Re: Buffer I/O error on device dm-28

2018-01-29 Thread Csaba Polgar 

Hello,

I got help, and found the solution.

The user direct was incorrect (on z/VM level). This line: IPL SUSE11DB PARM
DASD=300-3FF
   The correct line:IPL SUSE11DB PARM DASD=300-5FF

The DASD range was not extended.  This is working now!

About the root cause: Before the extension of LVM and creating of the new
LV and FS, the chccwdev was enough. But after and during the first boot,
the lvm was not able to find the necessary PV and building up the LV and
FS, so the filesystems was harmed or missed.

But I got another idea also:
use dasd_configure to configure the new dasds:
Example:
dasd_configure 0.0.0400 1 0
dasd_configure 0.0.0401 1 0
dasd_configure 0.0.0402 1 0
...
This command should bring the dasd online and create a udev rule to make it
persistent at ipl time.

As I see now, this second solution would be good also (it was not tested,
because of the server is very sensitive and in production).

About the affected system:
myhostname:~ # cat /etc/SuSE-release
SUSE Linux Enterprise Server 11 (s390x)
VERSION = 11
PATCHLEVEL = 4
myhostname:~ #



Kind Regards :
Csaba Polgar



From:   Mark Post <mp...@suse.com>
To: LINUX-390@VM.MARIST.EDU
Date:   2018.01.29 19:49
Subject:    Re: Buffer I/O error on device dm-28
Sent by:Linux on 390 Port <LINUX-390@VM.MARIST.EDU>



>>> On 1/28/2018 at 09:51 PM, Csaba Polgar <csaba_pol...@hu.ibm.com> wrote:

> Hello,
>
> I'm newbie on this forum, but I would like to ask your help. (Sorry, if
I
> brake any protocol/usage guideline.)
>   Could someone please help to solve the below issue?
>
-snip-
>
> How can it be online (active) before the Linux boot? Or what is missing
> from the configuration?

There's not nearly enough information to know what's going on, yet.  First,
what distribution is this?  What maintenance level?

Next, is this an LVM setup, or something else?  If LVM, then some output
from things like "vgs" "pvs" "vgdisplay -v" and so on would be helpful.


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
https://urldefense.proofpoint.com/v2/url?u=http-3A__www.marist.edu_htbin_wlvindex-3FLINUX-2D390=DwIFAg=jf_iaSHvJObTbx-siA1ZOg=aJQcehgOpJiq_Sot7u0bUO6bkzudWsGKwcorj1dj0KU=B4ffK2DG20KGfvVKnTWIeWwImYtx01KsKUkvKECsPmM=Twr2nuIkwZEVnNUlBuAWhk98iRob8yuoijQMCxCJvj8=

--
For more information on Linux on System z, visit
https://urldefense.proofpoint.com/v2/url?u=http-3A__wiki.linuxvm.org_=DwIFAg=jf_iaSHvJObTbx-siA1ZOg=aJQcehgOpJiq_Sot7u0bUO6bkzudWsGKwcorj1dj0KU=B4ffK2DG20KGfvVKnTWIeWwImYtx01KsKUkvKECsPmM=3MACH-NAJUwcTjztiD03fXY7FEvwu1Mdp3CTNqbZTxQ=





--
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: Buffer I/O error on device dm-28

2018-01-29 Thread Grzegorz Powiedziuk 
2018-01-29 13:47 GMT-05:00 Mark Post :

> >>> On 1/28/2018 at 09:51 PM, Csaba Polgar 
> wrote:
> > Hello,
> >
> > I'm newbie on this forum, but I would like to ask your help. (Sorry, if
> I
> > brake any protocol/usage guideline.)
> >   Could someone please help to solve the below issue?
> >
> -snip-
> >
> > How can it be online (active) before the Linux boot? Or what is missing
> > from the configuration?
>
> There's not nearly enough information to know what's going on, yet.
> First, what distribution is this?  What maintenance level?
>
> Next, is this an LVM setup, or something else?  If LVM, then some output
> from things like "vgs" "pvs" "vgdisplay -v" and so on would be helpful.
>

Probably that's an LVM with many dasds as PVs (judging by "dm-28" device
name) and some of them haven't been online-d in a permanent way.

