Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Alan Altmark]

On Friday, 04/22/2016 at 02:41 GMT, Grzegorz Powiedziuk 
 wrote:
>
> 3. Less common, useful in some cases - OSA is plugged into "trunk" port 
on
> real switch and in general same as (2). But, when you do grant, you can 
say
> that this specific grant should act as "porttype trunk" (and you specify
> which vlans are trunked)  so VSWITCH instead of removing the vlan tag,
> forwards the whole thing to linux guest. So linux guest should be
> configured to receive and send tagged frames.  As Mark mentioned, during
> install process it might be troublesome.

Please don't do this.  Instead, use a PORTBASED VSWITCH so that you can 
two vNICs, each assigned to a different VLAN on the same VSWITCH.

Except for a VSWITCH sniffer, there's no longer any need for Linux to be 
VLAN-aware when running as a guest.

Alan Altmark

Senior Managing z/VM and Linux Consultant
Lab Services System z Delivery Practice
IBM Systems & Technology Group
ibm.com/systems/services/labservices
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Grzegorz Powiedziuk]

2016-04-21 16:29 GMT-04:00 Mark Post :

> >>> On 4/21/2016 at 03:38 PM, Grzegorz Powiedziuk 
> wrote:
> -snip-
> > I believe Mark said that having linux to handle vlan tagging is hard.
> >
> > But what you are trying to do is different. In your case, vswitch is
> > removing/adding vlan tags from/to  frames on the fly.
>
> If Grzegorz is correct in _his_ assumption, what he posted is a good
> description of what goes on and that _is_ something the installer can
> handle with no problems since from the guest's perspective it is not VLAN
> aware.
>
>
You're right. That is just assumption and after I read again the whole
thread I am not sure anymore  if this is what Dave wanted
So to give a full picture, you have three options ( simplified ):

1. Basic - Osa is plugged into "access" port on the real switch so there
are no vlan tags between real switch and OSA, and there are no vlan tags in
vswitch or linux. VSWITCH has to be set as "vlan unware".

2. Most common and useful - OSA is plugged into a "trunk" port on a real
switch. That port has to have vlan(s) added to it.  VSWITCH handles vlan
tags. So between vswitch (osa) and real switch the traffic is tagged with
vlan ids. Vswitch removes and adds vlan tags depending on direction.
Traffic between VSWITCH and linux guest is a untagged - standard network
frames. So from linux perspective it is like (1) . VSWITCH has to be set as
vlan AWARE and grants should have a vlan tag.

3. Less common, useful in some cases - OSA is plugged into "trunk" port on
real switch and in general same as (2). But, when you do grant, you can say
that this specific grant should act as "porttype trunk" (and you specify
which vlans are trunked)  so VSWITCH instead of removing the vlan tag,
forwards the whole thing to linux guest. So linux guest should be
configured to receive and send tagged frames.  As Mark mentioned, during
install process it might be troublesome.
In that case you might want to change it to (2) just for the install and
later configure vlan interface in the linux and redefine your grant to
porttype trunk afterwards.

Gregory

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Mark Post]

>>> On 4/21/2016 at 05:33 PM, Grzegorz Powiedziuk  
>>> wrote: 
> then pvcreate, vgextend, lvextend and filesystem resize.

No, we're trying to get him set up with LVM, just btrfs.


Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Grzegorz Powiedziuk]

I cant remember how it is during install but I would make a bet that
default partitioning will not use more than one drive. Did you try to go
into advanced partitioning and see what the installator sugests? You should
be able manualy create one big lvm with seperate partition for boot (or you
can include boot in lvm but not all distros/versions support this). Unless
you want to do btrfs (I usually stick to ext3/4) then as we agreed lvm
doesn't make sense and btrfs volume pools should be used instead. But I
haven't done this before so can't tell.

