Re: DASD assignment different between SLES10 and SLES11
Added ?dasd_mod.dasd=0.0.150-0.0.15f? to the zipl.conf and ran zipl. These caused massive I/O errors and I had to back out. I analyzed this problem further and realized that I had two virtual DASD addresses mapped to this z/VM guest, leftover from a previous Reiserfs to EXT3 conversion. I had taken them off with the dasd_configure command months previouz. In other words, these I/O errors were ok and made now made sense. I updated the z/VM guest?s configuration and removed these DASD devices. I then added updated zipl.conf as follows: [ipl] image = /boot/image target = /boot/zipl ramdisk = /boot/initrd,0x200 parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 dasd_mod.dasd=0.0.0150-0.0.015f TERM=dumb I ran zipl and rebooted. Now everything is in its normal predictable slot as follows: linuxm02:~ lsdasd Bus-ID Status Name Device Type BlkSz Size Blocks == 0.0.0150 active dasda 94:0ECKD 4096 2347MB600840 0.0.0151 active dasdb 94:4ECKD 4096 2347MB600840 0.0.0152 active dasdc 94:8ECKD 4096 2347MB600840 0.0.0153 active dasdd 94:12 ECKD 4096 2347MB600840 0.0.0154 active dasde 94:16 ECKD 4096 2347MB600840 0.0.015f active dasdp 94:60 ECKD 4096 589MB 150840 Thanks to everyone that responded. Enjoy your weekend. Peter From: Ronald van der Laan nl50...@gmail.com To: LINUX-390@vm.marist.edu Date: 09/29/2011 04:26 PM Subject:Re: DASD assignment different between SLES10 and SLES11 Sent by:Linux on 390 Port LINUX-390@vm.marist.edu Peter, Have you tried: parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 dasd=0.0.0150-0.0.015f TERM=dumb Ronald van der Laan -- 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/ This Email message and any attachment may contain information that is proprietary, legally privileged, confidential and/or subject to copyright belonging to Pepco Holdings, Inc. or its affiliates (PHI). This Email is intended solely for the use of the person(s) to which it is addressed. If you are not an intended recipient, or the employee or agent responsible for delivery of this Email to the intended recipient(s), you are hereby notified that any dissemination, distribution or copying of this Email is strictly prohibited. If you have received this message in error, please immediately notify the sender and permanently delete this Email and any copies. PHI policies expressly prohibit employees from making defamatory or offensive statements and infringing any copyright or any other legal right by Email communication. PHI will not accept any liability in respect of such communications. -- 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/
DASD assignment different between SLES10 and SLES11
Under SLES10, we see the following: 0.0.0150(ECKD) at ( 94: 0) is dasda : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0151(ECKD) at ( 94: 4) is dasdb : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0152(ECKD) at ( 94: 8) is dasdc : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0153(ECKD) at ( 94: 12) is dasdd : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0154(ECKD) at ( 94: 16) is dasde : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.015f(ECKD) at ( 94: 60) is dasdp : active at blocksize 4096, 150840 blocks, 589 MB Notice that 0.0.015f is dasdp Under SLES11, we see the following: Bus-ID Status Name Device Type BlkSz Size Blocks == 0.0.0150 active dasda 94:0ECKD 4096 2347MB600840 0.0.0151 active dasdb 94:4ECKD 4096 2347MB600840 0.0.0152 active dasdc 94:8ECKD 4096 2347MB600840 0.0.0155 active dasdd 94:12 ECKD 4096 2347MB600840 0.0.0156 active dasde 94:16 ECKD 4096 2347MB600840 0.0.015f active dasdf 94:20 ECKD 4096 589MB 150840 Notice that 0.0.015f is dasdf rather than dasdp. How can I insure that 0.0.015f is always dasdp? I read that this behavior had change but naturally now that I need this doc, I cannot find it anywhere. I do not remember where I read it. E. Can anyone point me in the right direction? Thanks as always. Peter This Email message and any attachment may contain information that is proprietary, legally privileged, confidential and/or subject to copyright belonging to Pepco Holdings, Inc. or its affiliates (PHI). This Email is intended solely for the use of the person(s) to which it is addressed. If you are not an intended recipient, or the employee or agent responsible for delivery of this Email to the intended recipient(s), you are hereby notified that any dissemination, distribution or copying of this Email is strictly prohibited. If you have received this message in error, please immediately notify the sender and permanently delete this Email and any copies. PHI policies expressly prohibit employees from making defamatory or offensive statements and infringing any copyright or any other legal right by Email communication. PHI will not accept any liability in respect of such communications. -- 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: DASD assignment different between SLES10 and SLES11
Is this for a program or fstab entry which is expecting that disk to always
appear at /dev/dasdp?