If this is suse then as far as I remember just activate them using yast and
on exit it should save it. There also was another way. Just run
"dasd_configure 0.0.0400 1" and it  should do the trick. I don't think you
need to run mkinitrd unless this this device is a part of a root filesystem
but probably it is not otherwise you wouldn't complete the boot at all.

If this is redhat then you can add these to dasd.conf for example:

cat /etc/dasd.conf
0.0.0200
0.0.0201
0.0.0210

If this a really old redhat you might need to add these to /etc/zipl.com
for example like this(in redhat 6 and 7 you can either way) :

[root@lin00 ~]# cat /etc/zipl.conf
[defaultboot]
defaultauto
prompt=1
timeout=5
default=linux
target=/boot
[linux]
image=/boot/vmlinuz-3.10.0-123.el7.s390x
parameters="cio_ignore=all,!condev root=/dev/mapper/rhel_lin00-root
crashkernel=auto  rd.lvm.lv=rhel_lin00/root vconsole.keymap=us
vconsole.font=latarcyrheb-sun16 rd.dasd=0.0.0200 rd.dasd=0.0.0201
rd.dasd=0.0.0210 rd.dasd=0.0.0211 rd.dasd=0.0.0192 rd.lvm.lv=rhel_lin00/boot
LANG=en_US.UTF-8"
ramdisk=/boot/initramfs-3.10.0-123.el7.s390x.img


see above list of rd.dasd=  with addresses? After adding these you need to
run zipl to save it in the ipl record  (you can do zipl --dry-run -V to
verify that it does what it should)

Regards
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: Buffer I/O error on device dm-28

2018-01-29 Thread Mark Post 
>>> On 1/28/2018 at 09:51 PM, Csaba Polgar  wrote: 
> Hello,
> 
> I'm newbie on this forum, but I would like to ask your help. (Sorry, if  I
> brake any protocol/usage guideline.)
>   Could someone please help to solve the below issue?
> 
-snip-
> 
> How can it be online (active) before the Linux boot? Or what is missing
> from the configuration?

There's not nearly enough information to know what's going on, yet.  First, 
what distribution is this?  What maintenance level?

Next, is this an LVM setup, or something else?  If LVM, then some output from 
things like "vgs" "pvs" "vgdisplay -v" and so on would be helpful.


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/

Buffer I/O error on device dm-28

2018-01-28 Thread Csaba Polgar 
Hello,

I'm newbie on this forum, but I would like to ask your help. (Sorry, if  I
brake any protocol/usage guideline.)
  Could someone please help to solve the below issue?


We have created a big filesystems (with 2 TB) 1 and half month ago.
And there was no error message or any complain from the customer.

After a new boot, I got the below error message:
Jan 28 20:16:50 myhostname kernel: Buffer I/O error on device dm-28,
logical block 536870896
Jan 28 20:16:50 myhostname last message repeated 2 times
Jan 28 20:17:07 myhostname kernel: Buffer I/O error on device dm-28,
logical block 536870896
Jan 28 20:17:12 myhostname last message repeated 6 times

from dmesg:
...
Buffer I/O error on device dm-28, logical block 536870896
Buffer I/O error on device dm-28, logical block 536870896
Buffer I/O error on device dm-28, logical block 536870896
Buffer I/O error on device dm-28, logical block 536870896
...

Most of dasd were online, but I saw more offline dasd:
myhostname:~ # lsdasd -a
Bus-ID Status  Name  Device  Type  BlkSz  Size  Blocks
==
0.0.0400   offline
0.0.0401   offline
0.0.0402   offline
0.0.0403   offline
0.0.0404   offline
0.0.0405   offline
0.0.0406   offline
0.0.0407   offline
...

How can it be online (active) before the Linux boot? Or what is missing
from the configuration?

Kind Regards :
Csaba Polgar

--
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/