| From: | Alexey Orlov <aporlov(at)gmail(dot)com> |
|---|---|
| To: | wenhui qiu <qiuwenhuifx(at)gmail(dot)com> |
| Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: Patch: Show queries of processes holding a lock |
| Date: | 2024-10-02 09:43:03 |
| Message-ID: | CALjmG4LLL4qTje9N7HQwLniRAuOOxSPqikspC7XrXuPRUnYCMg@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, Oct 1, 2024 at 11:45 AM wenhui qiu <qiuwenhuifx(at)gmail(dot)com> wrote:
>
> Hi Alexey Orlov
> Thank you for your work on this path,The lock information is recorded in detail,Easy to trace the lock competition at that time there is a detailed lock competition log,But I have a concern,Frequent calls to this function (pgstat_get_backend_current_activity) in heavy lock contention or high concurrency environments may cause performance degradation, especially when processes frequently enter and exit lock waits. Can you add a guc parameter to turn this feature on or off?After all communities for this parameter( log_lock_waits )default values set to on many people concern (https://commitfest.postgresql.org/49/4718/)
>
>
>
> Thanks
>
Yeah, agree, thank you. I just think changing the parameter type would
be nice too.
typedef enum
{
LOG_LOCK_WAIT_NONE = 0,
LOG_LOCK_WAIT_TERSE,
LOG_LOCK_WAIT_VERBOSE,
} LogLockWaitVerbosity;
LOG_LOCK_WAIT_NONE is "off", LOG_LOCK_WAIT_TERSE is "on",
LOG_LOCK_WAIT_VERBOSE enables writing the query to to the log.
I've attached a slightly modified patch to use the new log_lock_wait values.
--
Regards,
Alexey Orlov!
| Attachment | Content-Type | Size |
|---|---|---|
| v-2-0001-Show-queries-in-log_lock_wait.patch | text/x-patch | 5.7 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jim Jones | 2024-10-02 09:47:42 | Re: Psql meta-command conninfo+ |
| Previous Message | Fujii Masao | 2024-10-02 09:40:08 | Re: [BUG FIX]Connection fails with whitespace after keepalives parameter value |