Because if you can edit your source, it would be safer to refer to your
disks by path (such as /dev/disk/by-path/ccw-0.0.0154) rather than by their
dasd[a-z][a-z]* names. The path will always correspond to the underlying
virtual device number, while the by-letter names are dependent on the order
in which they happen to be brought online.
Also, if you need to find the associated dasd* name for a disk whose vdev
you know, you can use the following:
readlink -f /dev/disk/by-path/ccw-0.0.${vdev}
...like this:
myhostname:/etc/udev/rules.d # readlink -f /dev/disk/by-path/ccw-0.0.0100
/dev/dasda
On Thu, Sep 29, 2011 at 10:59 AM, Peter E. Abresch Jr. - at Pepco
peabre...@pepco.com wrote:
Under SLES10, we see the following:
0.0.0150(ECKD) at ( 94: 0) is dasda : active at blocksize 4096,
600840 blocks, 2347 MB
0.0.0151(ECKD) at ( 94: 4) is dasdb : active at blocksize 4096,
600840 blocks, 2347 MB
0.0.0152(ECKD) at ( 94: 8) is dasdc : active at blocksize 4096,
600840 blocks, 2347 MB
0.0.0153(ECKD) at ( 94: 12) is dasdd : active at blocksize 4096,
600840 blocks, 2347 MB
0.0.0154(ECKD) at ( 94: 16) is dasde : active at blocksize 4096,
600840 blocks, 2347 MB
0.0.015f(ECKD) at ( 94: 60) is dasdp : active at blocksize 4096,
150840 blocks, 589 MB
Notice that 0.0.015f is dasdp
Under SLES11, we see the following:
Bus-ID Status Name Device Type BlkSz Size Blocks
==
0.0.0150 active dasda 94:0ECKD 4096 2347MB600840
0.0.0151 active dasdb 94:4ECKD 4096 2347MB600840
0.0.0152 active dasdc 94:8ECKD 4096 2347MB600840
0.0.0155 active dasdd 94:12 ECKD 4096 2347MB600840
0.0.0156 active dasde 94:16 ECKD 4096 2347MB600840
0.0.015f active dasdf 94:20 ECKD 4096 589MB 150840
Notice that 0.0.015f is dasdf rather than dasdp. How can I insure that
0.0.015f is always dasdp?
I read that this behavior had change but naturally now that I need this
doc, I cannot find it anywhere. I do not remember where I read it. E.
Can anyone point me in the right direction? Thanks as always.
Peter
This Email message and any attachment may contain information that is
proprietary, legally privileged, confidential and/or subject to copyright
belonging to Pepco Holdings, Inc. or its affiliates (PHI). This Email is
intended solely for the use of the person(s) to which it is addressed. If
you are not an intended recipient, or the employee or agent responsible for
delivery of this Email to the intended recipient(s), you are hereby
notified
that any dissemination, distribution or copying of this Email is strictly
prohibited. If you have received this message in error, please immediately
notify the sender and permanently delete this Email and any copies. PHI
policies expressly prohibit employees from making defamatory or offensive
statements and infringing any copyright or any other legal right by Email
communication. PHI will not accept any liability in respect of such
communications.
