You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to run c.relationhelpers.api.rebuild_relations on our database and it does the first 3 steps (store relations, purge them and clean up intids) but it has not enough memory for the fourth (restoring the relations).
We can of course increase the memory available, but looking at the code, could it be possible to make the restore process reentrant? i.e. that the process keeps committing the changes every x relations restored and removing them from the annotation so if it runs out of memory and the process is killed, the restore process can be continued rather than started from the very beginning?
The code is quite large, so I'm not sure if that's possible at all 🤔
@pbauer as you mostly wrote it, would a PR towards that direction welcomed? 😄
The text was updated successfully, but these errors were encountered:
Another thing that is not completely clear to me is if store_relations and purge_relations are fine to be run as isolated steps, and same goes with cleanup_intids ?
I'm expecting that the first and last (store relations and cleanup intids) are, but not sure about the purge relations...
I'm trying to run
c.relationhelpers.api.rebuild_relations
on our database and it does the first 3 steps (store relations, purge them and clean up intids) but it has not enough memory for the fourth (restoring the relations).We can of course increase the memory available, but looking at the code, could it be possible to make the restore process reentrant? i.e. that the process keeps committing the changes every x relations restored and removing them from the annotation so if it runs out of memory and the process is killed, the restore process can be continued rather than started from the very beginning?
The code is quite large, so I'm not sure if that's possible at all 🤔
@pbauer as you mostly wrote it, would a PR towards that direction welcomed? 😄
The text was updated successfully, but these errors were encountered: