Manu Tayal | 1 Sep 12:09 2004

What is the behavior on reboot/shutdown...

Hi,
	If I have created a snapshot for a device and some writes are in
progress on the original device, then how is the shutdown/reboot handled. 

Regards,
Manu

DISCLAIMER:
This message (including attachment if any) is confidential and may be privileged. Before opening
attachments please check them for viruses and defects. MindTree Consulting Private Limited (MindTree)
will not be responsible for any viruses or defects or any forwarded attachments emanating either from
within MindTree or outside. If you have received this message by mistake please notify the sender by
return  e-mail and delete this message from your system. Any unauthorized use or dissemination of this
message in whole or in part is strictly prohibited.  Please note that e-mails are susceptible to change and
MindTree shall not be liable for any improper, untimely or incomplete transmission.

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Michel Bouissou | 1 Sep 13:08 2004
Picon

Re: [O.T.] What is the behavior...

Manu Tayal a écrit :
>
> DISCLAIMER:
> This message (including attachment if any) is confidential and may be
> privileged. Before opening attachments please check them for viruses and
> defects. MindTree Consulting Private Limited (MindTree) will not be
> responsible for any viruses or defects or any forwarded attachments
> emanating either from within MindTree or outside. If you have received
> this message by mistake please notify the sender by return  e-mail and
> delete this message from your system. Any unauthorized use or
> dissemination of this message in whole or in part is strictly prohibited.
> Please note that e-mails are susceptible to change and MindTree shall not
> be liable for any improper, untimely or incomplete transmission.

That makes a good candidate for
http://www.goldmark.org/jeff/stupid-disclaimers/

--

-- 
Michel Bouissou <michel <at> bouissou.net> OpenPGP ID 0xDDE8AC6E

Disclaimer:
 By sending an email to ANY of my addresses you are agreeing that:
 - I am by definition, "the intended recipient"
 - All information in the email is mine to do with as I see fit and make
   such financial profit, political mileage, or good joke as it lends itself
   to.
   In particular, I may quote it on usenet.
 - I may take the contents as representing the views of your company.
 This overrides any disclaimer or statement of confidentiality that may be
 included on your message.
(Continue reading)

ram prakash | 1 Sep 08:18 2004

Why hdb_freespace2,3

 
I have attached My Second disk hdb for the testing purpose for evms,
I tried to create More logical partitions, Some times later it produced
hdb_freelist2, What is the technical issue behind this?
Can I Emulate the evms. that is , Say

1)  evms:c,s,hdb_freespace2,size=100MB,primary=true,bootable=false.
Now i want to emulate this operation,(Can i predict exactly What the output of this command would be?)

2)  I created a partition with bootable=true, Now i want to remove it.
  How can i do this?
Thanks
Ramprakash Jelari.





Suresh Balla | 1 Sep 13:43 2004

(no subject)

hi,
i am getting these message in evms-engine.log

Sep 01 09:55:02 sunday.comodo.net _3_ Engine: do_fork_and_execvp:
execvp() failed.  errno is 2: No such file or directory

Sep 01 09:55:02 sunday.comodo.net _0_ Engine: engine_user_message:
Message is: Engine: The plug-in SWAPFS in module
/lib/evms/2.3.3/swap-1.1.8.so failed to load.  The plug-in's
setup_evms_plugin() function failed with error code 38: Function not
implemented.

Sep 01 09:55:02 sunday.comodo.net _3_ Engine: engine_open_object: Open
of /dev/evms/.nodes/ram0 failed with error code 22: Invalid argument

Sep 01 09:55:03 sunday.comodo.net _2_ MDLinearRegMgr:
md_ioctl_flush_buffer_cache: Unable to open object hda8 to send ioctl

Sep 01 09:55:03 sunday.comodo.net _2_ MDRaid5RegMgr:
md_ioctl_flush_buffer_cache: Unable to flush buffer cache on md/md0
(major=9, minor=0)

Sep 01 09:55:03 sunday.comodo.net _2_ Engine: ensure_dev_node:
Major:minor 0:0 is not valid.