But nontheless you still should be able to change it now if you do it step
by step from Mark's instructions.
So first make sure that second dasd is activated and available after ipl
(yast)
then pvcreate, vgextend, lvextend and filesystem resize.
Gregory

On Thursday, April 21, 2016, Tom Huegel  wrote:

> I don't know.. I've done a couple of re-installs and I still don't
> see dasdb (251) being used.
> This is SLES12 SP1. I activate and format my two drives and then take all
> of the defaults.
>
> 蘿
>
> On Thu, Apr 21, 2016 at 11:50 AM, Tom Huegel  > wrote:
>
> > Since this is a new install I'll just go back to step one and start with
> > reinstalling..
> > I'll let you know how it works out.
> >
> > On Thu, Apr 21, 2016 at 11:08 AM, Robert J Brenneman  >
> > wrote:
> >
> >> On Thu, Apr 21, 2016 at 2:00 PM, Grzegorz Powiedziuk <
> >> gpowiedz...@gmail.com >
> >> wrote:
> >>
> >> > I see. I never used btrfs before so it makes sense.
> >> > So  isn't using LVM with btrfs on top of it complicating things
> >> more?
> >> >
> >>
> >> Basically yes.
> >>
> >> The default SLES 12 install does not use LVM though, it only uses the
> >> btrfs
> >> functions to achieve similar results.
> >>
> >> Doing both LVM and btrfs is not impossible, but you need to be
> thoughtful
> >> about how you combine those two.
> >>
> >> --
> >> Jay Brenneman
> >>
> >> --
> >> 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.marist.edu/htbin/wlvindex?LINUX-390
> >> --
> >> For more information on Linux on System z, visit
> >> http://wiki.linuxvm.org/
> >>
> >
> >
>
> --
> 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.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Mark Post]

>>> On 4/21/2016 at 05:03 PM, Tom Huegel  wrote: 
> I don't know.. I've done a couple of re-installs and I still don't
> see dasdb (251) being used.
> This is SLES12 SP1. I activate and format my two drives and then take all
> of the defaults.

That's because of the way the installer does things.  Once you have the install 
completed, use "yast dasd" to activate (and if needed format) the new volume. 
Then you'll need to use the btrfs command to add the new volume to the file 
system:
btrfs device add /dev/dasdb1 /

But, please, please, make sure you have all available maintenance installed 
before you do any of this, or make sure you have SLES12 *SP1* installed and 
updated with maintenance first.  Otherwise you're just going to wind up in the 
emergency shell again.


Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Mark Post]

>>> On 4/21/2016 at 04:38 PM, Alan Altmark  wrote: 
> On Thursday, 04/21/2016 at 07:39 GMT, Grzegorz Powiedziuk 
>  wrote:
>> I believe Mark said that having linux to handle vlan tagging is hard.
> 
> There is nothing difficult about getting Linux to perform VLAN tagging. 
> The question is not one of capability, but of network security management.

Perhaps on an already-installed system.  Working around the limitations of the 
SLE11 installer can indeed be very difficult.


Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Tom Huegel]

I don't know.. I've done a couple of re-installs and I still don't
see dasdb (251) being used.
This is SLES12 SP1. I activate and format my two drives and then take all
of the defaults.

蘿

On Thu, Apr 21, 2016 at 11:50 AM, Tom Huegel  wrote:

> Since this is a new install I'll just go back to step one and start with
> reinstalling..
> I'll let you know how it works out.
>
> On Thu, Apr 21, 2016 at 11:08 AM, Robert J Brenneman 
> wrote:
>
>> On Thu, Apr 21, 2016 at 2:00 PM, Grzegorz Powiedziuk <
>> gpowiedz...@gmail.com>
>> wrote:
>>
>> > I see. I never used btrfs before so it makes sense.
>> > So  isn't using LVM with btrfs on top of it complicating things
>> more?
>> >
>>
>> Basically yes.
>>
>> The default SLES 12 install does not use LVM though, it only uses the
>> btrfs
>> functions to achieve similar results.
>>
>> Doing both LVM and btrfs is not impossible, but you need to be thoughtful
>> about how you combine those two.
>>
>> --
>> Jay Brenneman
>>
>> --
>> 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.marist.edu/htbin/wlvindex?LINUX-390
>> --
>> For more information on Linux on System z, visit
>> http://wiki.linuxvm.org/
>>
>
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Alan Altmark]

On Thursday, 04/21/2016 at 07:39 GMT, Grzegorz Powiedziuk 
 wrote:
> I believe Mark said that having linux to handle vlan tagging is hard.

There is nothing difficult about getting Linux to perform VLAN tagging. 
The question is not one of capability, but of network security management.


