Tue Jan 16 06:08:06 PST 2007
- Previous message: [Slony1-general] sl_registry not created on upgrade
- Next message: [Slony1-general] sl_log_1 not cleaning out
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi,
we seem to have a problem with our replication cluster - the sl_log_1
table doesn't get cleaned out. Can anybody give some advice on how to resolve
the issue?
Details:
Output from test-slony-state.pl:
DSN: dbi:Pg:dbname=RPONLINE;host=localhost;port=5481
===========================
Rummage for DSNs
=============================
Query:
select p.pa_server, p.pa_conninfo
from "_RPO".sl_path p
where exists (select * from "_RPO".sl_subscribe s where
(s.sub_provider = p.pa_server or s.sub_receiver =
p.pa_server) and
sub_active = 't')
group by pa_server, pa_conninfo;
Tests for node 1 - DSN = dbname=RPONLINE host=10.1.16.1 port=5481 user=slony
password=vs17gs
========================================
pg_listener info:
Pages: 513
Tuples: 38312
Size Tests
================================================
sl_log_1 391410 13679500.000000
sl_log_2 0 0.000000
sl_seqlog 43234 5879820.000000
Listen Path Analysis
===================================================
No problems found with sl_listen
--------------------------------------------------------------------------------
Summary of event info
Origin Min SYNC Max SYNC Min SYNC Age Max SYNC Age
================================================================================
5 490546 637044 00:00:00 17 days 04:58:00 0
1 1525251 2017755 00:00:00 17 days 04:58:00 0
4 491105 637583 00:00:00 17 days 04:58:00 0
3 736904 737115 00:00:00 00:35:00 0
2 576284 721645 00:00:00 17 days 04:58:00 0
---------------------------------------------------------------------------------
Summary of sl_confirm aging
Origin Receiver Min SYNC Max SYNC Age of latest SYNC Age of eldest
SYNC
=================================================================================
1 2 2016597 2017756 00:00:00 00:33:00 0
1 3 1525251 1525251 17 days 04:58:00 17 days 04:58:00
1
1 4 2016597 2017756 00:00:00 00:33:00 0
1 5 2016535 2017756 00:00:00 00:35:00 0
2 1 721456 721645 00:00:00 00:33:00 0
2 3 576284 576284 17 days 04:58:00 17 days 04:58:00
1
2 4 721456 721645 00:00:00 00:33:00 0
2 5 721447 721645 00:00:00 00:35:00 0
3 1 736914 737115 00:00:00 00:33:00 0
3 2 736913 737114 00:00:00 00:33:00 0
3 4 736913 737114 00:00:00 00:33:00 0
3 5 736904 737115 00:00:00 00:35:00 0
4 1 637382 637583 00:00:00 00:33:00 0
4 2 637382 637583 00:00:00 00:33:00 0
4 3 491105 491105 17 days 04:59:00 17 days 04:59:00
1
4 5 637374 637583 00:00:00 00:35:00 0
5 1 636844 637044 00:00:00 00:35:00 0
5 2 636844 637044 00:00:00 00:35:00 0
5 3 490546 490546 17 days 04:59:00 17 days 04:59:00
1
5 4 636844 637044 00:00:00 00:35:00 0
------------------------------------------------------------------------------
Listing of old open connections
Database PID User Query Age
Query
================================================================================
Tests for node 4 - DSN = dbname=RPONLINE host=10.1.50.1 port=5481 user=slony
========================================
pg_listener info:
Pages: 513
Tuples: 38312
Size Tests
================================================
sl_log_1 391410 13679500.000000
sl_log_2 0 0.000000
sl_seqlog 43234 5879820.000000
Listen Path Analysis
===================================================
No problems found with sl_listen
--------------------------------------------------------------------------------
Summary of event info
Origin Min SYNC Max SYNC Min SYNC Age Max SYNC Age
================================================================================
5 490546 637044 00:00:00 17 days 04:59:00 0
1 1525251 2017756 00:00:00 17 days 04:58:00 0
4 491105 637583 00:00:00 17 days 04:59:00 0
3 736904 737115 00:00:00 00:35:00 0
2 576284 721645 00:00:00 17 days 04:58:00 0
---------------------------------------------------------------------------------
Summary of sl_confirm aging
Origin Receiver Min SYNC Max SYNC Age of latest SYNC Age of eldest
SYNC
=================================================================================
1 2 2016597 2017758 00:00:00 00:33:00 0
1 3 1525251 1525251 17 days 04:58:00 17 days 04:58:00
1
1 4 2016597 2017759 00:00:00 00:33:00 0
1 5 2016535 2017759 00:00:00 00:35:00 0
2 1 721456 721646 00:00:00 00:33:00 0
2 3 576284 576284 17 days 04:58:00 17 days 04:58:00
1
2 4 721456 721646 00:00:00 00:33:00 0
2 5 721447 721646 00:00:00 00:35:00 0
3 1 736914 737116 00:00:00 00:33:00 0
3 2 736913 737115 00:00:00 00:33:00 0
3 4 736913 737115 00:00:00 00:33:00 0
3 5 736904 737115 00:00:00 00:35:00 0
4 1 637382 637584 00:00:00 00:33:00 0
4 2 637382 637584 00:00:00 00:33:00 0
4 3 491105 491105 17 days 04:59:00 17 days 04:59:00
1
4 5 637374 637584 00:00:00 00:35:00 0
5 1 636844 637045 00:00:00 00:35:00 0
5 2 636844 637045 00:00:00 00:35:00 0
5 3 490546 490546 17 days 04:59:00 17 days 04:59:00
1
5 4 636844 637045 00:00:00 00:35:00 0
------------------------------------------------------------------------------
Listing of old open connections
Database PID User Query Age
Query
================================================================================
[Output from other nodes omitted as they are identical]
Node: 1 sl_log_1 tuples = 1.36795e+07 > 200000
================================================
Number of tuples in Slony-I table sl_log_1 is 1.36795e+07 which
exceeds 200000.
