Rsync'ing old server to Linode problems

So I just bought a linode and I'm following the guide on how to migrate my old server (slicehost) to my new linode. Problem is, the rsync keeps dying halfway through saying "No space left on device". I don't understand how that can be possible. Originally I started with a disk image of 15GB, then increased to 25GB and now I'm at 55GB and it continues to fail. My slicehost disk is only using 6.8GB out of a total of 38GB. I've searched around the internet but I cant find any solutions, can someone help me?

10 Replies

What's the command you're using to do the rsync? What does 'df' say on both ends, before and after? Is it excluding /proc and other weird places?

So, you have a Linode 2048? That's the only Linode with a disk big enough (64GB) to support 55GB of data.

If your slice isn't using any more than 6.8GB, then even the smallest linode should have enough disk space… What parameters are you calling rsync with, and what's using up the space when you fill up the disk?

Command:

rsync --exclude="/sys/*" --exclude="/proc/*" -aHSKDvz --numeric-ids -e ssh / root@<linode ip="">:/mnt/xvdb/</linode> 

Slicehost:

$ df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/sda1             39220672   7035100  30591740  19% /
tmpfs                   511268         0    511268   0% /lib/init/rw
udev                     10240        16     10224   1% /dev
tmpfs                   511268         4    511264   1% /dev/shm

Linode (before):

# df                                                                                     
Filesystem           1K-blocks      Used Available Use% Mounted on                                  
none                   1618464      1008   1617456   1% /                                           
/ramdisk/dev/xvda       127954    127954         0 100% /cdrom                                      
/dev/loop0              118784    118784         0 100% /FINNIX                                     
none                   1618464      1008   1617456   1% /UNIONFS                                    
tmpfs                    10240       116     10124   2% /dev                                        
devshm                 1024228         0   1024228   0% /dev/shm                                    
/dev/xvdb             56039780    131228  55908552   1% /media/xvdb

It'll take a while before I can get the "after" for Linode, but when I ran it last night, I remember that xvdb wasn't full, which is why I'm so confused about why I'm getting this message. And yes Guspax, I do have a Linode 2048.

@Telanor:

Command:

rsync --exclude="/sys/*" --exclude="/proc/*" -aHSKDvz --numeric-ids -e ssh / root@<linode ip="">:/mnt/xvdb/</linode> 
/dev/xvdb             56039780    131228  55908552   1% /media/xvdb

The rsync is putting the data in /mnt/xvdb, your target filesystem is at /media/xvdb. Are you dumping your data to the root filesystem?

The data is supposed to go to that xvdb filesystem. I thought it was odd that what was listed in df and what was in the rsync command were different. I did exactly as the guide said. So should I change the rsync command to use /media/xvdb instead of /mnt/xvdb?

Ok, I changed the command to target /media/xvdb instead and it seems to be working now. Thanks guys. BTW, unless I messed something up, it seems like the guide should be updated to fix that?

be helpful if you actually linked to it

Edit: If you were talking about this guide, then yes, you read it wrong, nowhere does it mention /media/

@OverlordQ:

If you were talking about this guide, then yes, you read it wrong, nowhere does it mention /media/
If you use the mount command with no target under finnix, as shown in the guide, it mounts the filesystem on /dev/media. The guide is wrong when it subsequently refers to /mnt/xdvb.

@pclissold:

If you use the mount command with no target under finnix, as shown in the guide, it mounts the filesystem on /dev/media. The guide is wrong when it subsequently refers to /mnt/xdvb.
To be fair to the guide, that was a reasonably recent change in finnix behavior, so I'm pretty sure when the guide was written the finnix image Linode was using was in fact auto-mounting under /mnt.

But yes, the guide should be updated to reflect the current behavior.

– David

@db3l:

To be fair to the guide, that was a reasonably recent change in finnix behavior, so I'm pretty sure when the guide was written the finnix image Linode was using was in fact auto-mounting under /mnt.
Yes – I should have been more clear -- the guide is now out of date, rather than being incorrect as written.

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Community Code of Conduct