I have an out of order partition table and /dev's. It is a twisted tail. Let's see you unravel this issue…
When I run the CloneZilla backup menus, I ask to image the whole disk it uses the /dev's sequence sda1, 2(primaries & 3 extended), then 5(logical). The partitions physical sequence matches /dev/sda2, /dev/sda1, /dev/sda5. Why the 1st two are so arranged I have no idea, they just are. I did add MS-DOS sda5 later.
The CloneZilla default restore order is…/dev/sda1, /dev/sda2, /dev/sda5. BTW: Since the new disk is larger I have chosen the write a new partition table, proportionally. (Is that correct?) The restore sequence I want is /dev/sda2(primary), /dev/sda1(primary), /dev/sda5(logical).
Should I manually modify the CloneZilla script comand line /dev's to match the physical sequence of partitions?
Later after I just get the thing cloned, I was thinking I would reorder the partitions to…
1. Recovery, 2.31 GB
2. MS-DOS 504 MB
3. Windows XP
4. Extended
5> OpenSUSE 11.3 dual boot
…then fix the partition table…but it may be that the laptop maker did this sequence diliberately…
I was not thinking of doing all of this at once for that reason.
The out of order partition table is like this…
1. /dev/sda2 Recovery partition, primary, (Win partition(0)-?)
2. /dev/sda1 Windows XP, 39 GB, primary, (Win partition(1)-?)
3. /dev/sda3, Extended partition, (Win partition(?)) <<<==== What's the MS Win XP partition(?)
4. /dev/sda5, MS-DOS, 504 MB, (Win partition(?)) <<<===== ..same here
….from sda-pt.parted…look at the out of order partition numbers….2, 1, 3, 5
Model: ATA HTS541040G9AT00 (scsi)
Disk /dev/sda: 78140160s
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
2 63s 4851629s 4851567s primary fat32 diag
1 4851630s 77144129s 72292500s primary ntfs boot
3 77144130s 78124094s 979965s extended lba
5 77144193s 78124094s 979902s logical fat16
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
While I'm waiting for a reply …I'm going to erase the partitions and try another restore using the saved partition table to see if the partitions restore sequence changes.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I ran the restore using most of the defaults except adding verbose comment printing and using the saved partition table setup. That did reproduce the out-of-order partition table as per my original posting. The restore is now running and I'm watching the partClone text screen updates.
What I'm seeing is the interaction as the backup is delivered to the client from the Samba server. The LAN is 100 Megabits/second raw rate with the SMB protocol running on top of that raw rate.
partClone reports that it is transferring at a rate on average of 80 MB's, Mega-Bytes/second. There is a total of 34 Gigagabytes of data to be transferred to a 37 Gigabyte sized disk partition. The calculation is that will take about 7 hours total time.
However since my last observation the transfer has completely stopped. I over to a command prompt to check things out.. I see error messages…
CIFS VFS: No response to cmd 46 mid 50714
CIFS VFS: Send Error in read=-11
CIFS VFS: Send Error in read=-11
…now it is continuing to do the restore…
…the screen now says 30 minutes have passed but my clock says over 60 minutes…partClone's clock is wrong….
…6.58% of 34 GB have been restored
…this is way too slow for any practical use…
…it is time to quit and go to bed…
..I have copied the entire /var/log directories & files to a server along with the run script…
..I can EMail these to the developers to help them determine the issues with the cifs failures and slow transfer rates…
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Normally this problem is about the network quality, or the driver for the network card.
You can try to use different versions of Clonezilla live, e.g. 1.2.5-38, 20100729-lucid or20100729-maverick to see if any difference.
Steven.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It may take a day or two but I'm trying to determine if the respective NIC's are in full or half-duplex modes. If different then some issues could arise, but then again maybe not. I'm also looking for a network test suite/model to try to test different TCP/IP issues.
The other is to test the SMB protocol to see what it expects the client to be doing and how soon. Concurrency issues have not been handled well by most programmers.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've got to switch to getting thing done. I'll have to just install the two drives in an old server and clone direct. Bummer. I so wanted this networked approach to work now. I'll have to investigate network performance and find a way to test separately the performance of partClone in different situations.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm running a CloneZilla disk 2 disk which is going well. I estimate it will copy 3 partitions with about 36 GB to the target in about 50 minutes. The rate is about 1.12 GB per minute according to the output on the screen. So far no problems. Good this feature works.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have an out of order partition table and /dev's. It is a twisted tail. Let's see you unravel this issue…
When I run the CloneZilla backup menus, I ask to image the whole disk it uses the /dev's sequence sda1, 2(primaries & 3 extended), then 5(logical). The partitions physical sequence matches /dev/sda2, /dev/sda1, /dev/sda5. Why the 1st two are so arranged I have no idea, they just are. I did add MS-DOS sda5 later.