--
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: DASD assignment different between SLES10 and SLES11
Peter, Actually we observed this behavior as far back as SLES10 SP3. We started using persistent device names and forgot about it. The documentation for this is in the release notes. From the release notes: Installation using Persistent Device names If you plan to add additional storage devices to your system after the OS installation, we strongly recommend to use persistent device names for all storage devices during installation. The installer by default uses the kernel device names. There is additional information in the SLES11 device driver manual on the developerworks site. From the Device Drivers, Features, and Commands manual SC34-2595-01 (from page 35) autodetect causes the DASD device driver to allocate device names and the corresponding minor numbers to all DASD devices and set them online during the boot process. See “DASD naming scheme” on page 31 for the naming scheme. The device names are assigned in order of ascending subchannel numbers. Auto-detection can yield confusing results if you change your I/O configuration and reboot, or if you are running as a guest operating system in VM because the devices might appear with different names and minor numbers after rebooting. What may have changed is how dasd= is specified and/or how it works. It is different between the SLES10 and SLES11 manuals. Ron Foster From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E. Abresch Jr. - at Pepco [peabre...@pepco.com] Sent: Thursday, September 29, 2011 9:59 AM To: LINUX-390@VM.MARIST.EDU Subject: DASD assignment different between SLES10 and SLES11 Under SLES10, we see the following: 0.0.0150(ECKD) at ( 94: 0) is dasda : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0151(ECKD) at ( 94: 4) is dasdb : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0152(ECKD) at ( 94: 8) is dasdc : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0153(ECKD) at ( 94: 12) is dasdd : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0154(ECKD) at ( 94: 16) is dasde : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.015f(ECKD) at ( 94: 60) is dasdp : active at blocksize 4096, 150840 blocks, 589 MB Notice that 0.0.015f is dasdp Under SLES11, we see the following: Bus-ID Status Name Device Type BlkSz Size Blocks == 0.0.0150 active dasda 94:0ECKD 4096 2347MB600840 0.0.0151 active dasdb 94:4ECKD 4096 2347MB600840 0.0.0152 active dasdc 94:8ECKD 4096 2347MB600840 0.0.0155 active dasdd 94:12 ECKD 4096 2347MB600840 0.0.0156 active dasde 94:16 ECKD 4096 2347MB600840 0.0.015f active dasdf 94:20 ECKD 4096 589MB 150840 Notice that 0.0.015f is dasdf rather than dasdp. How can I insure that 0.0.015f is always dasdp? I read that this behavior had change but naturally now that I need this doc, I cannot find it anywhere. I do not remember where I read it. E. Can anyone point me in the right direction? Thanks as always. Peter This Email message and any attachment may contain information that is proprietary, legally privileged, confidential and/or subject to copyright belonging to Pepco Holdings, Inc. or its affiliates (PHI). This Email is intended solely for the use of the person(s) to which it is addressed. If you are not an intended recipient, or the employee or agent responsible for delivery of this Email to the intended recipient(s), you are hereby notified that any dissemination, distribution or copying of this Email is strictly prohibited. If you have received this message in error, please immediately notify the sender and permanently delete this Email and any copies. PHI policies expressly prohibit employees from making defamatory or offensive statements and infringing any copyright or any other legal right by Email communication. PHI will not accept any liability in respect of such communications. -- 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: DASD assignment different between SLES10 and SLES11
Yes, our root is always dasda and our swap is always dasdp. This is what our original automated Linux cloning process expects and we wish to not re-engineer our cloning process at this time. Under SLES10, in /etc/zipl.conf we have the following: parameters = root=/dev/dasda1 dasd=150-15f TERM=dumb Under SLES11, in /etc/zipl.conf we have the following: parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 TERM=dumb I am thinking this is where I need to make my change, but my own memory is slow to swap back in for me to remember all of it. But . . . readlink -f /dev/disk/by-path/ccw-0.0.