[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [rdiff-backup-users] Result too large [Errno 34] on OS X
From: |
Ilari Scheinin |
Subject: |
Re: [rdiff-backup-users] Result too large [Errno 34] on OS X |
Date: |
Thu, 27 Sep 2007 11:34:26 +0300 |
I was also getting the 'Result too large' error, and found Andrew's
suggestion of running rdiff-backup with the option -v6. That way I
was able to pinpoint the error to this file:
Regular copying ('Applications', 'Remote Buddy.app', 'Contents',
'PlugIns', 'NetNewsWire.rbscriptedbehaviour', 'Contents',
'Resources', 'English.lproj', 'Localizable.strings') to /Volumes/
Turva/Applications/Remote Buddy.app/Contents/PlugIns/
NetNewsWire.rbscriptedbehaviour/Contents/Resources/English.lproj/
rdiff-backup.tmp.4177
I excluded /Applications/Remote Buddy.app and I am currently running
the backup again, and hopefully it will now finish without errors. I
also tried to look at the file, but got these errors:
$ more Localizable.strings
read error (press RETURN)
$ cat Localizable.strings
cat: Localizable.strings: Result too large
I have never seen those errors before. I tried some googling, but did
not find anything that would be truly eye-opening for my level
knowledge of low level file system access.
This mailing list thread gives the following explanation:
"This means that the accept() system call returned a file descriptor
that is larger than FD_SETSIZE. This means you have (for some reason)
lots and lots of milter threads active."
http://lists.roaringpenguin.com/pipermail/mimedefang/2006-June/
030492.html
It suggests that sendmail is a possible cause, but I am not running it.
This page says it is an an error from a Unix system call:
http://caml.inria.fr/pub/docs/manual-caml-light/node17.html
Does this ring a bell, anyone?
Ilari