[Date Prev][Date Next] [Chronological] [Thread] [Top]

Deferring operation & malloc crash in 2.1.22



Hello there,

So I think that I have seen all the previous posts on the list related to "deferred operation" errors showing up in the logs but none really gave me (a non-programmer) any clue on how to troubleshoot or otherwise correct this problem.

I am suffering this issue currently on my production servers (I have 2 (1 master, 1 replica), they are Mac OS X 10.2.6 with openldap 2.1.22 and LDBM with BDB 4.1.25 running on 2 Apple XServers with 2 Gb RAM each).

I get routinely 20,000 of these errors in a 24 hour period.

Is there some what that I can track down what is the cause of these? I don't see High CPU, tons of I/O, Swapping, etc on the machines. They simply handle a steady stream of requests (usually quickly).

Last night the master server actually crashed, the last entry in the logs showed (GMT timezone):

( you can see a deferring operation error as well.... then the malloc error then of course nothing!)

Oct 16 08:03:24 vars slapd[20991]: deferring operation
Oct 16 08:03:24 vars slapd[20991]: conn=78130 op=1 UNBIND
Oct 16 08:03:24 vars slapd[20991]: conn=78130 fd=30 closed
Oct 16 08:03:30 vars slapd[20991]: conn=78131 fd=30 ACCEPT from IP=17.128.115.151:1614 (IP=0.0.0.0:389)
Oct 16 08:03:30 vars slapd[20991]: conn=78131 op=0 BIND dn="" method=128
Oct 16 08:03:30 vars slapd[20991]: conn=78131 op=0 RESULT tag=97 err=0 text=
Oct 16 08:03:30 vars slapd[20991]: deferring operation
Oct 16 08:03:30 vars slapd[20991]: conn=78131 op=1 UNBIND
Oct 16 08:03:30 vars slapd[20991]: conn=78131 fd=30 closed
Oct 16 08:03:32 vars slapd[20991]: ldbm: ==> User-specified malloc function returned NULL
Oct 16 08:03:33 vars last message repeated 42 times
Oct 16 08:03:33 vars slapd[20991]: str2entry: entry -1 has invalid DN "appleDSID=706209,ou=GroupMembers,ou=groups,o=Apple Computer"
Oct 16 08:03:33 vars slapd[20991]: ldbm: ==> User-specified malloc function returned NULL
Oct 16 08:03:33 vars last message repeated 130 times


nothing after this.

I restarted the server and it is running OK now....

ANY THOUGHTS/SUGGESTIONS on what to tune, logging that I can turn on to capture more information, etc is very appreciated!

my configuration file looks like this:

#
# See slapd.conf(5) for details on configuration options.
# This file should NOT be world readable.
#
#
#
# Schema Includes
include         apple_openldap/etc/openldap/schema/core.schema
include         apple_openldap/etc/openldap/schema/cosine.schema
include         apple_openldap/etc/openldap/schema/inetorgperson.schema
# Custom Apple Schema
include         apple_openldap/etc/openldap/schema/apple_ist.schema

# Define global ACLs
# Give the replicator account the ability to update and everyone read access.


access to attr=userpassword
       by * auth
access to *
       by dn.base="cn=replicator,o=Apple Computer" write
       by * read

# Do not enable referrals until AFTER you have a working directory
# service AND an understanding of referrals.
#referral    ldap://root.openldap.org

# Put the database in read-only mode

readonly off

loglevel        256

gentlehup       on
idletimeout     300
sizelimit       2000
timelimit       300
password-hash   {SSHA}
allow bind_v2
threads         32

pidfile   /ngs/app/openldap/apple_openldap/etc/slapd.pid
argsfile  /ngs/app/openldap/apple_openldap/etc/slapd.args

replogfile /ngs/app/openldap/apple_openldap/var/openldap-slurp/replogfile.log

# Load dynamic backend modules:
# modulepath    /usr/local/libexec/openldap
# moduleload    back_bdb.la
# moduleload    back_ldap.la
# moduleload    back_ldbm.la
# moduleload    back_passwd.la
# moduleload    back_shell.la

# Set up the databases

database ldbm
suffix "ou=Groups, o=Apple Computer"
subordinate "o=Apple Computer"
dbcachesize 400000000
cachesize 400000000
rootdn "cn=Manager,o=Apple Computer"
replica host=ldap2p-grp.apple.com:389
binddn="cn=replicator,o=Apple Computer"
bindmethod=simple
credentials=xxx
directory /ngs/app/openldap/apple_openldap/var/openldap-data/db/groups
index objectClass eq
index appledsid eq
index cn eq,sub
index givenName eq,sub
index mail eq,sub
index sn eq,sub
index telephonenumber eq,sub
index appleGroupID eq
index appleUniqueMember eq
index appleGrpMbrEmail eq,sub


database ldbm
suffix "ou=Externals, o=Apple Computer"
subordinate "o=Apple Computer"
dbcachesize 400000000
cachesize 400000000
rootdn "cn=Manager,o=Apple Computer"
replica host=ldap2p-ext.apple.com:389
binddn="cn=Replicator,o=Apple Computer"
bindmethod=simple
credentials=xxx
directory /ngs/app/openldap/apple_openldap/var/openldap-data/db/externals
index objectClass eq
index appledsid eq
index cn eq,sub
index givenName eq,sub
index mail eq,sub
index sn eq,sub
index telephonenumber eq,sub


database ldbm
suffix "o=Apple Computer"
dbcachesize 300000000
cachesize 300000000
rootdn "cn=Manager,o=Apple Computer"
rootpw {SSHA}Kw24tUi4J4Um62aN1Gu/WJIHrxKTntlS
replica host=ldap2p.apple.com:389
binddn="cn=replicator,o=Apple Computer"
bindmethod=simple
credentials=xxx
directory /ngs/app/openldap/apple_openldap/var/openldap-data/db/apple
index objectClass eq
index appledsid eq
index cn eq,sub
index givenName eq,sub
index mail eq,sub
index sn eq,sub
index telephonenumber eq,sub
index l eq,sub
index c eq,sub
index applefloor eq,sub
index buildingname eq,sub
index description eq,sub
index st sub