Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google

Our Sites

covecube.com 
community.covecube.com 
blog.covecube.com 
wiki.covecube.com 
bitflock.com 
stablebit.com 

Poll

No poll attached to this discussion.

In order to better support our growing community we've set up a new more powerful forum.


The new forum is at: http://community.covecube.com


The new forum is running IP.Board and will be our primary forum from now on.


This forum is being retired, but will remain online indefinitely in order to preserve its contents. This forum is now read only.


Thank you,

M4 and Crashplan

edited January 2012 in DrivePool

Firstly M4 is just briliant ! Great work Alex, how you get all this done I don't know.

The upgrade went well except now Crashplan is trying to backup everything all again, all .5TB of it. Any ideas how to fix this, I was under the impression Crashplan would recognize the same files ?

Comments

  • I could be wrong, I suspect it only does this sort of detection when you select the "Adopt a computer" method during a reinstall of Crashplan.

    I know in the past when I've performed full computer reinstalls, Crashplan has required the file paths (including drive letter) to be the same. It then performed a local hash check and marked everything as OK. Minimal traffic was transferred in this situation to the Crashplan servers.

    The detection of an existing file is something I've only seen work when I move a file from one folder to another in an existing crashplan backup set. In this situation, it doesn't re-upload the file. The important thing here though is the root path remains the same.

    Short of recreating your original path using Subst or junction points, I'm not sure if this is achievable. Definitely happy to be proven wrong on this one :)
  • edited January 2012 Member
    I run Crashplan on my WHS 2011 and have just upgraded from M3 to M4, and all seems to be working OK. All I did was change my Crashplan backup set settings to backup from the new locations on the pool drive. If you also have incoming backups on the pool, these will each need to be re-attached (mine are on non-pooled drives for now). This triggered Crashplan to backup again to the destinations, which appears to be a full backup, but as Saltydh explained the way it should work is that only the hashes are exchanged, the actual files are not sent to the destinations unless they have changed.

    This can still take some time as I think all the hashes have to be recalculated locally (i.e. reading every file in its entirety), but still much faster than a backup to an empty destination.

    Edit... Here is my Crashplan Central backup of just over 300GB - still running
    since I upgraded to M4 yesterday, but not long to go. The network
    connection is throttled at 800 kbit/s, the much faster speed showing is
    because it measures this by the size of the files even though only the
    hashes are exchanged over the network.
    image

  • edaeda
    Member
    I also run CrashPlan for backing up not only to CrashPlan Central, but also to my original WHS 1.0 system (which no longer does anything except provide off-server backup for WHS 2011).  Datageek and Saltydh are correct.  While it appears that the entire backup is happening, CrashPlan matches the files to the new locations and "data dedup" prevents the need to resend the whole file.  That's why the upload speed is reported so much higher than the actual connection speed.  I confirmed that this is the case with CrashPlan support back when I made the shift to WHS 2011 in the first place.

    CrashPlan has done a really nice job of backup.  I'm a fan.
  • I'll try this, thanks for your help !
  • I also run crashplan, I can confirm "eda" is correct, is reverifying the files via hash and not having to reupload the entire data set.
  • edited January 2012 Member

    eda, do you use crashplan in conjunction with the server build in backup, or do you have the WHS internal backup disabled? Will crashplan backup to WHS V1?

     

    Thanks

    Rufus Roper

  • Yes it apears after 4 days of what looked like to be backing up it was just checking.

    It did use some bandwidth but nowhere near the amount of data. Good job Crashplan (again) 

  • @m2and3ingo
    1) I also use WHS Server backup, but just for the C drive. Crashplan doesn't do system image backups (i.e. to restore a system including OS, applications, settings etc), only files.

    2) Yes, Crashplan can backup to any system with Crashplan installed, including WHS 2011, WHS v1 (based on Win Server 2003), other versions of Windows, Linux and Mac OS X. The catch on WHS v1 is that UNC paths are not properly supported by Crashplan, and backing up Drive Extender data is not supported via D:\Shares\ due to concerns about potential data corruption (although no one has reported an actual problem with just reading the data). The workaround I used was here: http://www.mylifewired.com/686/properly-setting-up-crashplan-for-whs/
  • edaeda
    Member

    eda, do you use crashplan in conjunction with the server build in backup, or do you have the WHS internal backup disabled? Will crashplan backup to WHS V1?

    I currently have WHS 2011 internal backup unconfigured.  That is because my understanding is that the internal backup requires a drive attached directly to the server, which is not what I want.  I'd like to create some physical distance between my server and it's backup, which USB and ESATA don't provide.

    I currently use CrashPlan to backup to WHS V1.  I have configured my servers so that I have my new WHS 2011 system as my primary production server (expressing my confidence in Alex!), but with backups to both the CrashPlan online service as well as to the older V1 server.  Crashplan is installed on the V1 server, and I backup to a single directory on the V1.  However, since having applications write directly to directories on V1 was always a risk, I made sure the directory is not duplicated (after all, why duplicate a backup--it IS a duplicate).

    In this case, the V1 server is nothing more than a high-tech NAS, but I can run CrashPlan on it directly, which I can't do on a NAS.  I don't think.

    I am emailed automatic reports each day by Crashplan, and everything backs up reliably.  Occasional cursory checks confirm that.  With three copies of everything (plus duplication on the WHS 2011 server with DrivePool), I can sleep pretty soundly at night.
  • Has anyone been using inbound backups to the crashplan on whs2011?  Specifically storing those inbound backups on the pool?

    I want to back up my desktop to my homeserver just for quick restores if need be and I'd like to use the pool to not have to worry about space on a dedicated drive.  Eventually I'll get my parents offsite machine backing up to my server as well.
  • Yes, I run crashplan and use it for inbound backup. I created a folder on the pool for my parents PC. I have most of my backup on a 2TB external drive hooked to the PC. I do plan to move that offsite though at some point maybe to my parents house.
  • So you aren't storing the inbound backups to the pool?  Is anyone?
  • Yes, I am using the pool for inbound storage.
  • Just had to reboot my server because the crashplan service kept reporting the destination as unavailable to my Mother's crashplan client.

    Not quite sure things are 100% stable using the pool for inbound storage...
  • Member
    I have a question: When using CrashPlan to back up server folders (not client backups), do you add the pool drive or the individual disks in the pool to CrashPlan?

    Thanks
  • Resident Guru
    Hopefully I'm not leaping to a hasty conclusion, but I'd pick the pool drive. Crashplan has automatic de-duping; if your pool has duplication, then telling it to backup the individual disks would just be more work for the same result.
Sign In or Register to comment.