Log and SQL files taking up significant storage
Michael Ott
Posting this in this category because the 'Product Feedback' category the support guy told us to use doesn't seem to exist.
Our team uses LOCAL for push / pulling operations. As we work on around 50 sites, LOCAL gets used a lot.
What we have recently discovered (by chance) is that after each push and pull, the _wpeprivate directory is left cluttered with log files, sometimes tens of MB in size, and copies of the SQL database, which can over 100mb. Just to paint a picture, one dev could be working on a site and do several push / pulls throughout the day, and the storage used by these leftover files quickly start to add up, especially when the entire team is working across different sites.
We understand why the logs and database files are created in that directory (you need a remote temp directory), but that is not the issue. The issue is those files never get deleted and consequently contribute towards our limited storage. I’d hazard a guess most WPEngine customers (who use LOCAL) probably don’t even know this is happening.
We’ve had some discussions over the months with WP support on this subject. Ultimately there doesn’t seem to be any way to solve this from their perspective, although in the past they have manually deleted all the files across all sites for us. Well, that was until today, when this request was refused for ‘reasons’ (which we do understand).
From our perspective, it's unreasonable to be penalised for using WPEngine's own workflow solution (LOCAL).
But here's some solutions we would be happy with:
1) Add a setting to the LOCAL application that automatically deletes logs and database stored in _wpeprivate after each successful push / pull operation.
or
2) Exclude files stored in _wpeprivate from contributing towards our storage allocation.
or
3) A cron that runs daily to cleanup the _wpeprivate, although that might come with some technical challenges.
We're hoping you guys take this matter seriously.
Austin Wendt
Hi Michael Ott - thank you for raising this! I'm the PM of our Local product and work with our engineering team there.
I believe you've been communicating about this issue via a support ticket as well, but to close the loop here on Canny - I agree, Local should not be polluting that directory, and certainly not when it is counting against your storage totals. In the past, I know support has cleared up that directory at your request, but that isn't a scalable solution.
The engineering team met this morning to refine the issue, and the proposed solution is similar to your #1. We've pulled that ticket in and have started the work, so this should be resolved soon - I will follow up here once the fix is deployed. Thank you again for pushing for the fix on this one.