Sep 01 09:55:31 sunday.comodo.net _5_ Engine: is_volume_change_pending:
Change pending: Volume /dev/evms/hda1 is not active.

thanking you
suresh

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Kevin Corry | 1 Sep 15:35 2004
Picon

Re: What is the behavior on reboot/shutdown...

On Wednesday 01 September 2004 5:09 am, Manu Tayal wrote:
> Hi,
> 	If I have created a snapshot for a device and some writes are in
> progress on the original device, then how is the shutdown/reboot handled.

When the system shuts down, all filesystems are always unmounted or remounted 
read-only. Thus, before the system actually shuts down, all writes to the 
origin volume have stopped, and there can be no more copies to the snapshot 
in progress. When the system boots, at some point evms_activate runs (either 
during the init-ramdisk or during the init scripts) and both the origin and 
the snapshot are activated again, and reflect the state at the time the 
system was shut down.

--

-- 
Kevin Corry
kevcorry <at> us.ibm.com
http://evms.sourceforge.net/

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Kevin Corry | 2 Sep 18:37 2004
Picon

Re: (no subject)

Hi Suresh,

On Wednesday 01 September 2004 6:43 am, Suresh Balla wrote:
> hi,
> i am getting these message in evms-engine.log
>
> Sep 01 09:55:02 sunday.comodo.net _3_ Engine: do_fork_and_execvp:
> execvp() failed.  errno is 2: No such file or directory

This means EVMS tried to call some external program, but the binary couldn't 
be found. Could be bad, or could be normal. Hard to tell without any 
additional context.

> Sep 01 09:55:02 sunday.comodo.net _0_ Engine: engine_user_message:
> Message is: Engine: The plug-in SWAPFS in module
> /lib/evms/2.3.3/swap-1.1.8.so failed to load.  The plug-in's
> setup_evms_plugin() function failed with error code 38: Function not
> implemented.

This means the swap FSIM plugin didn't load. Most likely it couldn't find or 
run the mkswap program. This could be related to the above error. See if you 
have the mkswap program on your system and make sure it's in your PATH.

> Sep 01 09:55:02 sunday.comodo.net _3_ Engine: engine_open_object: Open
> of /dev/evms/.nodes/ram0 failed with error code 22: Invalid argument

Ignorable. It's just attempting to open a device node that doesn't have a real 
device attached to it.

> Sep 01 09:55:03 sunday.comodo.net _2_ MDLinearRegMgr:
> md_ioctl_flush_buffer_cache: Unable to open object hda8 to send ioctl
>
> Sep 01 09:55:03 sunday.comodo.net _2_ MDRaid5RegMgr:
> md_ioctl_flush_buffer_cache: Unable to flush buffer cache on md/md0
> (major=9, minor=0)

Also ignorable.

> Sep 01 09:55:03 sunday.comodo.net _2_ Engine: ensure_dev_node:
> Major:minor 0:0 is not valid.

A commonly seen warning. Just means some volume or object isn't active yet.

> Sep 01 09:55:31 sunday.comodo.net _5_ Engine: is_volume_change_pending:
> Change pending: Volume /dev/evms/hda1 is not active.

Again, this just means some volume isn't active yet. It will likely be 
activated when you press "save".

--

-- 
Kevin Corry
kevcorry <at> us.ibm.com
http://evms.sourceforge.net/

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Kevin Corry | 2 Sep 18:44 2004
Picon

Re: Why hdb_freespace2,3

Hi Ramprakash,

On Wednesday 01 September 2004 1:18 am, ram  prakash wrote:
> I have attached My Second disk hdb for the testing purpose for evms,
> I tried to create More logical partitions, Some times later it produced
> hdb_freelist2, What is the technical issue behind this?

All freespace portions of the disk are numbered in the order that they appear 
on the disk. When you switch from primary to logical partitions, because of 
the cylinder alignment requirements there is usually some freespace left 
between the last primary and the first logical. This usually becomes 
"freespace1". Then the freespace that follows the last logical partition is 
usually "freespace2". 

