[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
checksums misfiring
From: |
Tod Oace |
Subject: |
checksums misfiring |
Date: |
Wed, 10 Dec 2003 09:40:44 -0800 |
I'm having trouble with type=checksum copy:'s sporadically firing when
they shouldn't be. I'm using 2.1.0p1. Before I dig into the code, has
anyone else seen this? It happens on the UNIX platforms I'm supporting
and does not seem to be limited to any particular set of hosts or times
of day.
Here's one specific example. One of my Cfengine clients sends me this:
cfengine:jf22grid02: Image file /etc/securetty has a wrong MD5
checksum (should be copy of /var/cfengine/cvsexport/etc/securetty)
So I look at its /etc/securetty and see:
[root@jf22grid02 etc]# ls -l securetty
-r--r--r-- 1 root root 114 Jun 13 2001 securetty
[root@jf22grid02 etc]# openssl md5 securetty
MD5(securetty)= 9a3f3c442ed66695529e5e72c1a9bbe5
And there's no old version in its /var/cfengine/repository/. The
securetty on the Cfengine server looks like:
[root@nwlxsos02 etc]# ls -l securetty
-rw-r--r-- 1 root rpc 114 Dec 9 09:21 securetty
[root@nwlxsos02 etc]# openssl md5 securetty
MD5(securetty)= 9a3f3c442ed66695529e5e72c1a9bbe5
Same checksum. Different time-stamps, but that shouldn't matter because
I'm using type=checksum. I have ChecksumUpdates = ( on ) and
ChecksumDatabase = ( /var/cfengine/cfdb ) for client and server.
Ummm... Oh! Do I need separate database files for the client and server
checksum databases? Because the server machine is also a client. I'll
try separating them.
Anyway, if you've run into this problem I'd like to hear from you.
Thanks!
--
Tod Oace, Intel Corporation <tod@intel.com>
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- checksums misfiring,
Tod Oace <=