You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<tr><td>BDR</td><td>5.8.0</td><td>Improve pgd_bench error message related to CAMO</td><td></td></tr>
19
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>The <code>--summary</code> and <code>--options</code> flags for <code>pgd node show</code> CLI command.</summary><hr/><p>Add the <code>--summary</code> and <code>--options</code> flags to <code>pgd node show</code> command to filter the output of the <code>pgd node show</code> command.
20
+
This also maintains symmetry with other <code>show</code> commands.</p>
21
+
</details></td><td></td></tr>
22
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>More GUCs verfied in <code>pgd cluster verify</code> CLI command.</summary><hr/><p>Add the <code>bdr.lock_table_locking</code> and <code>bdr.truncate_locking</code> GUCs to list of GUCs verfied in <code>pgd cluster verify</code> command.</p>
23
+
</details></td><td></td></tr>
24
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Ensure commit scope logic only runs on data nodes</summary><hr/><p>While commit scope processing does not have any direct affect on
25
+
non-data nodes, by skipping it altogether we can avoid potentially
26
+
confusing error messages.</p>
27
+
</details></td><td></td></tr>
22
28
<tr><td>BDR</td><td>5.8.0</td><td>Added `bdr.galloc_chunk_info()` function to simplify sequences</td><td></td></tr>
29
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>Improve the CLI debug messages.</summary><hr/><p>Improve the formating of the log messages to be more readable and symmetrical with Postgres log messages.</p>
30
+
</details></td><td></td></tr>
31
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>New column for <code>pgd cluster verify --settings</code> CLI command output.</summary><hr/><p>Add the <code>recommended_value</code> column to the result of the <code>pgd cluster verify --settings</code> command.
32
+
The column will not be displayed in tabular output but will be displayed in JSON output.</p>
33
+
</details></td><td></td></tr>
34
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>Display sorted output for CLI.</summary><hr/><p>The output for the commands with tabular output will be sorted by the resource name.
35
+
For the commands that display more than one resource, the output will be sorted by each resource column in order.</p>
36
+
</details></td><td></td></tr>
37
+
<tr><td>BDR</td><td>5.8.0</td><td>Improve pgd_bench error message related to CAMO</td><td></td></tr>
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Add safety nets for CDC failover</summary><hr/><p>CDC failover now has additional safety nets to ensure that the consumer
45
+
does not start replication from a node that is not the creator of the
46
+
replication slot. This is to prevent data loss or duplicate transactions.
47
+
The changes also add additional checks to ensure that the consumer does
48
+
not start replication from a node that does not have the required WAL
49
+
files to decode the transactions that are missing on the consumer but
50
+
were included in the initial snapshot that the newly joined node had
51
+
obtained (physical or logical).</p>
52
+
</details></td><td></td></tr>
53
+
<tr><td>BDR</td><td>5.8.0</td><td>Update to non-unique index needs to compare tuple on lookup to ensure it is the same</td><td>43523, 43802, 45244, 47815, 48007</td></tr>
54
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Use primary_conninfo GUC instead of trying to read from shared memory</summary><hr/><p>The shared memory information may get obfuscated and hence become
55
+
useless. So instead of reading from there, simply use the GUC which is
56
+
now available on all supported PG versions.</p>
57
+
</details></td><td>41776</td></tr>
58
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>Fix the CLI <code>pgd cluster show</code> command issues on a degraded cluster.</summary><hr/><p>The <code>pgd cluster show</code> command failed with error for clock drift if only one node is up and running in a N node cluster.
59
+
The command is fixed to return valid output for other components viz., <code>health</code> and <code>summary</code> while reporting a valid error for <code>clock-drift</code>.</p>
60
+
</details></td><td></td></tr>
61
+
<tr><td>PGD CLI</td><td>5.8.0</td><td><details><summary>Fix the CLI <code>pgd node show</code> command issue if a non-existent node is specified.</summary><hr/><p>The <code>pgd node show</code> command crashed if a non-existent node is specified to the command.
62
+
The command is fixed to fail gracefully with appropriate error message.</p>
63
+
</details></td><td></td></tr>
64
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Prevent segfault in bdr.taskmgr_set_leader</summary><hr/><p>Add a note in test 00200 to at least give a hint about which
65
+
function may be to blame.</p>
66
+
</details></td><td></td></tr>
67
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Replacing TRUNCATEs with DELETEs from all BDR catalogs on local node drop (#3685)</summary><hr/><p>This is to avoid deadlock in bdr_init_physical IF customer happens to run monitoring
68
+
queries during node joining / cleaning unwanted source node data.</p>
69
+
</details></td><td>46952</td></tr>
70
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Ensure new joiner processes watermark message in the CATCHUP phase</summary><hr/><p>Setting nci->min_lsn to XactLastCommitEnd of watermark message Tx to ensure
71
+
CATCHUP phase finishes on new joiner only after watermark is processed.</p>
72
+
</details></td><td></td></tr>
73
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Fixed deadlock issue in bdr_init_physical.</summary><hr/><p>Fixed deadlock between bdr_init_physical cleaning unwanted node data and concurrent monitoring queries.</p>
74
+
</details></td><td>46952</td></tr>
75
+
<tr><td>BDR</td><td>5.8.0</td><td><details><summary>Fixed new cluster node consistency issue.</summary><hr/><p>Fixed an issue when new node joining the cluster finishes CATCHUP phase before getting its replication progress against all data nodes. This may cause new node being out of sync with the cluster.</p>
- relnote: Replacing TRUNCATEs with DELETEs from all BDR catalogs on local node drop (#3685)
84
+
component: BDR
85
+
details: |
86
+
This is to avoid deadlock in bdr_init_physical IF customer happens to run monitoring
87
+
queries during node joining / cleaning unwanted source node data.
