| From: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
|---|---|
| To: | Greg Stark <stark(at)mit(dot)edu> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: measuring lwlock-related latency spikes |
| Date: | 2012-03-31 14:21:48 |
| Message-ID: | CA+U5nMLAFGe5PEm79uRBrPFQvorKbkVPyfhAHbRGyAOg7MX=2Q@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Sat, Mar 31, 2012 at 1:58 PM, Greg Stark <stark(at)mit(dot)edu> wrote:
> On Sat, Mar 31, 2012 at 4:41 AM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>> But I didn't think we were ever supposed to hold content
>> locks for that long.
>
> Isn't that lock held while doing visibility checks? What about I/O
> waiting for a clog page to be read?
So what we should be logging is the list of lwlocks held when the lock
wait occurred.
That would differentiate call paths somewhat better than just looking
at the current lock request.
--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2012-03-31 14:56:36 | pgsql: Add PGDLLIMPORT to ScanKeywords and NumScanKeywords. |
| Previous Message | Dobes Vandermeer | 2012-03-31 13:37:14 | Re: HTTP Frontend? (and a brief thought on materialized views) |