$15F reveals /dev/disk/by-path/ccw-0.0.5F - it seems to strip the 1 from the 15f. In my SLES10 cloning script, where I have the following: echo Creating DASD swap space on /dev/dasdp1 swapoff -a echo Formatting DASD swap space on /dev/dasdp1 dasdfmt -vy -f /dev/dasdp -d cdl -l swap01 -b 4096 echo Partitioning DASD swap space on /dev/dasdp1 fdasd -l swap01 -a /dev/dasdp echo Making DASD swap space on /dev/dasdp1 sleep 5 mkswap /dev/dasdp1 echo Activating DASD swap space on /dev/dasdp1 swapon -a echo Verifying DASD swap space on /dev/dasdp1 swapon ?s I can change in my SLES11 cloning script to: echo Creating DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapoff -a echo Formatting DASD swap space on /dev/disk/by-path/ccw-0.0.5F dasdfmt -vy -f /dev/disk/by-path/ccw-0.0.5F -d cdl -l swap01 -b 4096 echo Partitioning DASD swap space on /dev/disk/by-path/ccw-0.0.5F fdasd -l swap01 -a /dev/disk/by-path/ccw-0.0.5F echo Making DASD swap space on /dev/disk/by-path/ccw-0.0.5F sleep 5 mkswap /dev/disk/by-path/ccw-0.0.5F echo Activating DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapon -a echo Verifying DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapon ?s Is this what I should do? I am hesitant as there might me other scripted mounts that do /dev/dasdx# . Should I eat the bullet now or defer? Peter From: Ron Foster at Baldor-IS rfos...@baldor.com To: LINUX-390@vm.marist.edu Date: 09/29/2011 11:38 AM Subject:Re: DASD assignment different between SLES10 and SLES11 Sent by:Linux on 390 Port LINUX-390@vm.marist.edu Peter, Actually we observed this behavior as far back as SLES10 SP3. We started using persistent device names and forgot about it. The documentation for this is in the release notes. From the release notes: Installation using Persistent Device names If you plan to add additional storage devices to your system after the OS installation, we strongly recommend to use persistent device names for all storage devices during installation. The installer by default uses the kernel device names. There is additional information in the SLES11 device driver manual on the developerworks site. From the Device Drivers, Features, and Commands manual SC34-2595-01 (from page 35) autodetect causes the DASD device driver to allocate device names and the corresponding minor numbers to all DASD devices and set them online during the boot process. See ?DASD naming scheme? on page 31 for the naming scheme. The device names are assigned in order of ascending subchannel numbers. Auto-detection can yield confusing results if you change your I/O configuration and reboot, or if you are running as a guest operating system in VM because the devices might appear with different names and minor numbers after rebooting. What may have changed is how dasd= is specified and/or how it works. It is different between the SLES10 and SLES11 manuals. Ron Foster From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E. Abresch Jr. - at Pepco [peabre...@pepco.com] Sent: Thursday, September 29, 2011 9:59 AM To: LINUX-390@VM.MARIST.EDU Subject: DASD assignment different between SLES10 and SLES11 Under SLES10, we see the following: 0.0.0150(ECKD) at ( 94: 0) is dasda : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0151(ECKD) at ( 94: 4) is dasdb : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0152(ECKD) at ( 94: 8) is dasdc : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0153(ECKD) at ( 94: 12) is dasdd : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0154(ECKD) at ( 94: 16) is dasde : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.015f(ECKD) at ( 94: 60) is dasdp : active at blocksize 4096, 150840 blocks, 589 MB Notice that 0.0.015f is dasdp Under SLES11, we see the following: Bus-ID Status Name Device Type BlkSz Size Blocks == 0.0.0150 active dasda 94:0ECKD 4096 2347MB600840 0.0.0151 active dasdb 94:4ECKD 4096 2347MB600840 0.0.0152 active dasdc 94:8ECKD 4096 2347MB600840 0.0.0155 active dasdd 94:12 ECKD 4096 2347MB600840 0.0.0156 active dasde 94:16 ECKD
Re: DASD assignment different between SLES10 and SLES11
Peter, We do not have SLES11, but from the manual, on page 21 (Module parameters on the kernel parameter line), it appears that the appropriate format of the dasd= parameter might be: dasd_mod.dasd=0.0.150-0.0.15f so your parameter line would be: parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 dasd_mod.dasd=0.0.150-0.0.15f TERM=dumb Ron Foster From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E. Abresch Jr. - at Pepco [peabre...@pepco.com] Sent: Thursday, September 29, 2011 11:09 AM To: LINUX-390@VM.MARIST.EDU Subject: Re: DASD assignment different between SLES10 and SLES11 Yes, our root is always dasda and our swap is always dasdp. This is what our original automated Linux cloning process expects and we wish to not re-engineer our cloning process at this time. Under SLES10, in /etc/zipl.conf we have the following: parameters = root=/dev/dasda1 dasd=150-15f TERM=dumb Under SLES11, in /etc/zipl.conf we have the following: parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 TERM=dumb I am thinking this is where I need to make my change, but my own memory is slow to swap back in for me to remember all of it. But . . . readlink -f /dev/disk/by-path/ccw-0.0.