88
+
jira: BDR-6313
89
+
addresses: "46952"
90
+
type: Bug fix
91
+
impact: Medium
92
+
93
+
- relnote: Ensure new joiner processes watermark message in the CATCHUP phase
94
+
component: BDR
95
+
details: |
96
+
Setting nci->min_lsn to XactLastCommitEnd of watermark message Tx to ensure
97
+
CATCHUP phase finishes on new joiner only after watermark is processed.
98
+
jira: BDR-6397
99
+
addresses: ""
100
+
type: Bug fix
101
+
impact: Medium
102
+
103
+
- relnote: Improve the CLI debug messages.
104
+
details: |
105
+
Improve the formating of the log messages to be more readable and symmetrical with Postgres log messages.
106
+
component: PGD CLI
107
+
jira: BDR-6101
108
+
type: Enhancement
109
+
impact: Medium
110
+
111
+
- relnote: The `--summary` and `--options` flags for `pgd node show` CLI command.
112
+
details: |
113
+
Add the `--summary` and `--options` flags to `pgd node show` command to filter the output of the `pgd node show` command.
114
+
This also maintains symmetry with other `show` commands.
115
+
component: PGD CLI
116
+
jira: BDR-6145
117
+
addresses: ""
118
+
type: Enhancement
119
+
impact: High
120
+
121
+
- relnote: More GUCs verfied in `pgd cluster verify` CLI command.
122
+
details: |
123
+
Add the `bdr.lock_table_locking` and `bdr.truncate_locking` GUCs to list of GUCs verfied in `pgd cluster verify` command.
124
+
component: PGD CLI
125
+
jira: BDR-5308
126
+
addresses: ""
127
+
type: Enhancement
128
+
impact: High
129
+
130
+
- relnote: New column for `pgd cluster verify --settings` CLI command output.
131
+
details: |
132
+
Add the `recommended_value` column to the result of the `pgd cluster verify --settings` command.
133
+
The column will not be displayed in tabular output but will be displayed in JSON output.
134
+
component: PGD CLI
135
+
jira: BDR-5308
136
+
addresses: ""
137
+
type: Enhancement
138
+
impact: Medium
139
+
140
+
- relnote: Display sorted output for CLI.
141
+
details: |
142
+
The output for the commands with tabular output will be sorted by the resource name.
143
+
For the commands that display more than one resource, the output will be sorted by each resource column in order.
144
+
component: PGD CLI
145
+
jira: BDR-6094
146
+
addresses: ""
147
+
type: Enhancement
148
+
impact: Medium
149
+
150
+
- relnote: Fix the CLI `pgd cluster show` command issues on a degraded cluster.
151
+
details: |
152
+
The `pgd cluster show` command failed with error for clock drift if only one node is up and running in a N node cluster.
153
+
The command is fixed to return valid output for other components viz., `health` and `summary` while reporting a valid error for `clock-drift`.
154
+
component: PGD CLI
155
+
jira: BDR-6135
156
+
addresses: ""
157
+
type: Bug Fix
158
+
impact: High
159
+
160
+
- relnote: Fix the CLI `pgd node show` command issue if a non-existent node is specified.
161
+
details: |
162
+
The `pgd node show` command crashed if a non-existent node is specified to the command.
163
+
The command is fixed to fail gracefully with appropriate error message.
164
+
component: PGD CLI
165
+
jira: BDR-6292
166
+
addresses: ""
167
+
type: Bug Fix
168
+
impact: High
169
+
27
170
- relnote: Improve pgd_bench error message related to CAMO
28
171
description: |
29
172
If executed with `--mode=camo`, but the provided test script is not wrapped
30
173
in an explicit transaction, pgd_bench will not be able to retrieve the
31
174
expected `transaction_id` value. Improve the emitted error message to
32
175
contain a hint about a possible missing transaction.
33
176
component: BDR
34
-
impact: Minor
177
+
impact: Low
35
178
jira: BDR-6411
36
179
type: Enhancement
180
+
181
+
- relnote: Fixed deadlock issue in bdr_init_physical.
182
+
component: BDR
183
+
details: |
184
+
Fixed deadlock between bdr_init_physical cleaning unwanted node data and concurrent monitoring queries.
185
+
jira: BDR-6313
186
+
addresses: 46952
187
+
type: Bug Fix
188
+
impact: Low
189
+
190
+
- relnote: Fixed new cluster node consistency issue.
191
+
component: BDR
192
+
details: |
193
+
Fixed an issue when new node joining the cluster finishes CATCHUP phase before getting its replication progress against all data nodes. This may cause new node being out of sync with the cluster.
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?v=4&size=40){kind=avatar}
0 commit comments