> In your case, you let the vswitch to do vlan tagging and untaging. So
> vswitch has to be vlan aware so it can receive and understand tagged 
frames
> from the outside real, switch.
> When it receives a tagged frame from  the outside world it removes the 
vlan
> tag, and send the actual frame to guests which have  "granted" access to
> that speciffic vlan.  And when vswitch receives an untaged frame from
> guest, it adds a vlan tag to it and sends it to real switch (trunk port
> specifically) .
> So from linux perspective, there are no vlan tags and everything is nice 
an
> easy.

A nice explanation.

> define vswitch vlan aware vs vswitch .. vlan xxx I don't believe makes 
any
> difference in your case
> According to manual
> "defvid
> defines the virtual switch as a VLAN-aware switch supporting IEEE 
standard
> 802.lQ. The defvid defines the default VLAN ID to be assigned to guest
> ports when no VLAN ID is coded on the SET VSWITCH GRANT VLAN command"

Yes, there is a difference.  Please see my post.

Alan Altmark

Senior Managing z/VM and Linux Consultant
Lab Services System z Delivery Practice
IBM Systems & Technology Group
ibm.com/systems/services/labservices
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Mark Post]

>>> On 4/21/2016 at 03:38 PM, Grzegorz Powiedziuk  
>>> wrote: 
-snip-
> I believe Mark said that having linux to handle vlan tagging is hard.
> 
> But what you are trying to do is different. In your case, vswitch is
> removing/adding vlan tags from/to  frames on the fly.

If Grzegorz is correct in _his_ assumption, what he posted is a good 
description of what goes on and that _is_ something the installer can handle 
with no problems since from the guest's perspective it is not VLAN aware.


Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Grzegorz Powiedziuk]

2016-04-21 14:52 GMT-04:00 Dave Myers :

> Since Mark has told us that what we're attempting to do is not supported
> by the RAM system...I'll close this thread and thank you all for your
> feedback.
>
> I do have one last question for Alan.
>
> In this statement does the VLAN AWARE specify TRUNKING to the uplink OSA ?
> If not what is the difference between what we coded VLAN 229 and  VLAN
> AWARE?
>
> DEFINE VSWITCH VSW1  RDEV 400.P1  VLAN AWARE  NATIVE NONE
>
> Thanks,
> Dave
>
>
I believe Mark said that having linux to handle vlan tagging is hard.

But what you are trying to do is different. In your case, vswitch is
removing/adding vlan tags from/to  frames on the fly.

To do what Mark mentioned (and I doubt this is what you want to do) you
have to specify porttype trunk when you do the grant command. This will let
linux guest to see vlan tags. But I am pretty sure that you don't want to
do this.

In your case, you let the vswitch to do vlan tagging and untaging. So
vswitch has to be vlan aware so it can receive and understand tagged frames
from the outside real, switch.
When it receives a tagged frame from  the outside world it removes the vlan
tag, and send the actual frame to guests which have  "granted" access to
that speciffic vlan.  And when vswitch receives an untaged frame from
guest, it adds a vlan tag to it and sends it to real switch (trunk port
specifically) .
So from linux perspective, there are no vlan tags and everything is nice an
easy.

So what you are doing, should work perfectly fine as long the real switch
is set up correctly (port is set to trunk and your vlan was added to it)

When it comes to syntax

define vswitch vlan aware vs vswitch .. vlan xxx I don't believe makes any
difference in your case
According to manual
"defvid
defines the virtual switch as a VLAN-aware switch supporting IEEE standard
802.lQ. The defvid defines the default VLAN ID to be assigned to guest
ports when no VLAN ID is coded on the SET VSWITCH GRANT VLAN command"


Gregory

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

2016 VMWORKSHOP call for papers

[Philip Tully]

2016 VMWORKSHOP



The 2016 VMWorkshop is scheduled for June 23-25, 2016,it is being held at
Rutgers University, Busch Campus Center. 604 Bartholomew Rd,
Piscataway Township, NJ 08854.

This years VMWORKSHOP is the 6 since they were reanimated at Ohio
State in 2010.

 Call for papers is open.

One of the most valuable parts of the VMWORKSHOP are the sessions
given in small less formal settings with
access to the speakers.   Have you done something unique at your shop?
Written an interesting piece of code?
Integrated your z/VM, zVSE and/or Linux with other systems? Are you
willing to talk about these?

If you have never presented at the VMWORKSHOP before as an incentive
we will giving first time presenters
a $50 check*.

If you need help brushing up your presentation we have volunteers that
can help.
So submit your paper today here: Session submission
 and also Register for the
2016 VMWORKSHOP on your calendar, and submit your ideas for session
now.

