I am no RRD expert, but it must be storing data in binary format. I believe the x86 and SPARC processors store integer values differently -- one is "big-endian" and the other is "little-endian" (I don't remember which is which) and this is the root cause of the problem. Used to have the same problems as a young lad when moving data between the IBM mainframe and VAX.
GLH
-----Original Message----- From: Ralph Mitchell [mailto:ralphmitchell at gmail.com] Sent: Tuesday, February 28, 2006 11:02 PM To: hobbit at hswn.dk Subject: Re: [hobbit] Moving to new Hobbit server
On 2/28/06, ZanDAhaR <allan at zandahar.net> wrote:
Ok cool so basically the graphs stop updating if theyre NQR after the move, I guess thats pretty effortless to test, move a host onto the new box and come back in a day and see what happens I suppose.
I think you could find out sooner than that - just copy an rrd over to the new system and try to dump it:
rrdtool dump datafile.rrdIf you get XML, it's probably all good. If it complains, you need to do the export at the original host and import it on the new host. I just copied an rrd over from a Sparc system to an x86 system, and rrdtool dump tells me:
*** glibc detected *** double free or corruption (out): 0x0804a170
Abort
whereas doing the same with a locally-created rrd I get a stream of xml. Likewise, going back to the original system I can dump the original file just fine.
Ralph Mitchell
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk