Support for more granular session lock behaviors [possible patch]

[ericmerrill]

I do programing with a large LMS, Moodle (moodle.org), and support has been added for memcached sessions to improve performance, but something we have noticed is that both lock max wait and lock expiration are tied to max_execution_time at the time of session_start(), which is not optimal for us (and I'm sure we're not alone).

I propose two new settings, memcached.sess_lock_max_wait and memcached.sess_lock_expire which allow the setting of the max time to wait for a lock before dying, and the time for a lock to expire, respectively.

I've done a first pass of the patch at:
https://github.com/merrill-oakland/php-memcached/compare/c10de3699faa3ecb4cb6861d6c92157e0c7e18ce...sessionlocks

Current behavior should be maintained with settings of 0 (or unset).
I've never written stuff for this project, so I followed the coding styles that I observed, but I may have missed something. I also need to look at the test system to see about adding tests, but the same tests pass/fail both before and after this patch.

Thoughts? Recommendations? Is this a possibility?

[ericmerrill]

In case any Moodlers find this, the related ticket is https://tracker.moodle.org/browse/MDL-42485

[mkoppanen]

Looks pretty ok to me, I'll review properly soon and merge

[mkoppanen]

Can you create a PR for this?

[ericmerrill]

I would be happy to... Although I don't know what PR stands for in this
context (and googling didn't get me anywhere).
On Oct 27, 2013 3:14 AM, "Mikko Koppanen" notifications@github.com wrote:

Can you create a PR for this?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/89#issuecomment-27164243
.

[ericmerrill]

Yeah it just dawned on me. I'll open it shortly.

head desk
On Oct 27, 2013 5:57 AM, "Eric Merrill" merrill@oakland.edu wrote:

I would be happy to... Although I don't know what PR stands for in this
context (and googling didn't get me anywhere).
On Oct 27, 2013 3:14 AM, "Mikko Koppanen" notifications@github.com
wrote:

Can you create a PR for this?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/89#issuecomment-27164243
.

[ericmerrill]

That's what I get for responding at 5am. Opened as #91

[mkoppanen]

#91 has been merged. Thanks!

[ericmerrill]

Excellent, thanks!

I see a pretty big uptick in activity in the project - is there a planned new release, or just when one happens :)

[mkoppanen]

I've had some free time lately, which means more time to fix bugs and answer enquiries. However, I cannot do releases at the moment as @andreiz is the only one listed as lead developer in PECL.

I will update changelogs etc at some point and prepare for another beta.

[andreiz]

MIkko, I added you as a lead on PECL memcached package.

On Sun, Oct 27, 2013 at 6:06 AM, Mikko Koppanen notifications@github.comwrote:

I've had some free time lately, which means more time to fix bugs and
answer enquiries. However, I cannot do releases at the moment as @andreizhttps://github.com/andreizis the only one listed as lead developer in PECL.

I will update changelogs etc at some point and prepare for another beta.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/89#issuecomment-27168663
.

[mkoppanen]

@andreiz Thanks!