Please go to VMWORKSHOP.ORG to register.

The cost of the conference remains at $100 for 2 ½ days which includes
a polo shirt and the Thursday

evening banquet.

Student attendance cost $10 (no Polo shirt ).


* (Max $50 pers session, check will be mailed to you within 4-6 weeks
after the workshop)

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Alan Altmark]

On Thursday, 04/21/2016 at 06:59 GMT, Dave Myers 
 wrote:
> I do have one last question for Alan.
>
> In this statement does the VLAN AWARE specify TRUNKING to the uplink OSA 
?
> If not what is the difference between what we coded VLAN 229 and  VLAN 
AWARE?
>
> DEFINE VSWITCH VSW1  RDEV 400.P1  VLAN AWARE  NATIVE NONE

Yes, VLAN AWARE specifies trunking on the uplink.  The difference is that 
you authorized everyone who was not explicitly authorized to a VLAN to use 
VLAN 229, and I required everyone to have an explicit GRANT.

By the way, if you have an ESM and the ESM returns an empty list of VLANs, 
your syntax would give the guest access to VLAN 229.  [This is fixed in 
6.4.]

I've been harping on the use of "VLAN AWARE NATIVE NONE" for all z/VM 6.2 
and later systems since ... z/VM 6.2.  And I expect (hope?) that there are 
going to be some changes in the future to disable the use of a VLAN ID on 
the DEFINE VSWITCH.

Alan Altmark

Senior Managing z/VM and Linux Consultant
Lab Services System z Delivery Practice
IBM Systems & Technology Group
ibm.com/systems/services/labservices
office: 607.429.3323
mobile; 607.321.7556
alan_altm...@us.ibm.com
IBM Endicott

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

[Dave Myers]

Since Mark has told us that what we're attempting to do is not supported by the 
RAM system...I'll close this thread and thank you all for your feedback.

I do have one last question for Alan.

In this statement does the VLAN AWARE specify TRUNKING to the uplink OSA ?
If not what is the difference between what we coded VLAN 229 and  VLAN AWARE?

DEFINE VSWITCH VSW1  RDEV 400.P1  VLAN AWARE  NATIVE NONE

Thanks,
Dave



From: Dave Myers
Sent: Wednesday, April 20, 2016 10:37 PM
To: 'linux-390@VM.MARIST.EDU' 
Subject: RE: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

Hi all,
I appreciate your help..please keep suggestions coming.
Marcy, the last post from you was not readable on the list.
Thanks,
Dave

From: Dave Myers
Sent: Wednesday, April 20, 2016 9:29 PM
To: 'linux-390@VM.MARIST.EDU' 
>
Subject: RE: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

VLAN def in system config is:
define vswitch vsw1 rdev 400.p1  vlan 229

In LNXDFLT profile we have:

COMMAND SET VSWITCH VSW1 GRANT   VLAN 229
COMMAND DEFINE NIC 300 TYPE QDIO
COMMAND COUPLE 300 TO SYSTEM VSW1

When we logon to the guest from which we are executing the SLES11SP4, we get 
positive msgs that the vnic is coupled to the vswitch.



[cid:image001.gif@01D14C4B.903B2B30]

Dave Myers  |  Senior Systems Engineer - System z
Office: 303.996.7125 | Mobile: 303.619.0782 |  
dave.my...@siriuscom.com
Sirius Computer Solutions | www.siriuscom.com
10100 Reunion Place, Suite 500, San Antonio, TX 78216



From: Dave Myers
Sent: Wednesday, April 20, 2016 8:56 PM
To: 'linux-390@VM.MARIST.EDU' 
>
Subject: RE: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

The define VSWITCH and the COUPLE both produce no errors.  That part looks good.
Yes. We are using this VLAN for some z/OS interfaces, so it is not new.

The SLES11SP4 EXEC does not show any errors.
The only message we get, after a 1-2 minute hang in the SLES11SP4 EXEC is:
" unable to find ... repository".

When I switch to a different VLAN that is not trunked and I keep the same NFS 
location of the repository is works OK and finds my NFS server.



