
