[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
sigpipes on Solaris 8
From: |
Marco Marongiu |
Subject: |
sigpipes on Solaris 8 |
Date: |
Fri, 04 Mar 2005 14:19:04 +0100 |
User-agent: |
Debian Thunderbird 1.0 (X11/20050116) |
Hello
I am running a self-compiled 2.1.13 on a Solaris 8 platform. I used the
following line in configure:
./configure --prefix=/opt/cfengine-2.1.13 --with-pcre
--with-berkeleydb=/usr/local --with-openssl=/usr/local/ssl
9 times over 10 that I run: /var/cfengine/bin/cfagent
I get an output like this during the copy phase:
cfengine:: Received signal 13 (SIGPIPE) while doing [pre-lock-state]
cfengine:: Logical start time Fri Mar 4 13:17:11 2005
cfengine:: This sub-task started really at Fri Mar 4 13:17:11 2005
I've read this problem happens
<http://lists.gnu.org/archive/html/help-cfengine/2005-01/msg00090.html>,
and someone that solved it playing with SplayTime and similar things
<http://nakedape.cc/wiki/ApplicationNotes/CfEngine>, but I can't get
through and I'm stuck.
I also tried today's snapshot at about 13:00, but the problem is still
there...
My setup is as follows:
---cfservd.conf------------------------------------------------------
control:
domain = ( ourdomain.com )
IfElapsed = ( 0 )
AllowUsers = ( root )
cfrunCommand = ( /var/cfengine/bin/cfagent )
trustees = ( "127.0.0.1 10.39.17.16 10.39.17.40" )
TrustKeysFrom = ( ${trustees} )
admit:
/var/cfengine/repository ${trustees} root=${trustees}
---8<----------------------------------------------------------------
---update.conf-------------------------------------------------------
control:
domain = ( ourdomain.com )
sysadm = ( anaddress@of.ours )
syslog = ( on )
access = ( root )
policyhost = ( policy )
policyhostip = ( 10.39.17.16 )
smtpserver = ( smtp.of.ours )
IfElapsed = ( 0 )
SplayTime = ( 0 )
childlibpath = ( /usr/local/lib )
cfdir = ( /var/cfengine )
cfbin = ( ${cfdir}/bin )
cfinputs = ( ${cfdir}/inputs )
actionsequence = (
copy
)
copy:
!policy::
${cfbin} dest=${cfbin}
recurse=1
mode=500
owner=root
group=bin
type=checksum
server=${policyhost}
trustkey=true
ignore=CVS
exclude=*~
exclude=#*
${cfinputs} dest=${cfinputs}
recurse=1
mode=444
owner=root
group=bin
type=checksum
server=${policyhost}
trustkey=true
ignore=CVS
include=*.conf
---8<----------------------------------------------------------------
---cfagent.conf------------------------------------------------------
control:
domain = ( ourdomain.com )
sysadm = ( anaddress@of.ours )
syslog = ( on )
access = ( root )
policyhost = ( policy )
policyhostip = ( 10.39.17.16 )
smtpserver = ( smtp.of.ours )
IfElapsed = ( 0 )
SplayTime = ( 0 )
EmptyResolvConf = ( true )
IfElapsed = ( 0 )
SplayTime = ( 0 )
AddInstallable = ( reload_crontab restart_xntpd )
childlibpath = ( /usr/local/lib )
randmin = ( RandomInt(10,50) )
repo = ( /var/cfengine/repository/config )
actionsequence = (
shellcommands.dump_crontab
editfiles
resolve
directories
copy
processes
shellcommands
tidy
)
[...OMMISIS...]
copy:
any::
${repo}/ntp.conf dest=/etc/inet/ntp.conf
mode=400
owner=root
group=other
recurse=0
type=checksum
define=restart_xntpd
server=${policyhost}
trustkey=true
[...OMMISIS...same things for other files...OMMISIS...]
---8<----------------------------------------------------------------
I can't imagine any solution but to give up with remote copy and set up
an NFS share. Any help? clues? screwdrivers?
Thanks in advance
Ciao
--bronto
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- sigpipes on Solaris 8,
Marco Marongiu <=