Difference between revisions of "pgpool-II 3.7 development"

From pgpool Wiki
Jump to: navigation, search
Line 1: Line 1:
Last update: 2016/11/24 (gray items are done, yellow items are behind schedule)
+
Last update: 2016/12/07 (gray items are done, yellow items are behind schedule)
  
 
{| class="wikitable"
 
{| class="wikitable"
Line 10: Line 10:
 
! Status
 
! Status
 
|-
 
|-
| 5
+
| 1
 
| pgpool-hackers:1034
 
| pgpool-hackers:1034
 
| 2015/09/03
 
| 2015/09/03
Line 18: Line 18:
 
| need to assign, postpone to 3.7
 
| need to assign, postpone to 3.7
 
|- style="background:white"
 
|- style="background:white"
| 10
+
| 2
 
| Make PCP set configuration command public  
 
| Make PCP set configuration command public  
 
| 2016/2/15
 
| 2016/2/15
Line 26: Line 26:
 
| Low priority
 
| Low priority
 
|-
 
|-
| 14
+
| 3
 
| Allow to use multiple pgpool-II instances with in-memory query cache enabled
 
| Allow to use multiple pgpool-II instances with in-memory query cache enabled
 
| 2016/3/4
 
| 2016/3/4
Line 35: Line 35:
 
|-
 
|-
 
|-
 
|-
| 15
+
| 4
 
| Health-check timeout for each backend node
 
| Health-check timeout for each backend node
 
| 2016/3/4
 
| 2016/3/4
Line 43: Line 43:
 
| Low priority
 
| Low priority
 
|-
 
|-
| 49
+
| 5
 
| pgpool routes queries with "SET TRANSACTION READ ONLY" to master
 
| pgpool routes queries with "SET TRANSACTION READ ONLY" to master
 
| 2016/5/18
 
| 2016/5/18
Line 51: Line 51:
 
| low priority
 
| low priority
 
|-
 
|-
| 54
+
| 6
 
| testing of new failover code
 
| testing of new failover code
 
| 2016/6/6
 
| 2016/6/6
Line 59: Line 59:
 
| assigned (Not started yet, plan to start after #04)
 
| assigned (Not started yet, plan to start after #04)
 
|-
 
|-
| 74
+
| 7
 
| packet kind of backend 1 ['3'] does not match with master/majority nodes packet kind ['2']
 
| packet kind of backend 1 ['3'] does not match with master/majority nodes packet kind ['2']
 
| 2016/8/15
 
| 2016/8/15
Line 67: Line 67:
 
| <del>[3.6 beta]</del> In progress
 
| <del>[3.6 beta]</del> In progress
 
|-
 
|-
| 79
+
| 8
 
| watchdog can start even in case VIP is brought up in other nodes.
 
| watchdog can start even in case VIP is brought up in other nodes.
 
| 2016/8/23
 
| 2016/8/23
Line 75: Line 75:
 
| assigned
 
| assigned
 
|-
 
|-
| 80
+
| 9
 
| watchdog can start even in case trusted servers are not reachable.
 
| watchdog can start even in case trusted servers are not reachable.
 
| 2016/8/23
 
| 2016/8/23
Line 83: Line 83:
 
| assigned
 
| assigned
 
|-
 
|-
| 81
+
| 10
 
| watchdogs with different VIP can exist in a cluster.
 
| watchdogs with different VIP can exist in a cluster.
 
| 2016/8/23
 
| 2016/8/23
Line 91: Line 91:
 
| assigned
 
| assigned
 
|-
 
|-
| 82
+
| 11
 
| Behavior when watchdog process is killed
 
| Behavior when watchdog process is killed
 
| 2016/8/23
 
| 2016/8/23
Line 99: Line 99:
 
| assigned
 
| assigned
 
|-
 
|-
| 84
+
| 12
 
| Changed behavior when all heartbeat LAN are disconnected
 
| Changed behavior when all heartbeat LAN are disconnected
 
| 2016/8/23
 
| 2016/8/23
Line 107: Line 107:
 
| assigned
 
| assigned
 
|-
 
|-
| 90
+
| 13
 
| pgpool fails to run failover script if the sever running primary pgpool node and postgresql server dies
 
| pgpool fails to run failover script if the sever running primary pgpool node and postgresql server dies
 
| 2016/9/6
 
| 2016/9/6
Line 115: Line 115:
 
| <del>[3.6 beta]</del>(same as #79)assigned  
 
| <del>[3.6 beta]</del>(same as #79)assigned  
 
|-
 
|-
| 97
+
| 14
 
| watchdog sometimes fails de-escalation
 
| watchdog sometimes fails de-escalation
 
| 2016/9/21
 
| 2016/9/21

Revision as of 05:37, 7 December 2016

Last update: 2016/12/07 (gray items are done, yellow items are behind schedule)

ID Registered/start Date Target Date Person in charge Overview Status
1 pgpool-hackers:1034 2015/09/03 - - add new child if it goes down need to assign, postpone to 3.7
2 Make PCP set configuration command public 2016/2/15 - Usama - Low priority
3 Allow to use multiple pgpool-II instances with in-memory query cache enabled 2016/3/4 http://pgpool.net/mediawiki/index.php/TODO#Allow_to_use_pg_rewind_in_online_recovery Low priority
4 Health-check timeout for each backend node 2016/3/4 http://pgpool.net/mediawiki/index.php/TODO#Health-check_timeout_for_each_backend_node Low priority
5 pgpool routes queries with "SET TRANSACTION READ ONLY" to master 2016/5/18 2016/6/30 t-ishii http://www.pgpool.net/mantisbt/view.php?id=195 low priority
6 testing of new failover code 2016/6/6 Ahsan assigned (Not started yet, plan to start after #04)
7 packet kind of backend 1 ['3'] does not match with master/majority nodes packet kind ['2'] 2016/8/15 2016/9/30 Tatsuo http://www.pgpool.net/mantisbt/view.php?id=231 [3.6 beta] In progress
8 watchdog can start even in case VIP is brought up in other nodes. 2016/8/23 Usama assigned
9 watchdog can start even in case trusted servers are not reachable. 2016/8/23 Usama assigned
10 watchdogs with different VIP can exist in a cluster. 2016/8/23 Usama assigned
11 Behavior when watchdog process is killed 2016/8/23 Usama assigned
12 Changed behavior when all heartbeat LAN are disconnected 2016/8/23 Usama assigned
13 pgpool fails to run failover script if the sever running primary pgpool node and postgresql server dies 2016/9/6 2016/9/30 Usama http://www.pgpool.net/mantisbt/view.php?id=234 [3.6 beta](same as #79)assigned
14 watchdog sometimes fails de-escalation 2016/9/21 Usama http://www.pgpool.net/mantisbt/view.php?id=249 assigned
98 Not proper behaviour of Pgpool when one node of cluster goes disconnect by down interface 2016/9/28 Usama http://www.pgpool.net/mantisbt/view.php?id=250 assigned
99 failover_command not executed 2016/9/28 Usama http://www.pgpool.net/mantisbt/view.php?id=251 assigned
109 pgpoolAdmin 3.6 release 2016/12/31 -