You may wish to investigate whether or not a node is down, or perhaps
if sl_confirm entries have not been propagating properly.
Node: 1 sl_seqlog tuples = 5.87982e+06 > 200000
================================================
Number of tuples in Slony-I table sl_seqlog is 5.87982e+06 which
exceeds 200000.
You may wish to investigate whether or not a node is down, or perhaps
if sl_confirm entries have not been propagating properly.
[Output from other nodes omitted as they are identical]
More information (this is the replication master):
RPONLINE=# select count(1) from "_RPO".sl_confirm;
count
-------
7059
(1 row)
RPONLINE=# select count(1) from "_RPO".sl_event;
count
--------
931492
(1 row)
RPONLINE=# select * from "_RPO".sl_status;
-[ RECORD 1 ]-------------+---------------------------
st_origin | 1
st_received | 2
st_last_event | 2018097
st_last_event_ts | 2007-01-16 14:54:44.99581
st_last_received | 2018096
st_last_received_ts | 2007-01-16 14:54:44.053113
st_last_received_event_ts | 2007-01-16 14:54:43.978249
st_lag_num_events | 1
st_lag_time | 00:00:01.245307
-[ RECORD 2 ]-------------+---------------------------
st_origin | 1
st_received | 3
st_last_event | 2018097
st_last_event_ts | 2007-01-16 14:54:44.99581
st_last_received | 1525251
st_last_received_ts | 2006-12-30 09:46:15.781741
st_last_received_event_ts | 2006-12-30 09:46:15.774831
st_lag_num_events | 492846
st_lag_time | 17 days 05:08:29.448725
-[ RECORD 3 ]-------------+---------------------------
st_origin | 1
st_received | 4
st_last_event | 2018097
st_last_event_ts | 2007-01-16 14:54:44.99581
st_last_received | 2018095
st_last_received_ts | 2007-01-16 14:54:41.435216
st_last_received_event_ts | 2007-01-16 14:54:40.964031
st_lag_num_events | 2
st_lag_time | 00:00:04.259525
-[ RECORD 4 ]-------------+---------------------------
st_origin | 1
st_received | 5
st_last_event | 2018097
st_last_event_ts | 2007-01-16 14:54:44.99581
st_last_received | 2018095
st_last_received_ts | 2007-01-16 14:54:41.075192
st_last_received_event_ts | 2007-01-16 14:54:40.964031
st_lag_num_events | 2
st_lag_time | 00:00:04.259525
RPONLINE=# select * from "_RPO".sl_event where ev_timestamp < '2007-01-01
00:00:00' order by ev_timestamp limit 1;
-[ RECORD 1 ]+---------------------------
ev_origin | 4
ev_seqno | 491105
ev_timestamp | 2006-12-30 09:46:07.122286
ev_minxid | 171751695
ev_maxxid | 171751696
ev_xip |
ev_type | SYNC
ev_data1 |
ev_data2 |
ev_data3 |
ev_data4 |
ev_data5 |
ev_data6 |
ev_data7 |
ev_data8 |
I've also checked that there are no old transactions hanging around. Any clue
Ciao,
Thomas
--
Thomas Pundt <thomas.pundt at rp-online.de> ---- http://rp-online.de/ ----
- Previous message: [Slony1-general] sl_registry not created on upgrade
- Next message: [Slony1-general] sl_log_1 not cleaning out
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list