$15F reveals /dev/disk/by-path/ccw-0.0.5F - it seems to strip the 1 from the 15f. In my SLES10 cloning script, where I have the following: echo Creating DASD swap space on /dev/dasdp1 swapoff -a echo Formatting DASD swap space on /dev/dasdp1 dasdfmt -vy -f /dev/dasdp -d cdl -l swap01 -b 4096 echo Partitioning DASD swap space on /dev/dasdp1 fdasd -l swap01 -a /dev/dasdp echo Making DASD swap space on /dev/dasdp1 sleep 5 mkswap /dev/dasdp1 echo Activating DASD swap space on /dev/dasdp1 swapon -a echo Verifying DASD swap space on /dev/dasdp1 swapon ?s I can change in my SLES11 cloning script to: echo Creating DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapoff -a echo Formatting DASD swap space on /dev/disk/by-path/ccw-0.0.5F dasdfmt -vy -f /dev/disk/by-path/ccw-0.0.5F -d cdl -l swap01 -b 4096 echo Partitioning DASD swap space on /dev/disk/by-path/ccw-0.0.5F fdasd -l swap01 -a /dev/disk/by-path/ccw-0.0.5F echo Making DASD swap space on /dev/disk/by-path/ccw-0.0.5F sleep 5 mkswap /dev/disk/by-path/ccw-0.0.5F echo Activating DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapon -a echo Verifying DASD swap space on /dev/disk/by-path/ccw-0.0.5F swapon ?s Is this what I should do? I am hesitant as there might me other scripted mounts that do /dev/dasdx# . Should I eat the bullet now or defer? Peter From: Ron Foster at Baldor-IS rfos...@baldor.com To: LINUX-390@vm.marist.edu Date: 09/29/2011 11:38 AM Subject:Re: DASD assignment different between SLES10 and SLES11 Sent by:Linux on 390 Port LINUX-390@vm.marist.edu Peter, Actually we observed this behavior as far back as SLES10 SP3. We started using persistent device names and forgot about it. The documentation for this is in the release notes. From the release notes: Installation using Persistent Device names If you plan to add additional storage devices to your system after the OS installation, we strongly recommend to use persistent device names for all storage devices during installation. The installer by default uses the kernel device names. There is additional information in the SLES11 device driver manual on the developerworks site. From the Device Drivers, Features, and Commands manual SC34-2595-01 (from page 35) autodetect causes the DASD device driver to allocate device names and the corresponding minor numbers to all DASD devices and set them online during the boot process. See ?DASD naming scheme? on page 31 for the naming scheme. The device names are assigned in order of ascending subchannel numbers. Auto-detection can yield confusing results if you change your I/O configuration and reboot, or if you are running as a guest operating system in VM because the devices might appear with different names and minor numbers after rebooting. What may have changed is how dasd= is specified and/or how it works. It is different between the SLES10 and SLES11 manuals. Ron Foster From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E. Abresch Jr. - at Pepco [peabre...@pepco.com] Sent: Thursday, September 29, 2011 9:59 AM To: LINUX-390@VM.MARIST.EDU Subject: DASD assignment different between SLES10 and SLES11 Under SLES10, we see the following: 0.0.0150(ECKD) at ( 94: 0) is dasda : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0151(ECKD) at ( 94: 4) is dasdb : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0152(ECKD) at ( 94: 8) is dasdc : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0153(ECKD) at ( 94: 12) is dasdd : active at blocksize 4096, 600840 blocks, 2347 MB 0.0.0154(ECKD) at ( 94: 16) is dasde : active at blocksize 4096, 600840 blocks, 2347
Re: DASD assignment different between SLES10 and SLES11
On the readlink example, the $ indicates a variable. So in a script you
could set vdev='015f' and then reference that value as ${vdev}. Or you could
just hard code in the 015f.
...also note that the by-path disk nodes always zero-pad the beginning of a
virtual device number, so it'll be four characters in length.
It's stripping out the 1 in $15f because it's assuming you mean $1 as a
variable, and (since you never initialized a $1 or passed it in as an
argument/parameter) interpreting it as an empty string.
If your script is always dealing with the same virtual device numbers,
though, and not dealing with something that *has* to use the dasd* format,
you may as well just hard-code the by-path name of the disk.
You can find other mounts by /dev/dasd* in your scripts by running a grep
through them, if that's a concern. Unlike network interfaces (which have a
configurable persistent mapping in
/etc/udev/rules.d/70-persistent-net.rules), there is not any way I'm aware
of to force the assignment of a virtual disk to a particular letter in the
/dev/dasd* scheme. Those assignments happen to usually remain stable - but,
as you've seen, they can change if something changes the order in which they
are brought on line.
On Thu, Sep 29, 2011 at 12:59 PM, Ron Foster at Baldor-IS
rfos...@baldor.com wrote:
Peter,
We do not have SLES11, but from the manual, on page 21 (Module parameters
on the kernel parameter line), it appears that the appropriate format of
the dasd= parameter might be:
dasd_mod.dasd=0.0.150-0.0.15f
so your parameter line would be:
parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1
dasd_mod.dasd=0.0.150-0.0.15f
TERM=dumb
Ron Foster
From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E.
