Had to tweak statusnet.ini to remove the DB_DATAOBJECT_MYSQLTIMESTAMP bitfield constant on session.modified; while it sounds like a useful and legit setting, it actually just means that DB_DataObject silently fails to pass through any attempts to explicitly set the value. As a result, MySQL does its default behavior which is to insert the current *LOCAL* time, which is useless.
This was leading to early GC west of GMT, or late GC east of it. Early GC could at worst destroy all live sessions (whoever's session *triggered* GC is fine, as the session then gets saved right back.)
$session->id = $id;
$session->session_data = $session_data;
$session->created = common_sql_now();
+ $session->modified = common_sql_now();
$result = $session->insert();
$orig = clone($session);
$session->session_data = $session_data;
+ $session->modified = common_sql_now();
$result = $session->update($orig);
id = 130
session_data = 34
created = 142
-modified = 384
+modified = 142
+; Warning: using DB_DATAOBJECT_MYSQLTIMESTAMP (256) causes DB_DataObject
+; to SILENTLY REMOVE ATTEMPTS TO SET THIS FIELD DIRECTLY, which is pretty
+; bad because the default behavior for auto-updated TIMESTAMP fields is
+; to use local time. Local time can't be compared to UTC in any useful
+; way, so doing that breaks session GC.
+;
+; Instead we'll use the plain datetime settings so it'll actually save the
+; UTC value we provide when updating.
+;
+; Long-term fix: punch MySQL in the face until it understands that local
+; time is a tool of the cyber-devil.
+;
+;modified = 384
[session__keys]
id = K