Page MenuHomePhabricator
Create Task
Maniphest T205005

Reindex produces a lot of Undefined index messages
Closed, DuplicatePublic
Actions

Description

While reindexing Wikidata:

        Task: 1lpe9N1FQxOGsfOcY969RA:2610140 Search Retries: 0 Bulk Retries: 0 Indexed: 584100 / 50278595                                             
        Task: 1lpe9N1FQxOGsfOcY969RA:2610140 Search Retries: 0 Bulk Retries: 0 Indexed: 623300 / 50278595                                             
[Thu Sep 20 18:59:19 2018] [hphp] [1252:7ffb46e4d3c0:0:000001] []                                                                                     
Notice: Undefined index: total in /srv/mediawiki/php-1.32.0-wmf.22/extensions/CirrusSearch/includes/Elastica/ReindexTask.php on line 150              
[Thu Sep 20 18:59:19 2018] [hphp] [1252:7ffb46e4d3c0:0:000002] []                                                                                     
Notice: Undefined index: updated in /srv/mediawiki/php-1.32.0-wmf.22/extensions/CirrusSearch/includes/Elastica/ReindexTask.php on line 151

etc. lots of messages like this.

Related Objects

Event Timeline

Smalyshev created this task.Sep 20 2018, 7:27 PM
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptSep 20 2018, 7:27 PM
Smalyshev mentioned this in T147505: [tracking] CirrusSearch: what is updated during re-indexing.Sep 20 2018, 7:29 PM
Smalyshev edited projects, added Discovery-Search; removed Discovery-Search (Current work).Sep 20 2018, 7:33 PM
EBernhardson added a comment.Sep 20 2018, 8:16 PM

Is this from the task reindexing the same cluster that is having a restart (codfw, iirc)? I don't think elasticsearch tasks or scrolls are able to move between hosts and will fail in the face of one of the nodes participating restarting. The error message also means we don't detect that case, there is probably no particular error handling in the Cirrus side of the reindexing code to detect errors.

Smalyshev added a subscriber: Gehel.Sep 20 2018, 8:20 PM

Yep, according to @Gehel there were restarts going on in the same time (my luck... sigh, should have checked) so maybe it's caused by restarts. Will try again and see. We probably should somehow detect this case anyway.

• EBjune triaged this task as Medium priority.Sep 27 2018, 5:05 PM
• EBjune moved this task from needs triage to Tech Debt/Misc on the Discovery-Search board.
Smalyshev added a comment.Sep 27 2018, 5:17 PM

For clarity: the message does not appear anymore, but I don't close the ticket so that we add some proper handling for this case.

debt moved this task from Tech Debt/Misc to elastic / cirrus on the Discovery-Search board.Jan 29 2019, 6:33 PM
Gehel moved this task from elastic / cirrus to Bugs on the Discovery-Search board.Aug 28 2020, 9:08 AM
dcausse closed this task as a duplicate of T279598: Track down "empty error" during reindexing.Jul 13 2021, 9:36 AM
Content licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 unless otherwise noted; code licensed under GNU General Public License (GPL) 2.0 or later and other open source licenses. By using this site, you agree to the Terms of Use, Privacy Policy, and Code of Conduct. · Wikimedia Foundation · Privacy Policy · Code of Conduct · Terms of Use · Disclaimer · CC-BY-SA · GPL