From: Dave Myers
Sent: Wednesday, April 20, 2016 7:19 PM
To: 'linux-390@VM.MARIST.EDU' 
>
Subject: Install not working for SLES11 SP4 thru TRUNKED VSWITCH

I am trying to install SLES11 SP4 under zVM VSWITCH.
The VSWITCH is attached to an OSA that is TRUNKED to a switch port.
We defined VLAN ID on the VSWITCH def and also on the GRANT ACCESS statement 
for the guest.
We cannot connect when we run the SLES11SP4  EXEC.

Does anyone know if this is supported for install by the SLES11 SP4 RAM system?
Or does anyone know how to debug this connection from the SLES11SP4  EXEC ?

Thanks,
Dave


This message (including any attachments) is intended only for the use of the 
individual or entity to which it is addressed and may contain information that 
is non-public, proprietary, privileged, confidential, and exempt from 
disclosure under applicable law. If you are not the intended recipient, you are 
hereby notified that any use, dissemination, distribution, or copying of this 
communication is strictly prohibited. This message may be viewed by parties at 
Sirius Computer Solutions other than those named in the message header. This 
message does not contain an official representation of Sirius Computer 
Solutions. If you have received this communication in error, notify Sirius 
Computer Solutions immediately and (i) destroy this message if a facsimile or 
(ii) delete this message immediately if this is an electronic communication. 
Thank you.

Sirius Computer Solutions

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Tom Huegel]

Since this is a new install I'll just go back to step one and start with
reinstalling..
I'll let you know how it works out.

On Thu, Apr 21, 2016 at 11:08 AM, Robert J Brenneman 
wrote:

> On Thu, Apr 21, 2016 at 2:00 PM, Grzegorz Powiedziuk <
> gpowiedz...@gmail.com>
> wrote:
>
> > I see. I never used btrfs before so it makes sense.
> > So  isn't using LVM with btrfs on top of it complicating things more?
> >
>
> Basically yes.
>
> The default SLES 12 install does not use LVM though, it only uses the btrfs
> functions to achieve similar results.
>
> Doing both LVM and btrfs is not impossible, but you need to be thoughtful
> about how you combine those two.
>
> --
> Jay Brenneman
>
> --
> 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.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Robert J Brenneman]

On Thu, Apr 21, 2016 at 2:00 PM, Grzegorz Powiedziuk 
wrote:

> I see. I never used btrfs before so it makes sense.
> So  isn't using LVM with btrfs on top of it complicating things more?
>

Basically yes.

The default SLES 12 install does not use LVM though, it only uses the btrfs
functions to achieve similar results.

Doing both LVM and btrfs is not impossible, but you need to be thoughtful
about how you combine those two.

--
Jay Brenneman

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Grzegorz Powiedziuk]

2016-04-21 13:39 GMT-04:00 Tom Huegel :

> Oh well I must have done something wrong, it won't boot (IPL) now.
> Booting default
> (grub2)
>
>
>
> [  OK  ] Found device
> /dev/disk/by-uuid/f75ec84d-5640-4a2c-bced-e026dd7ec2b9.
>
> 
> Warning: /dev/disk/by-uuid/2103a597-f950-4b63-931e-675d2c52dbd9 does not
> exist
>
>
>
It seems like the device is still missing. Did you run yast or
dasd_configure before doing all that and did it update initrd to have the
device available at boot time?
In dracut try a couple of basic commands like lsdasd  or cat
/proc/dasd/devices  to see if the second disk is actually there (probably
not)

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Mark Post]

 >>> On 4/21/2016 at 01:39 PM, Tom Huegel  wrote: 
> Starting Dracut Emergency
> Shell...
> 
> Warning: /dev/disk/by-uuid/2103a597-f950-4b63-931e-675d2c52dbd9 does not
> exist

This looks like the problem I warned you about previously.  When you're in the 
emergency shell, can you see both DASD volumes in /dev?
ls /dev/dasd*

If it is not there, then
echo 1 > /sys/bus/ccw/devices/0.0.0251/online

In some cases, the DASD volume would come online just after entering the 
emergency shell, and you could then see it:
ls /dev/disk/by-uuid/

If /dev/disk/by-uuid/2103a597-f950-4b63-931e-675d2c52dbd9 is there, you should 
be able to continue by doing a "systemctl default" command.

Once the system is up run mkinitrd and grub2-install.


Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Grzegorz Powiedziuk]

2016-04-21 13:27 GMT-04:00 Robert J Brenneman :

>
> That's part of what the new btrfs in SLES 12 does for you. It does
> management of multiple physical block devices as a single logical entity,
> as well as talking that single logical entity and using it to back multiple
> mount points ( now called subvolumes ? ) to do things like snapshot just
> the /home section so you can do a rollback if you need to.
>
> It's part of the new SLES support to restore off service if required.
>
>
I see. I never used btrfs before so it makes sense.
So  isn't using LVM with btrfs on top of it complicating things more?

Gregory

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Mark Post]

>>> On 4/21/2016 at 12:04 PM, Tom Huegel  wrote: 
> Now I need just a little more help.
> Would this be correct to use all of the new volume? " lvextend -l +100%PVS
> /dev/system "

Almost, but not quite.  You would need to specify the logical volume name as 
/dev/system/root, not just /dev/system.

Perhaps an easier way would be
lvextend -l +1761 /dev/system/root

since that is the number of free physical extents you have as shown in the 
vgdisplay command.

After that, you have a couple more things to do, one of which is to actually 
resize the file system
btrfs filesystem resize max /

Having said all that, perhaps the better route is to re-install the system:
1. using SLES12 SP1 instead of just SLES12
2. not using LVM for volume management since btrfs can handle it by itself 
(SUSE recommends against using btrfs on top of LVM). I would have to check but 
btrfs on LVM might not be a supported configuration.

If you decide not to do that, make sure you have all available maintenance 
installed to avoid the problem I mentioned earlier with new DASD volumes not 
being activated properly at boot time.

Mark Post

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Tom Huegel]

Oh well I must have done something wrong, it won't boot (IPL) now.
Booting default
(grub2)

[  OK  ] Started Show Plymouth Boot
Screen.

[  OK  ] Reached target
Paths.

[  OK  ] Reached target Basic
System.

[  OK  ] Found device
/dev/disk/by-uuid/f75ec84d-5640-4a2c-bced-e026dd7ec2b9.

[  OK  ] Found device
/dev/mapper/system-swap.

Buffer I/O error on device dm-2, logical block
3172336

Buffer I/O error on device dm-2, logical block
3172336

[  OK  ] Started Show Plymouth Boot
Screen.

[  OK  ] Reached target
Paths.

[  OK  ] Reached target Basic
System.

[  OK  ] Found device
/dev/disk/by-uuid/f75ec84d-5640-4a2c-bced-e026dd7ec2b9.

[  OK  ] Found device
/dev/mapper/system-swap.

 Starting Dracut Emergency
Shell...

Warning: /dev/disk/by-uuid/2103a597-f950-4b63-931e-675d2c52dbd9 does not
exist


Generating
"/run/initramfs/rdsosreport.txt"

Buffer I/O error on device dm-2, logical block
3172336

Buffer I/O error on device dm-2, logical block
3172336

Buffer I/O error on device dm-2, logical block
3172336

Buffer I/O error on device dm-2, logical block
3172336





Entering emergency mode. Exit the shell to
continue.

Type "journalctl" to view system
logs.

You might want to save "/run/initramfs/rdsosreport.txt" to a USB stick or
/boot
after mounting them and attach it to a bug
report.





dracut:/#
1




HOLDING   SCZVMLX2

On Thu, Apr 21, 2016 at 10:27 AM, Robert J Brenneman 
wrote:

> On Thu, Apr 21, 2016 at 12:27 PM, Grzegorz Powiedziuk <
> gpowiedz...@gmail.com
> > wrote:
>
> > /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /opt
> > /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /home
> > /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /boot/grub2/s390x-emu
> >
> >
> > According to this,  the same logical volume is mounted multiple times.
> > Honestly it doesn't make sense to me.
> >
>
> That's part of what the new btrfs in SLES 12 does for you. It does
> management of multiple physical block devices as a single logical entity,
> as well as talking that single logical entity and using it to back multiple
> mount points ( now called subvolumes ? ) to do things like snapshot just
> the /home section so you can do a rollback if you need to.
>
> It's part of the new SLES support to restore off service if required.
>
> --
> Jay Brenneman
>
> --
> 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.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Robert J Brenneman]

On Thu, Apr 21, 2016 at 12:27 PM, Grzegorz Powiedziuk  wrote:

> /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /opt
> /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /home
> /dev/mapper/system-root  5.3G  3.0G  2.0G  60% /boot/grub2/s390x-emu
>
>
> According to this,  the same logical volume is mounted multiple times.
> Honestly it doesn't make sense to me.
>

That's part of what the new btrfs in SLES 12 does for you. It does
management of multiple physical block devices as a single logical entity,
as well as talking that single logical entity and using it to back multiple
mount points ( now called subvolumes ? ) to do things like snapshot just
the /home section so you can do a rollback if you need to.

