[rdiff-backup-users] rdiff-backup process gone wild - any diagnostics I should do before killing it?

[Willard Korfhage]

My nightly backup normally takes less than 15 minutes, but I have an 
rdiff-backup process that has now been running for 17 hours. It is very 
busy, but I have no idea what it is doing. When I do strace, I got 
something like this:

read(5, 
"\0\0\1\272D\3667\252|\201\1\211\303\370\0\0\1\340\7\354\201\0\0\257\t\2037\350\374\v;\1\343"..., 
131072) = 131072
write(7, 
"v\fA\241\332\0323\7\332Ttw9\275,\206\325-e#\362\233&>\332v7e\221\224\227%\314"..., 
4096) = 4096
write(7, "\272 
\225P\262\367\240\\=U\324\323\373\244\355\32\266o_H\327\376\244\231Z\224!\261\201\332\37\366q"..., 
131072) = 131072
read(5, 
"\0\0\1\272D\366>#E\327\1\211\303\370\0\0\1\340\7\354\201\300\n1=\221\223a\21=\2215\211"..., 
131072) = 131072
write(7, 
"address@hidden 
\324NV\0-\321\323\31\200\324\373\26s"..., 4096) = 4096
write(7, 
"\16A\266\216\34EL\r\22\227\340\364\360\303\255\226\225\177\270\201\331p\333\307a/\36\271\242\247&'B"..., 
126976) = 126976
read(5, 
"\0\0\1\272D\366?G\334+\1\211\303\370\0\0\1\275\7\354\201\200\5!=\217\375S\201\2\0\1\v"..., 
131072) = 131072
write(7, 
"\336\260\322W\f>'\200\4\\!\372\26*v\211g\307\217\274\36\346\230\330\244\31\347/|\265\266\370/"..., 
4096) = 4096
write(7, 
"eY2\307A\304wp\300\302Hy9\363\202Y\355m\205\362)YUo\16\\g\35\274\270V\262\10"..., 
126976) = 126976
read(5, 
"\0\0\1\272D\366D\203E\327\1\211\303\370\0\0\1\340\7\354\201\0\0\326\204+\217q\3\251\234A\310"..., 
131072) = 131072

spewing by as fast as the screen can scroll. I can't tell what it is 
reading and writing to. It doesn't seem to be hurting anything right 
now, beyond taking up 100% of one CPU core, so I'll leave it run for a 
bit in case a developer wants me to do something to it for information 
before I kill it.