Abresch Jr. - at Pepco [peabre...@pepco.com]
Sent: Thursday, September 29, 2011 11:09 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: DASD assignment different between SLES10 and SLES11
Yes, our root is always dasda and our swap is always dasdp. This is what
our original automated Linux cloning process expects and we wish to not
re-engineer our cloning process at this time.
Under SLES10, in /etc/zipl.conf we have the following:
parameters = root=/dev/dasda1 dasd=150-15f TERM=dumb
Under SLES11, in /etc/zipl.conf we have the following:
parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 TERM=dumb
I am thinking this is where I need to make my change, but my own memory is
slow to swap back in for me to remember all of it.
But . . .
readlink -f /dev/disk/by-path/ccw-0.0.$15F
reveals /dev/disk/by-path/ccw-0.0.5F - it seems to strip the 1 from the
15f.
In my SLES10 cloning script, where I have the following:
echo Creating DASD swap space on /dev/dasdp1
swapoff -a
echo Formatting DASD swap space on /dev/dasdp1
dasdfmt -vy -f /dev/dasdp -d cdl -l swap01 -b 4096
echo Partitioning DASD swap space on /dev/dasdp1
fdasd -l swap01 -a /dev/dasdp
echo Making DASD swap space on /dev/dasdp1
sleep 5
mkswap /dev/dasdp1
echo Activating DASD swap space on /dev/dasdp1
swapon -a
echo Verifying DASD swap space on /dev/dasdp1
swapon ?s
I can change in my SLES11 cloning script to:
echo Creating DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapoff -a
echo Formatting DASD swap space on /dev/disk/by-path/ccw-0.0.5F
dasdfmt -vy -f /dev/disk/by-path/ccw-0.0.5F -d cdl -l swap01 -b 4096
echo Partitioning DASD swap space on /dev/disk/by-path/ccw-0.0.5F
fdasd -l swap01 -a /dev/disk/by-path/ccw-0.0.5F
echo Making DASD swap space on /dev/disk/by-path/ccw-0.0.5F
sleep 5
mkswap /dev/disk/by-path/ccw-0.0.5F
echo Activating DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapon -a
echo Verifying DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapon ?s
Is this what I should do? I am hesitant as there might me other scripted
mounts that do /dev/dasdx# . Should I eat the bullet now or defer?
Peter
From: Ron Foster at Baldor-IS rfos...@baldor.com
To: LINUX-390@vm.marist.edu
Date: 09/29/2011 11:38 AM
Subject:Re: DASD assignment different between SLES10 and SLES11
Sent by:Linux on 390 Port LINUX-390@vm.marist.edu
Peter,
Actually we observed this behavior as far back as SLES10 SP3. We started
using persistent device names and forgot about it.
The documentation for this is in the release notes. From the release
notes:
Installation using Persistent Device names
If you plan to add additional storage devices to your system after the
OS
installation, we strongly recommend to use persistent device names for
all
storage devices during installation. The installer by default uses the
kernel device names.
There is additional information in the SLES11 device driver manual on the
developerworks site.
From the Device Drivers, Features, and Commands manual
SC34-2595-01
(from page 35)
autodetect
causes the DASD device driver to allocate device names
Re: DASD assignment different between SLES10 and SLES11
Wow, you have to get that command just right:
linuxm02:~ readlink -f /dev/disk/by-path/ccw-0.0.15f
/dev/disk/by-path/ccw-0.0.15f
linuxm02:~ readlink -f /dev/disk/by-path/ccw-0.0.15F
/dev/disk/by-path/ccw-0.0.15F
linuxm02:~ readlink -f /dev/disk/by-path/ccw-0.0.Peter
/dev/disk/by-path/ccw-0.0.Peter
linuxm02:~ readlink -f /dev/disk/by-path/ccw-0.0.015f
/dev/dasdf
I am going to do some more analysis.
I also added the following to the zipl.conf:
parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1
dasd_mod.dasd=0.0.150-0.0.15f
TERM=dumb
and ran a zipl but that generated pages of pages of errors during boot. I
will have to look into that a little deeper as well.
Coding the device path in the cloning script generated other errors.
Again, more analysis on my part.