> Can I Emulate the evms. that is , Say
>
> 1)
> evms:c,s,hdb_freespace2,size=100MB,primary=true,bootable=false.
> Now i want to emulate this operation,(Can i predict exactly What the output
> of this command would be?)

I'm not sure I understand your question. This command should produce a 100MB 
segment, and since you specified primary=true, it will be called hdb1, hdb2, 
hdb3, or hdb4 (depending on which primary segments already exist).

> 2) I created a partition with bootable=true, Now i want to remove
> it. How can i do this?

The "bootable" flag is completely irrelevant to Linux. The only reason you'd 
ever need to set it or unset it is if you're dual-booting this machine 
between Linux and Windows.

--

-- 
Kevin Corry
kevcorry <at> us.ibm.com
http://evms.sourceforge.net/

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Ken Hwang | 3 Sep 00:58 2004

Re: howto change disk in a evns configured host

Hi,

I also have a change disk in EVMS question please help if it's possible. I
can only use EVMS cli instead of evmsgui due to it is an embedded linux
system. The system has 3 disks in several raid-5 such as (sdb1, sdc1, sdd1)
in md/md1, (sdb5, sdc5, sdd5) in md/md2, (sdb6, sdc6, sdd6) in md/md3. After
one disk failed, I bought a new and bigger disk to replace the failed one. I
want to write a script program to rebuild the 3 raid-5 after power-on but
don't know how. I enter EVMS and it reports md1, md2, and md3 are in
degraded mode but did not tell me which disk was failed. I assume I will
need to partition the new disk to be the same as the old one, and then for
each raid-5 I need to do remstale (with the correct partition#), and then
addspare to rebuild them. But I don't know how to identify the failed one,
and then re-create the partitions (I have extended partitions so I assume
copy mbr will not work). Also, I had hard time to find all the commands I
can use in CLI such as the "Replace", "Remstale", and "Addspare", can
someone please send me a pointer to their document?

Thanks,

Ken

Message: 15
Subject: Re: [Evms-devel] howto change disk in a evns configured host
To: Anthony Mutiso <anm.mlist01 <at> limuru.com>
Cc: evms-devel <at> lists.sourceforge.net
From: Steve Dobbelstein <steved <at> us.ibm.com>
Date: Tue, 31 Aug 2004 16:06:29 -0500

> Hi folks,
>
> I am looking for some information or a howto on how to change the disks
> in a host with evms.
>
> The host in question currently had two small disks with evms configured
> with reiserfs on 4 lv's. I would like to replace the small disks with on
> large disk while maintaining the same lv data. I have searched the mail
> list archives and the evms user guide but have not found anything that I
> recognize as what steps to take. I am familiar with the traditional
> steps for a none-evms host, in which I install the drive, reboot from
> cd,  partition the new disk, mount the old partitions, mount the new
> partitions, copy the data from the old partition to the new partitions,
> halt, remove the old disk(s), and set the up the new disk as the
> replacement and reboot. I guess what I am unable to figure out is how to
> create the new lv's that match the old ones in name so that I can copy
> the data over and reboot with on the new disk and have the names of
> things be the same.
>
> Thanks for the help
>
> Anthony

Hi, Anthony.

I assume you have a simple setup where your disks are partitioned and you
made volumes from the partitions.

If you made EVMS volumes, not compatibility volumes, then you can use
EVMS's "replace" function to swap out the storage object that was used to
make the volume.  In most cases, replace can be done while the volume is
mounted.  EVMS will copy all the data over to the new storage object and
setup all the metadata so that the EVMS volume uses the new object.

EVMS can't replace a segment (partition) that was made into a compatibility
volume, for instance, hda6 made into /dev/evms/hda6.  The volume derives
its name from the object.  If hda6 were to be replaced with hda7 the new
volume would be /dev/evms/hda7, not /dev/evms/hda6.  Anything that want to
mount /dev/evms/hda6 will fail.

