[pgpool-general: 5066] FATAL: unable to process backend response

[Ken Young]

Hey folks,

Using pgpool 3.4.3 with streaming replication (yes, I'm in the process of moving to 3.5).  I'm seeing this error - an SQL query attempts to create a table (which already exists) and the PostgreSQL server returns the proper error message.  However, pgpool raises this fatal message:

FATAL:  unable to process backend response
DETAIL:  invalid message kind sent by backend connection
DEBUG:  shmem_exit(2): 0 callbacks to make
DEBUG:  proc_exit(2): 0 callbacks to make
DEBUG:  exit(2)
DEBUG:  shmem_exit(-1): 0 callbacks to make
DEBUG:  proc_exit(-1): 0 callbacks to make

I tried executing the same SQL statement from pgAdmin, but could not generate the above error in the pgpool log.  So I'm not sure what the original client is doing differently.  Any ideas about how to avoid this error?  This is only a fatal error with pgpool - the PostgreSQL server doesn't appear to have any issues.  I'll try pgpool 3.5 to see if I see the same results.

The detailed pgpool.log  follows:

2016-10-17 23:00:08: pid 30584: LOG:  pool_send_and_wait: Error or notice message from backend: : DB node id: 0 backend pid: 30876 statement: "create table if not exists EAM_ENCRYPTION_KEYS (encryption_key varchar(2000) not null, creation_date timestamp primary key default now()) " message: "NOTICE:  relation "eam_encryption_keys" already exists, skipping
        "
2016-10-17 23:00:08: pid 30584: DEBUG:  detect error: kind: N
2016-10-17 23:00:08: pid 30584: DEBUG:  detect error: kind: N
2016-10-17 23:00:08: pid 30584: DEBUG:  detect error: kind: N
2016-10-17 23:00:08: pid 30584: DEBUG:  detect error: kind: N
2016-10-17 23:00:08: pid 30584: DEBUG:  reading backend data packet kind
2016-10-17 23:00:08: pid 30584: DETAIL:  backend:0 kind:'N'
2016-10-17 23:00:08: pid 30584: DEBUG:  reading backend data packet kind
2016-10-17 23:00:08: pid 30584: DETAIL:  backend:0 of 2 kind = 'N'
2016-10-17 23:00:08: pid 30584: DEBUG:  processing backend response
2016-10-17 23:00:08: pid 30584: DETAIL:  received kind 'N'(4e) from backend
2016-10-17 23:00:08: pid 30584: DEBUG:  reading string data
2016-10-17 23:00:08: pid 30584: DETAIL:  read all from pending data. po:2 len:80
2016-10-17 23:00:08: pid 30584: DEBUG:  reading backend data packet kind
2016-10-17 23:00:08: pid 30584: DETAIL:  backend:0 kind:'
2016-10-17 23:00:08: pid 30584: DEBUG:  reading backend data packet kind
2016-10-17 23:00:08: pid 30584: DETAIL:  backend:0 of 2 kind = '
2016-10-17 23:00:08: pid 30584: FATAL:  unable to process backend response
2016-10-17 23:00:08: pid 30584: DETAIL:  invalid message kind sent by backend connection
2016-10-17 23:00:08: pid 30584: DEBUG:  shmem_exit(2): 0 callbacks to make
2016-10-17 23:00:08: pid 30584: DEBUG:  proc_exit(2): 0 callbacks to make
2016-10-17 23:00:08: pid 30584: DEBUG:  exit(2)
2016-10-17 23:00:08: pid 30584: DEBUG:  shmem_exit(-1): 0 callbacks to make
2016-10-17 23:00:08: pid 30584: DEBUG:  proc_exit(-1): 0 callbacks to make
2016-10-17 23:00:08: pid 30335: DEBUG:  reaper handler
2016-10-17 23:00:08: pid 30335: LOG:  child process with pid: 30584 exits with status 512
2016-10-17 23:00:08: pid 30335: LOG:  fork a new child process with pid: 30877




Ken Young
Ken.Young at sas.com
Data Management R&D
SAS Institute, Inc.
919-531-9319
http://www.sas.com<http://www.sas.com/>
SAS...  The Power to Know

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20161019/9dfe7439/attachment-0001.html>