Peter
From: Christian Paro christian.p...@gmail.com
To: LINUX-390@vm.marist.edu
Date: 09/29/2011 01:23 PM
Subject:Re: DASD assignment different between SLES10 and SLES11
Sent by:Linux on 390 Port LINUX-390@vm.marist.edu
On the readlink example, the $ indicates a variable. So in a script you
could set vdev='015f' and then reference that value as ${vdev}. Or you
could
just hard code in the 015f.
...also note that the by-path disk nodes always zero-pad the beginning of
a
virtual device number, so it'll be four characters in length.
It's stripping out the 1 in $15f because it's assuming you mean $1 as
a
variable, and (since you never initialized a $1 or passed it in as an
argument/parameter) interpreting it as an empty string.
If your script is always dealing with the same virtual device numbers,
though, and not dealing with something that *has* to use the dasd* format,
you may as well just hard-code the by-path name of the disk.
You can find other mounts by /dev/dasd* in your scripts by running a grep
through them, if that's a concern. Unlike network interfaces (which have a
configurable persistent mapping in
/etc/udev/rules.d/70-persistent-net.rules), there is not any way I'm aware
of to force the assignment of a virtual disk to a particular letter in
the
/dev/dasd* scheme. Those assignments happen to usually remain stable -
but,
as you've seen, they can change if something changes the order in which
they
are brought on line.
On Thu, Sep 29, 2011 at 12:59 PM, Ron Foster at Baldor-IS
rfos...@baldor.com wrote:
Peter,
We do not have SLES11, but from the manual, on page 21 (Module
parameters
on the kernel parameter line), it appears that the appropriate format
of
the dasd= parameter might be:
dasd_mod.dasd=0.0.150-0.0.15f
so your parameter line would be:
parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1
dasd_mod.dasd=0.0.150-0.0.15f
TERM=dumb
Ron Foster
From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] On Behalf Of Peter E.
Abresch Jr. - at Pepco [peabre...@pepco.com]
Sent: Thursday, September 29, 2011 11:09 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: DASD assignment different between SLES10 and SLES11
Yes, our root is always dasda and our swap is always dasdp. This is what
our original automated Linux cloning process expects and we wish to not
re-engineer our cloning process at this time.
Under SLES10, in /etc/zipl.conf we have the following:
parameters = root=/dev/dasda1 dasd=150-15f TERM=dumb
Under SLES11, in /etc/zipl.conf we have the following:
parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 TERM=dumb
I am thinking this is where I need to make my change, but my own memory
is
slow to swap back in for me to remember all of it.
But . . .
readlink -f /dev/disk/by-path/ccw-0.0.$15F
reveals /dev/disk/by-path/ccw-0.0.5F - it seems to strip the 1 from the
15f.
In my SLES10 cloning script, where I have the following:
echo Creating DASD swap space on /dev/dasdp1
swapoff -a
echo Formatting DASD swap space on /dev/dasdp1
dasdfmt -vy -f /dev/dasdp -d cdl -l swap01 -b 4096
echo Partitioning DASD swap space on /dev/dasdp1
fdasd -l swap01 -a /dev/dasdp
echo Making DASD swap space on /dev/dasdp1
sleep 5
mkswap /dev/dasdp1
echo Activating DASD swap space on /dev/dasdp1
swapon -a
echo Verifying DASD swap space on /dev/dasdp1
swapon ?s
I can change in my SLES11 cloning script to:
echo Creating DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapoff -a
echo Formatting DASD swap space on /dev/disk/by-path/ccw-0.0.5F
dasdfmt -vy -f /dev/disk/by-path/ccw-0.0.5F -d cdl -l swap01 -b 4096
echo Partitioning DASD swap space on /dev/disk/by-path/ccw-0.0.5F
fdasd -l swap01 -a /dev/disk/by-path/ccw-0.0.5F
echo Making DASD swap space on /dev/disk/by-path/ccw-0.0.5F
sleep 5
mkswap /dev/disk/by-path/ccw-0.0.5F
echo Activating DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapon -a
echo Verifying DASD swap space on /dev/disk/by-path/ccw-0.0.5F
swapon ?s
Is this what I should do? I am hesitant as there might me other scripted
mounts that do /dev/dasdx# . Should I eat the bullet now or defer?
Peter
Re: DASD assignment different between SLES10 and SLES11
Peter, Have you tried: parameters = root=/dev/disk/by-path/ccw-0.0.0150-part1 dasd=0.0.0150-0.0.015f TERM=dumb Ronald van der Laan -- 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/
