- Notifications
You must be signed in to change notification settings - Fork28
Commit5db4fdc
committed
Further second thoughts about idle_session_timeout patch.
On reflection, the order of operations in PostgresMain() is wrong.These timeouts ought to be shut down before, not after, we do thepost-command-read CHECK_FOR_INTERRUPTS, to guarantee that anytimeout error will be detected there rather than at some ill-definedlater point (possibly after having wasted a lot of work).This is really an error in the original idle_in_transaction_timeoutpatch, so back-patch to 9.6 where that was introduced.1 parent0f8977b commit5db4fdc
1 file changed
+12
-10
lines changedLines changed: 12 additions & 10 deletions
Original file line number | Diff line number | Diff line change | |
---|---|---|---|
| |||
4269 | 4269 |
| |
4270 | 4270 |
| |
4271 | 4271 |
| |
4272 |
| - | |
| 4272 | + | |
| 4273 | + | |
| 4274 | + | |
| 4275 | + | |
| 4276 | + | |
| 4277 | + | |
| 4278 | + | |
| 4279 | + | |
| 4280 | + | |
| 4281 | + | |
| 4282 | + | |
| 4283 | + | |
4273 | 4284 |
| |
4274 | 4285 |
| |
4275 | 4286 |
| |
| |||
4280 | 4291 |
| |
4281 | 4292 |
| |
4282 | 4293 |
| |
4283 |
| - | |
4284 |
| - | |
4285 |
| - | |
4286 |
| - | |
4287 |
| - | |
4288 |
| - | |
4289 |
| - | |
4290 |
| - | |
4291 |
| - | |
4292 | 4294 |
| |
4293 | 4295 |
| |
4294 | 4296 |
| |
|
0 commit comments
Comments
(0)