Hope this helps.

Steve D.

--__--__--

_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

End of Evms-devel Digest
---
Incoming mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.687 / Virus Database: 448 - Release Date: 5/16/2004

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.687 / Virus Database: 448 - Release Date: 5/16/2004

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

Kevin Corry | 3 Sep 03:30 2004
Picon

Re: howto change disk in a evns configured host

Hi Ken,

On Thursday 02 September 2004 17:58, Ken Hwang wrote:
> Hi,
>
> I also have a change disk in EVMS question please help if it's possible. I
> can only use EVMS cli instead of evmsgui due to it is an embedded linux
> system. The system has 3 disks in several raid-5 such as (sdb1, sdc1, sdd1)
> in md/md1, (sdb5, sdc5, sdd5) in md/md2, (sdb6, sdc6, sdd6) in md/md3.
> After one disk failed, I bought a new and bigger disk to replace the failed
> one. I want to write a script program to rebuild the 3 raid-5 after
> power-on but don't know how. I enter EVMS and it reports md1, md2, and md3
> are in degraded mode but did not tell me which disk was failed. I assume I
> will need to partition the new disk to be the same as the old one, and then
> for each raid-5 I need to do remstale (with the correct partition#), and
> then addspare to rebuild them. But I don't know how to identify the failed
> one, and then re-create the partitions (I have extended partitions so I
> assume copy mbr will not work). Also, I had hard time to find all the
> commands I can use in CLI such as the "Replace", "Remstale", and
> "Addspare", can someone please send me a pointer to their document?

If one of your three disks physically failed, then I'm assuming it's no longer 
in the machine, and thus isn't seen at all by EVMS. If this is the case, then 
there are no "stale" objects to remove from the RAID regions. The only time 
an object is "stale" is when that object is only temporarily missing from the 
region.

You said you added an entirely new disk to your system to replace the failed 
one. So you'll need to start by creating three segments on that disk 
corresponding to each of the segments on the failed disk.

In the cli, you can run "query:disks" to get a list of disks, and you can run 
"query:segments" to get the list of segments. Obviously the disk that doesn't 
also have any segments is the disk you just added. If the new disk is 
completely blank, you'll need to first remove the compatibility volume from 
the disk, which you can do with "delete:/dev/evms/≤disk_name>".

Next you'll assign the DOS plugin to that disk, with the command:
assign:DosSegMgr={},<disk_name>
Then create the three segments on that disk, with the command:
create:segment,<disk_name>_freespace1,Size=<size>,Primary=true
create:segment,<disk_name>_freespace1,Size=<size>,Primary=false
create:segment,<disk_name>_freespace2,Size=<size>,Primary=false

Now that you have three new segments, you're going to add each one to the 
appropriate RAID region as a spare object. Since each RAID region is 
degraded, once you add the spare the kernel will activate it and begin to 
sync the data and parity. Add the spares with the commands:
task:addspare,md/md1,<disk_name>1
task:addspare,md/md2,<disk_name>5
task:addspare,md/md3,<disk_name>6

Then you can watch the sync progress in /proc/mdstat. Hope this helps. Let us 
know if you have additional questions.

--

-- 
Kevin Corry
kevcorry <at> us.ibm.com
http://evms.sourceforge.net

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel

sushmitha | 3 Sep 13:04 2004

Reg:Change of Partition ID

Hello,

I was unable to change the id(typebynumber) for a segment that already
exists.
After creating segments I want to create Raid or LVM by using those
existing segments.I tried using modify and set commands in evms cli.But no
use.

Is there any possible way to change the id and creating a mount point for
an existing segments for conversion of Raid and LVM using EVMS CLI?

Because I saw through fdisk we can change id for existing partitions.

Thanks,
Susmitha

-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
_______________________________________________
Evms-devel mailing list
Evms-devel <at> lists.sourceforge.net
To subscribe/unsubscribe, please visit:
https://lists.sourceforge.net/lists/listinfo/evms-devel


Gmane