On Wed, Mar 19, 2008 at 10:30:00AM -0400, Damian Cunniff wrote: ... > Recently however it continues to increase in the amount of time that it > is taking to backup. It has reached the point of taking as much as 8-10 > hours to complete.
I'm just guessing here, but this seems like what would happen if rsync was running out of memory. Have you watched what "mem" reports, during a run? > I'm stumped. What could possibly be causing this problem? Why would an > otherwise seemingly healthy setup be malfunctioning here? Should I fsck > the RAID array in hopes that that may be the trouble? rsync can really thrash hardware, because it milks every last drop of available performance to run as quickly as possible. It keeps the disks very busy. Indeed, you should fsck the array, and you should also thrash your hardware with equally hard but simpler-to-diagnose tests. I write my own simple tests that rapidly fill and read the disk, but I imagine there are some good tools out there for download that do a more complete job. For example, some types of USB2 drives have been known to lock up after tens of gigabytes have moved. That is not your problem here, but it shows the kinds of subtle flaws you sometimes have to contend with. Worse yet, no piece of hardware is 100% tested, and sometimes individual devices will exhibit flaws under extreme stress. So I would check to make sure there is plenty of memory, then I would find some open source tools to thrash the hardware and look for flaws. If nothing is found, then watching rsync do its work (keeping track of files as they move, looking at logs as they are built) may be your only recourse. When you reach that point, share some information about your hardware and software (kernel used, version of rsync, which controllers and disks, etc) and some of the statistics from the logs, and we might be able to help you with that difficult task. Keith -- Keith Lofstrom [EMAIL PROTECTED] Voice (503)-520-1993 KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon" Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs _______________________________________________ Dirvish mailing list [email protected] http://www.dirvish.org/mailman/listinfo/dirvish
