# Lazy writing in the session, session-lock-ini, and bug #68331 - Thread: https://externals.io/message/78699 - Original archive: https://news-web.php.net/php.internals/78699 - Started by Damian Wadley on 2014-11-05T01:54:45+00:00 - Messages: 8 - Reply by email to internals@lists.php.net - More for agents: search with https://externals.io/search.md?q={query} · JSON API at https://externals.io/api · documentation at https://externals.io/llms.txt --- ## #78699 — Damian Wadley — 2014-11-05T01:54:45+00:00 URL: https://externals.io/message/78699#78699 https://bugs.php.net/bug.php?id=68331 I was hoping the submitter (or one of their coworkers who commented on it) would reach out to the list themselves to get more information since I don't know the whole situation. I did ask them to... Story thus far: Yasuo's session-lock-ini RFC [1] in February 2014 was (partially) approved - the relevant change being that it adds support for lazy writing. Optionally. There's a pull request with it that has NOT been merged; as far as I can tell it's because there was discussion after the vote raising problems and there was another RFC [2] (since withdrawn) to change the first one. I also vaguely remember there being merge problems between 5.6 and something? Unrelated and prior to all this was a commit [3] in August 2013 for request #17860 [4] that made session writes lazy for everyone. [see session.c] Not optionally. It went into 5.6. Apparently this caused problems for some people as they made 68331 a few days ago. I'm not entirely sure that the userland code couldn't be changed to deal with the lazy writing without requiring the crazy workaround they described, but whatever I don't know their codebase. I tried explaining that session-lock-ini is not yet merged, despite being approved and being listed under the PHP 5.6 heading at /rfc, and that the commit in question was entirely separate, to no avail. Meanwhile I did NAB the bug on the grounds that the change was intentional. So what's up with the RFC and its code? Is it still happening for 5.6 (too late?) or 5.7 (I lost track if we're doing that) or 7? Any suggestions for what I can tell the folks concerned? -Damian [1] https://wiki.php.net/rfc/session-lock-ini [2] https://wiki.php.net/rfc/session-read_only-lazy_write [3] https://github.com/php/php-src/commit/554021d21e1b2517313a377676260c188152c2eb [4] https://bugs.php.net/bug.php?id=17860 --- ## #78748 — Yasuo Ohgaki — 2014-11-05T21:30:57+00:00 URL: https://externals.io/message/78699#78748 · In reply to: #78699 Hi all, > Apparently this caused > problems for some people as they made 68331 a few days ago. > Just a quick note for this. The user would like to access session data(session handler) regardless of data modification. I suppose it could be solved if session has user space "update" handler as I suggested originally. Alternatively, `session_write_close()`/session_commit() may be changed, so that it is called unconditionally during shutdown. I have to read the code to be sure if this is possible. BTW, anyone know the reason why the user need to call `session_write_close()`/session_commit() unconditionally? Accounting, perhaps? Regards, -- Yasuo Ohgaki yohgaki@ohgaki.net --- ## #78758 — Andrey Andreev — 2014-11-05T23:26:04+00:00 URL: https://externals.io/message/78699#78758 · In reply to: #78748 Hi, > > BTW, anyone know the reason why the user need to call > `session_write_close()`/session_commit() > unconditionally? Accounting, perhaps? Releasing locks is one example. In userland implementations though, it could be all kinds of application-specific logic. Cheers, Andrey. --- ## #78770 — Yasuo Ohgaki — 2014-11-06T02:02:47+00:00 URL: https://externals.io/message/78699#78770 · In reply to: #78748 HI all, > > > > […] > > Just a quick note for this. The user would like to access session > data(session handler) > regardless of data modification. I suppose it could be solved if session > has user space > "update" handler as I suggested originally. > > Alternatively, `session_write_close()`/session_commit() may be changed, so > that it > is called unconditionally during shutdown. I have to read the code to be > sure if this is possible. > > BTW, anyone know the reason why the user need to call > `session_write_close()`/session_commit() > unconditionally? Accounting, perhaps? > The issue user has reported could be solved easily by calling write API during shutdown, IIRC. An explicit flag for this might be needed. I'm just curious for what reason the user needs to call `session_write_close()` during shutdown. It could be used as user states in old PHP, but it does not make it valid usage/design automatically, so I wander the reason. Regards, -- Yasuo Ohgaki yohgaki@ohgaki.net --- ## #78780 — Yasuo Ohgaki — 2014-11-06T05:01:21+00:00 URL: https://externals.io/message/78699#78780 · In reply to: #78699 Hi Damian, > https://bugs.php.net/bug.php?id=68331 > I've reverted incomplete patch for the RFC. Since the RFC patch https://github.com/yohgaki/php-src/compare/PHP-5.6-rfc-session-lock was not merged yet, it has to be removed for now. We have to decide what to do with the RFC, but it should be different thread. Thank you for reporting. Regards, -- Yasuo Ohgaki yohgaki@ohgaki.net --- ## #78784 — Ferenc Kovacs — 2014-11-06T06:38:29+00:00 URL: https://externals.io/message/78699#78784 · In reply to: #78780 2014.11.06. 6:02 ezt írta ("Yasuo Ohgaki" ): > > Hi Damian, > > > > > […] > > I've reverted incomplete patch for the RFC. Thanks! > > Since the RFC patch > https://github.com/yohgaki/php-src/compare/PHP-5.6-rfc-session-lock > was not merged yet, it has to be removed for now. > > We have to decide what to do with the RFC, but it should be different > thread. > Thank you for reporting. I think it could have stayed in the same thread (as this thread wasn't just about the bugreport but also about the partially accepted but never merged rfc) but maybe it is better to separate, I'm just a bit afraid that the context will be missing for those who only read that thread, but not this. --- ## #78795 — Andrey Andreev — 2014-11-06T08:42:33+00:00 URL: https://externals.io/message/78699#78795 · In reply to: #78770 Hi, > HI all, > > > > > […] > > The issue user has reported could be solved easily by calling write API > during shutdown, IIRC. > An explicit flag for this might be needed. There already is such explicit flag in PHP 5.4's OOP version of `session_set_save_handler()`. The linked test script in the bug report also explicitly registers `session_write_close()` as a shutdown function. The problem isn't that there's no call, it's that _the call itself is ignored_. Cheers, Andrey. --- ## #78797 — Yasuo Ohgaki — 2014-11-06T08:55:39+00:00 URL: https://externals.io/message/78699#78797 · In reply to: #78795 Hi Andrey, The problem was "there was partial patch, while patch that overwrite it was not merged". I was confused. Offensive patch is removed for now already and this issue is closed. Regards, -- Yasuo Ohgaki yohgaki@ohgaki.net > Hi, > > > > […] > wrote: > > […] > > There already is such explicit flag in PHP 5.4's OOP version of > `session_set_save_handler()`. > The linked test script in the bug report also explicitly registers > `session_write_close()` as a shutdown function. > > The problem isn't that there's no call, it's that _the call itself is > ignored_. > > > Cheers, > Andrey.