It's part of the new SLES support to restore off service if required.

--
Jay Brenneman

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Grzegorz Powiedziuk]

I am not sure about "PVS"  but "FREE" will work
You want to extend the LV so it should be rather something like this:

lvextend -l +100%FREE /dev/system/root(or /dev/mapper/system-root)

(/dev/system is a Volume group which from I can see has been already
extended - it has 7GB free )

Of course it that is what you want to do? Extend the size of root logical
volume?
Next you will have to resize the filesystem. If it is ext3 or ext4 you can
just do resize2fs /dev/system/root   and it will use all free space.


But, what I don't like, is the output of your "df" command. Nobody said
anything about it before so I guess it is ok, but I've never seen something
like this:

/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/tmp
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/spool
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/opt
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/log
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/lib/pgsql
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/lib/mailman
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/crash
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /usr/local
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /var/lib/named
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /tmp
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /srv
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /opt
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /home
/dev/mapper/system-root  5.3G  3.0G  2.0G  60% /boot/grub2/s390x-emu


According to this,  the same logical volume is mounted multiple times.
Honestly it doesn't make sense to me.
There should be just one mount point if you have only one logical volume
"root" without separate volumes for other mount points.
/dev/mapper/system-root  xxx xxx xxx 60% /

And that's it.


Gregory

2016-04-21 12:04 GMT-04:00 Tom Huegel :

> Now I need just a little more help.
> Would this be correct to use all of the new volume? " lvextend -l +100%PVS
> /dev/system "
> Thanks
>
> sles12:~ # vgdisplay
>   --- Volume group ---
>   VG Name   system
>   System ID
>   Formatlvm2
>   Metadata Areas2
>   Metadata Sequence No  4
>   VG Access read/write
>   VG Status resizable
>   MAX LV0
>   Cur LV2
>   Open LV   2
>   Max PV0
>   Cur PV2
>   Act PV2
>   VG Size   13.55 GiB
>   PE Size   4.00 MiB
>   Total PE  3470
>   Alloc PE / Size   1709 / 6.68 GiB
>   Free  PE / Size   1761 / 6.88 GiB
>   VG UUID   8wdm3K-4TNN-yqqn-v4bG-vBFE-INMc-AIZCxT
>
>
> On Wed, Apr 20, 2016 at 8:06 AM, Mark Post  wrote:
>
> > >>> On 4/19/2016 at 04:42 PM, Michael J Nash 
> wrote:
> > > Greeting Mark, please tell us why  use_lvmetad  has been disabled!
> >
> > It wasn't considered ready for enterprise use.
> >
> >
> > Mark Post
> >
> > --
> > 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.marist.edu/htbin/wlvindex?LINUX-390
> > --
> > For more information on Linux on System z, visit
> > http://wiki.linuxvm.org/
> >
>
> --
> 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.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Simple DASD question

[Tom Huegel]

Now I need just a little more help.
Would this be correct to use all of the new volume? " lvextend -l +100%PVS
/dev/system "
Thanks

sles12:~ # vgdisplay
  --- Volume group ---
  VG Name   system
  System ID
  Formatlvm2
  Metadata Areas2
  Metadata Sequence No  4
  VG Access read/write
  VG Status resizable
  MAX LV0
  Cur LV2
  Open LV   2
  Max PV0
  Cur PV2
  Act PV2
  VG Size   13.55 GiB
  PE Size   4.00 MiB
  Total PE  3470
  Alloc PE / Size   1709 / 6.68 GiB
  Free  PE / Size   1761 / 6.88 GiB
  VG UUID   8wdm3K-4TNN-yqqn-v4bG-vBFE-INMc-AIZCxT


On Wed, Apr 20, 2016 at 8:06 AM, Mark Post  wrote:

> >>> On 4/19/2016 at 04:42 PM, Michael J Nash  wrote:
> > Greeting Mark, please tell us why  use_lvmetad  has been disabled!
>
> It wasn't considered ready for enterprise use.
>
>
> Mark Post
>
> --
> 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.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
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.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/