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

cannot allocate 0 byte



Today, one of our OpenLDAP syncrepl consumers died (again) with "ch_malloc of 0 bytes failed".
Are there any known issues in 2.4.15 leading to that? It's somewhat annoying since it's the replication out mail servers are using.

The last lines before the server dies are:

Nov  2 16:38:09 kinzig slapd[17902]: conn=570825 fd=215 closed (connection lost) 
Nov  2 16:38:12 kinzig slapd[17902]: do_syncrep2: cookie=rid=001,csn=20091102153812.781434Z#000000#000#000000 
Nov  2 16:38:12 kinzig slapd[17902]: syncrepl_entry: rid=001 LDAP_RES_SEARCH_ENTRY(LDAP_SYNC_MODIFY) 
Nov  2 16:38:12 kinzig slapd[17902]: syncrepl_entry: rid=001 be_search (0) 
Nov  2 16:38:12 kinzig slapd[17902]: syncrepl_entry: rid=001 cn=<CENSORED>
Nov  2 16:38:12 kinzig slapd[17902]: slap_queue_csn: queing 0xa296a00 20091102153812.781434Z#000000#000#000000 
Nov  2 16:38:12 kinzig slapd[17902]: slap_graduate_commit_csn: removing 0x11ac34c0 20091102153812.781434Z#000000#000#000000 
Nov  2 16:38:12 kinzig slapd[17902]: syncrepl_entry: rid=001 be_modrdn (0) 
Nov  2 16:38:12 kinzig slapd[17902]: slap_queue_csn: queing 0xa296a00 20091102153812.781434Z#000000#000#000000 
Nov  2 16:38:12 kinzig slapd[17902]: ch_malloc of 0 bytes failed