Running archive.php concurrently should process different websites

[anonymous-matomo-user]

If you schedule the archiving of Piwik by cron jobs in a regular basis time, if you got a lot of data to archive during the period the script keeps more than one time.

That's ilogical because there is more than one process calculating the same data.

Example output of msqyl showprocesslist:

| 119784 | piwik | localhost | piwik | Query | 1 | query end | INSERT IGNORE
INTO piwik_archive_blob_2013_01
(idarchive,idsite,date1,date2,period,ts_arch |
| 119859 | piwik | localhost | piwik | Query | 0 | query end | INSERT IGNORE
INTO piwik_archive_blob_2013_01
(idarchive,idsite,date1,date2,period,ts_arch |

To process are archiving the same data.

So the archiving process should be smart enought to determine if there is another process archiving the data and kill itself until the other process ends.

[mattab]

Thanks for the report, it's a regression as it used to work. I'll investigate

[mattab]

Thanks for the report, duplicate of: #3708

[mattab]

See follow up #4903 Add possibility to run multiple archiver in parallel