2015-12-22 My main notebook (deseo) has a 8 GB CF card as main storage. The current card is or became pretty slow. I exchanged it today. This is how I did it: 1) Remove the card and plug it into a card reader of another com- puter. 2) Make a backup: :-* dd /data/backup/deseo-cf_2015-12-22.dd 15904161+0 records in 15904161+0 records out 8142930432 bytes (8.1 GB) copied, 9062.13 s, 899 kB/s It was a failure to copy only its first partition and not the en- tire disk -- no real problem, but a hazzle. :-* file /data/backup/deseo*dd /data/backup/deseo-cf_2015-02-07.dd: x86 boot sector, LInux i386 boot LOader; partition 1: ID=0x83, active, starthead 1, startsector 63, 15904161 sectors, code offset 0 xeb /data/backup/deseo-cf_2015-12-22.dd: Linux rev 1.0 ext3 filesystem data, UUID=85b95127-9298-4a82-89a3-d6c5e6a7aa8d (large files) 3) Restore to the new card. As I haven't saved the whole image, I couldn't just write the image back. I first needed to restore the partition table. -- Well, I could have made a new (full) copy of the original disk, but I didn't wanted to wait for it to finish; and anyway, there's no reason to always walk the known path ... at least not in Unix. -- So I copied to partition table from the old (full) backup. I had a look how large it is, i.e. the size difference between the images: :-* ll /data/backup/deseo*dd | awk '{print $5, $9}' 8142962688 /data/backup/deseo-cf_2015-02-07.dd 8142930432 /data/backup/deseo-cf_2015-12-22.dd Then I copied the 32k plus some more, to be sure: :-* dd bs=1024 count=64 /dev/sdb 64+0 records in 64+0 records out 65536 bytes (66 kB) copied, 0.147113 s, 445 kB/s Afterwards, the partition was there, as expected: :-* lsblk -f NAME FSTYPE LABEL MOUNTPOINT sdb └─sdb1 ext3 Now, I was able to write the image to the first partition: :-* dd /dev/sdb1 15904161+0 records in 15904161+0 records out 8142930432 bytes (8.1 GB) copied, 9702.67 s, 839 kB/s 4) Take the card from the card reader and insert it back into deseo. Yay! The machine was running again ... at much faster speed. For whatever reason, today my home server had a hangup, after 276 days of non-stop operation. It might have been a thermal problem, because it is passively cooled and I was working hard on it to- day. After the hard reset, the software RAID had to rebuild itself. While I was on that, I decided to start a backup to an external 1 TB disk. I should do this regularly! It's a strange feeling for me to have two 320 GB disks lying around, without really knowing what to do with them. They are too small to be real backup disks; they would only consume power if I put them into the home server; my desktop computers don't need that much storage space. I have to realize that these disks are just spares, currently. There will be uses for them in the future, when other disks become old or too small, but currently they are lying in a box and waiting ... strange. http://marmaro.de/lue/ markus schnalke