The CloneZilla default restore order is…/dev/sda1, /dev/sda2, /dev/sda5. BTW: Since the new disk is larger I have chosen the write a new partition table, proportionally. (Is that correct?) The restore sequence I want is /dev/sda2(primary), /dev/sda1(primary), /dev/sda5(logical).
Should I manually modify the CloneZilla script comand line /dev's to match the physical sequence of partitions?
Later after I just get the thing cloned, I was thinking I would reorder the partitions to…
1. Recovery, 2.31 GB
2. MS-DOS 504 MB
3. Windows XP
4. Extended
5> OpenSUSE 11.3 dual boot
…then fix the partition table…but it may be that the laptop maker did this sequence diliberately…
I was not thinking of doing all of this at once for that reason.
The out of order partition table is like this…
1. /dev/sda2 Recovery partition, primary, (Win partition(0)-?)
2. /dev/sda1 Windows XP, 39 GB, primary, (Win partition(1)-?)
3. /dev/sda3, Extended partition, (Win partition(?)) <<<==== What's the MS Win XP partition(?)
4. /dev/sda5, MS-DOS, 504 MB, (Win partition(?)) <<<===== ..same here
….from sda-pt.parted…look at the out of order partition numbers….2, 1, 3, 5
Model: ATA HTS541040G9AT00 (scsi)
Disk /dev/sda: 78140160s
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
2 63s 4851629s 4851567s primary fat32 diag
1 4851630s 77144129s 72292500s primary ntfs boot
3 77144130s 78124094s 979965s extended lba
5 77144193s 78124094s 979902s logical fat16
While I'm waiting for a reply …I'm going to erase the partitions and try another restore using the saved partition table to see if the partitions restore sequence changes.
I ran the restore using most of the defaults except adding verbose comment printing and using the saved partition table setup. That did reproduce the out-of-order partition table as per my original posting. The restore is now running and I'm watching the partClone text screen updates.
What I'm seeing is the interaction as the backup is delivered to the client from the Samba server. The LAN is 100 Megabits/second raw rate with the SMB protocol running on top of that raw rate.
partClone reports that it is transferring at a rate on average of 80 MB's, Mega-Bytes/second. There is a total of 34 Gigagabytes of data to be transferred to a 37 Gigabyte sized disk partition. The calculation is that will take about 7 hours total time.
However since my last observation the transfer has completely stopped. I over to a command prompt to check things out.. I see error messages…
CIFS VFS: No response to cmd 46 mid 50714
CIFS VFS: Send Error in read=-11
CIFS VFS: Send Error in read=-11
…now it is continuing to do the restore…
…the screen now says 30 minutes have passed but my clock says over 60 minutes…partClone's clock is wrong….
…6.58% of 34 GB have been restored
…this is way too slow for any practical use…
…it is time to quit and go to bed…
..I have copied the entire /var/log directories & files to a server along with the run script…
..I can EMail these to the developers to help them determine the issues with the cifs failures and slow transfer rates…
Normally this problem is about the network quality, or the driver for the network card.
You can try to use different versions of Clonezilla live, e.g. 1.2.5-38, 20100729-lucid or20100729-maverick to see if any difference.
Steven.
It may take a day or two but I'm trying to determine if the respective NIC's are in full or half-duplex modes. If different then some issues could arise, but then again maybe not. I'm also looking for a network test suite/model to try to test different TCP/IP issues.
The other is to test the SMB protocol to see what it expects the client to be doing and how soon. Concurrency issues have not been handled well by most programmers.
I've got to switch to getting thing done. I'll have to just install the two drives in an old server and clone direct. Bummer. I so wanted this networked approach to work now. I'll have to investigate network performance and find a way to test separately the performance of partClone in different situations.
I'm running a CloneZilla disk 2 disk which is going well. I estimate it will copy 3 partitions with about 36 GB to the target in about 50 minutes. The rate is about 1.12 GB per minute according to the output on the screen. So far no problems. Good this feature works.