[Logcheck-devel] Bug#311216: logcheck: mails "egrep: Invalid content of \{\}"
John A. Martin
jam at jamux.com
Mon May 30 18:28:16 UTC 2005
>>>>> "Jamie" == Jamie L Penman-Smithson
>>>>> "Re: Bug#311216: logcheck: mails "egrep: Invalid content of \{\}""
>>>>> Mon, 30 May 2005 17:58:30 +0100
Jamie> Hi John,
>> Mail sent by logcheck
>>
>> ,----
>> From: root at lists.jamux.org (Cron Daemon)
>> Subject: Cron <logcheck at lists> if [ -x /usr/sbin/logcheck ]; then
>> nice -n10 /usr/sbin/logcheck; fi
>> To: root at lists.jamux.org
>> Date: Sun, 29 May 2005 18:02:09 -0400 (EDT)
>>
>> egrep: Invalid content of \{\}
>> `----
>>
>> egrep: Invalid content of \{\} `----
Jamie> I can't reproduce this at all, which makes me think there's
Jamie> something different on your system that is causing this
Jamie> bug.
Hmm... I have several boxen each doing the same with yesterday's
upgrade. It looks like the problem occurs on boxen with mixed
stable/unstable but not on Sid boxen.
Jamie> ^\w{3} [ :0-9]{11} [._[:alnum:]-]+ squid\[[0-9]+\]:
Jamie> WARNING: found whitespace in HTTP header name {Cache
Jamie> Control: no-cache}$
Jamie> With this rule, egrep parses it fine:
Jamie> D: [1117469713] cleanchecked - dir - /tmp/logcheck.fb7dSc/ignore/spamd
Jamie> D: [1117469713] cleanchecked - dir - /tmp/logcheck.fb7dSc/ignore/squid
Jamie> D: [1117469714] cleanchecked - dir - /tmp/logcheck.fb7dSc/ignore/ssh
Jamie> However, since there is nothing to lose by escaping the
Jamie> {}'s, I've modified that rule in CVS.
It looks like maybe grep-2.4.2-3 gives the problem while
grep-2.5.1.ds1-4 does not. That is not tested, just something that
differs between the boxen showing the problem and one that doesn't. I
didn't notice before because the Sid box doesn't normally do logcheck.
jam
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 154 bytes
Desc: not available
Url : http://lists.alioth.debian.org/pipermail/logcheck-devel/attachments/20050530/50fefbb3/attachment.pgp
More information about the Logcheck-devel
mailing list