At translatewiki.net, if I initiate refresh data, then run runJobs.php with --procs=3, I see the other two threads to stop running immediately.
This might also be a problem with the core jobqueue, but I don't remember having this issue for other type of jobs.
Version: master
Severity: normal
To clarify it first run only lots of SMWRefreshJobs. After that reached prog=1 it started running both SMWRefreshJobs (with smaller prog) and lots of SMWUpdateJobs with multiple threads.
What is the status on this ticket? If the problem still persists with the current master, are they any suggestions how to solve this issue?
This is still the case. Unfortunately I do not know how to fix it. Adding Aaron to CC in case he has ideas.
Just to clarify, since we have to support MW 1.19 we can't do any fancy stuff that is implemented by the new JobQueue class (as in 1.21/1.22) therefore the current code is as simple as:
class SMWRefreshJob extends Job {
...
$nextjob = new SMWRefreshJob( $this->title, array( ... ) );
$nextjob->insert();
...
}
The Semantic MediaWiki developers requested in https://phabricator.wikimedia.org/T64114 to move their task tracking to https://github.com/SemanticMediaWiki/SemanticMediaWiki/issues and to close remaining tasks in Wikimedia Phabricator. If you still face the problem reported in this task in a supported version of SMW, please feel free to transfer your report to https://github.com/SemanticMediaWiki/SemanticMediaWiki/issues . We are sorry for the inconvenience.
This issue is now tracked with https://github.com/SemanticMediaWiki/SemanticMediaWiki/issues/927