jenkins-bot has submitted this change and it was merged. (
https://gerrit.wikimedia.org/r/378480 )
Change subject: Speed up populateIpChanges maintenance script.
......................................................................
Speed up populateIpChanges maintenance script.
Use BETWEEN in populateIpChanges maintenance script, which will make it
more efficient when copying revisions with a high rev_id. Also adding a
'max-rev-id' option to prevent the script from looping through IP
changes that have already been copied since the core patch was deployed.
Bug: T175962
Change-Id: I1df10c9b7237ad5002f76f9d354c36ce879d9d9f
---
M maintenance/populateIpChanges.php
1 file changed, 15 insertions(+), 4 deletions(-)
Approvals:
Reedy: Looks good to me, approved
jenkins-bot: Verified
diff --git a/maintenance/populateIpChanges.php
b/maintenance/populateIpChanges.php
index ffb8c43..c173270 100644
--- a/maintenance/populateIpChanges.php
+++ b/maintenance/populateIpChanges.php
@@ -47,6 +47,12 @@
);
$this->addOption( 'rev-id', 'The rev_id to start copying from.
Default: 0', false, true );
$this->addOption(
+ 'max-rev-id',
+ 'The rev_id to stop at. Default: result of MAX(rev_id)',
+ false,
+ true
+ );
+ $this->addOption(
'throttle',
'Wait this many milliseconds after copying each batch
of revisions. Default: 0',
false,
@@ -57,20 +63,25 @@
public function doDBUpdates() {
$lbFactory =
MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
+ $dbr = $this->getDB( DB_REPLICA, [ 'vslow' ] );
$dbw = $this->getDB( DB_MASTER );
$throttle = intval( $this->getOption( 'throttle', 0 ) );
+ $maxRevId = intval( $this->getOption( 'max-rev-id', 0 ) );
$start = $this->getOption( 'rev-id', 0 );
- $end = $dbw->selectField( 'revision', 'MAX(rev_id)', false,
__METHOD__ );
+ $end = $maxRevId > 0
+ ? $maxRevId
+ : $dbw->selectField( 'revision', 'MAX(rev_id)', false,
__METHOD__ );
$blockStart = $start;
$revCount = 0;
$this->output( "Copying IP revisions to ip_changes, from rev_id
$start to rev_id $end\n" );
while ( $blockStart <= $end ) {
- $rows = $dbw->select(
+ $blockEnd = min( $blockStart + 200, $end );
+ $rows = $dbr->select(
'revision',
[ 'rev_id', 'rev_timestamp', 'rev_user_text' ],
- [ "rev_id >= $blockStart", 'rev_user' => 0 ],
+ [ "rev_id BETWEEN $blockStart AND $blockEnd",
'rev_user' => 0 ],
__METHOD__,
[ 'ORDER BY' => 'rev_id ASC', 'LIMIT' =>
$this->mBatchSize ]
);
@@ -80,7 +91,7 @@
}
$this->output( "...checking $this->mBatchSize revisions
for IP edits that need copying, " .
- "starting with rev_id $blockStart\n" );
+ "between rev_ids $blockStart and $blockEnd\n" );
$insertRows = [];
foreach ( $rows as $row ) {
--
To view, visit https://gerrit.wikimedia.org/r/378480
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I1df10c9b7237ad5002f76f9d354c36ce879d9d9f
Gerrit-PatchSet: 3
Gerrit-Project: mediawiki/core
Gerrit-Branch: master
Gerrit-Owner: MusikAnimal <musikani...@gmail.com>
Gerrit-Reviewer: Jcrespo <jcre...@wikimedia.org>
Gerrit-Reviewer: MaxSem <maxsem.w...@gmail.com>
Gerrit-Reviewer: Parent5446 <tylerro...@gmail.com>
Gerrit-Reviewer: Reedy <re...@wikimedia.org>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
